United States v. Askew

United States Court of Appeals
FOR THE DISTRICT OF COLUMBIA CIRCUIT

Argued October 11, 2007 Decided June 20, 2008

No. 04-3092

UNITED STATES OF AMERICA,
APPELLEE

v.

PAUL ASKEW,
APPELLANT

Appeal from the United States District Court
for the District of Columbia
(No. 04cr00010-01)

Sandra G. Roland, Assistant Federal Public Defender,
argued the cause for appellant. With her on the briefs were A.J.
Kramer, Federal Public Defender, and Sharon R. Rice, Assistant
Federal Public Defender.
Blair G. Brown, Jonathan E. Nuechterlein, and Sambhav N.
Sankar were on the brief for amicus curiae National Association
of Criminal Defense Lawyers in support of appellant.
Florence Y. Pan, Assistant U.S. Attorney, argued the cause
for appellee. With her on the brief were Jeffrey A. Taylor, U.S.
Attorney, and Roy W. McLeese, III, Assistant U.S. Attorney.
 2

Before: SENTELLE, Chief Judge, GINSBURG,* HENDERSON,
RANDOLPH, ROGERS, TATEL, GARLAND, BROWN, GRIFFITH, and
KAVANAUGH, Circuit Judges, and EDWARDS, Senior Circuit
Judge.
Opinion for the Court filed by Senior Circuit Judge
EDWARDS, with whom Circuit Judges ROGERS, TATEL, and
BROWN join, and with whom Circuit Judge GRIFFITH joins
except as to Part III.D, and with whom Circuit Judges
GINSBURG and GARLAND join as to Parts I, III.D, III.E, and IV.
Concurring opinion filed by Circuit Judge GRIFFITH, with
whom Circuit Judges ROGERS and TATEL join, except as to
footnote 2.
Dissenting opinion filed by Circuit Judge KAVANAUGH,
with whom Chief Judge SENTELLE and Circuit Judges
HENDERSON and RANDOLPH join.
EDWARDS, Senior Circuit Judge: On the evening of
December 19, 2003, police officers received a broadcast
“lookout” for an armed robber. Appellant Paul Askew, who
wore clothing similar, but not identical to that described in the
lookout, was stopped. The police then conducted a Terry “frisk”
which produced nothing. Some time after the frisk was
completed, the police moved appellant to a place where he could
be seen by the complaining witness. The officers’ purpose was
to determine whether the complainant could identify appellant
as her assailant. The District Court’s findings of fact indicate
that appellant complied during the stop and was not handcuffed
during the identification show-up. Preparatory to the show-up,
but without appellant’s consent, one of the officers attempted to
unzip appellant’s outer jacket to reveal to the complainant what
appellant had on under the jacket. The officer’s unfastening of
the jacket was interrupted when the zipper hit a hard object at

*
Judge Ginsburg was Chief Judge at the time of oral argument.
 3

appellant’s waist. Appellant then pushed the officer’s hand
away from his jacket. These latter events aroused the officer’s
suspicion, but the officer did nothing and the show-up
continued. Although appellant was not implicated by the
complaining witness, the police officers continued to detain him,
walked him backwards towards a police vehicle, placed him on
the hood of the car, and then fully unzipped his jacket. The
officers found a gun in an open waist pouch and arrested
appellant.
In April 2004, after the District Court denied his Fourth
Amendment motion to suppress the Government’s evidence,
appellant entered a conditional guilty plea to a one-count
indictment charging him with possession of a firearm by a
convicted felon in violation of 18 U.S.C. § 922(g)(1). Appellant
reserved his right to appeal the District Court’s denial of his
motion to suppress. See FED. R. CRIM. P. 11(a)(2). On June 29,
2004, the District Court sentenced appellant to 36 months’
imprisonment, followed by three years’ supervised release.
Gov’t En Banc Br. at 2; Appellant En Banc Br. at 2.
On April 6, 2007, a divided panel of the court affirmed the
District Court’s denial of appellant’s motion to suppress. On
July 12, 2007, the panel’s judgment was vacated and an order
was issued granting appellant’s petition for rehearing en banc.
The order granting en banc review instructed the parties to
address the following issue:
[W]hether during a Terry stop police officers may unzip a
suspect’s jacket solely to facilitate a show-up. In
addressing this question, the parties should consider
whether the officers’ action was a lawful search under Terry
v. Ohio, 392 U.S. 1 (1968), and its progeny.
United States v. Askew, No. 04-3092, Order Granting En Banc
Review (D.C. Cir. July 12, 2007). The order made clear that the
only issue before the en banc court was whether the first, partial
 4

unzipping was unlawful. There is no dispute that if the partial
unzipping was unlawful, the discovery of the hard object at
appellant’s waist during that unzipping cannot justify the second
full unzipping that yielded the gun.
On April 10, 2008, after oral arguments were heard by the
en banc court, an order was issued instructing the parties to
submit supplemental briefs addressing the following questions:
1. Assuming, arguendo, that it is not dispositive that the
unzipping was a search, was the gun evidence
nonetheless inadmissible as the product of steps taken
to facilitate a show-up witness’ identification, on a
theory that there were not reasonable grounds for
believing that unzipping the jacket would establish or
negate the suspect’s connection with the crime under
investigation?
2. Was the gun evidence admissible as the product of a
valid protective search, on a theory that regardless of
the officer’s subjective intent the initial unzipping was
an objectively reasonable response to the suspect’s
conduct during the pat-down?
3. Was the gun evidence admissible under the doctrine of
inevitable discovery, on a theory that the officers had
not completed the pat-down but would have done so
after the show-up?
United States v. Askew, No. 04-3092, Order (directing
supplemental briefing) (D.C. Cir. Apr. 10, 2008).
As described in its opening brief, the Government submits
that the principal question for this court is whether the police
“violate[d] appellant’s Fourth Amendment rights by partially
unzipping [his] outer jacket during a show-up identification
procedure, so that a robbery victim could see whether
appellant’s sweatshirt matched that of the robbery perpetrator.”
 5

Gov’t En Banc Br. at 13; see also id. at 22, 24. Applying
Minnesota v. Dickerson, 508 U.S. 366 (1993), and the precedent
on which it rests, to the District Court’s uncontested findings of
fact, a five-judge plurality of this court concludes that the
answer to this question is yes. Because the police officer’s
unzipping of appellant’s jacket went beyond what was necessary
to protect the investigating officers or others nearby, it amounted
to precisely the sort of evidentiary search that is impermissible
in the context of a Terry stop.
Even assuming, arguendo, that an unzipping to facilitate a
show-up is permissible under some circumstances, a majority of
the court is nonetheless satisfied that the police officer’s actions
cannot be justified here since there were no reasonable grounds
for believing that the unzipping would establish or negate
appellant’s identification as the robber in question.1 A majority
of the court is also satisfied that the Government’s alternative
argument, that the search of appellant can be justified as an
objectively reasonable continuation of the protective frisk, is
both contrary to the District Court’s factual findings and
unsupportable on any plausible reading of the record.
Finally, the Government concedes that “[t]he gun is not
admissible under a theory of ‘inevitable discovery.’” Gov’t
Supplemental Br. at 12. As the Government explains, it “did not
make an inevitable-discovery argument before the district court,
and thus failed to elicit” the testimony necessary to support such
a theory under Nix v. Williams, 467 U.S. 431 (1984). Gov’t
Supplemental Br. at 14. Moreover, the Government
acknowledges that “[b]ecause the inevitable-discovery theory
raises factual issues that could have been addressed at the

1
In this respect, assuming, arguendo, that an unzipping to
facilitate a show-up is permissible under some circumstances, a
majority of the court has resolved the legal question about the
permissible scope of police activity at show-ups.
 6

suppression hearing but were not, [it does] not believe that [it is]
in a position to request a remand for further development of the
record.” Id. at 15 n.7.
I. THE DISTRICT COURT’S FACTUAL FINDINGS
Following completion of the hearing on appellant’s
suppression motion, the District Court set forth its factual
findings in a published opinion. See United States v. Askew, 313
F. Supp. 2d 1 (D.D.C. 2004). “[A]ppellate courts must
constantly have in mind that their function is not to decide
factual issues de novo.” Anderson v. City of Bessemer City, 470
U.S. 564, 573 (1985) (quotation marks omitted). Thus, as the
Government rightly points out, “[t]his court must accept the
district court’s findings of fact unless clearly erroneous.” Gov’t
En Banc Br. at 15. This rule is firmly entrenched in Supreme
Court precedent, see Ornelas v. United States, 517 U.S. 690, 699
(1996), and in applying it we “overstep[] the bounds of [our]
duty . . . if [we] undertake[] to duplicate the role of the lower
court,” Bessemer City, 470 U.S. at 573. This is especially so
when, as here, the trial court is required to reconcile differences
in testimony in order to make factual findings. See Mar. 26 Tr.
at 29-31; see generally Bessemer City, 470 U.S. at 573-76.
Notably, neither party challenged the District Court’s
findings in this case. In fact, in its brief filed with the panel, the
Government characterized the District Court’s factual findings
as “consistent with the government’s evidence at the suppression
hearing.” Gov’t Panel Br. at 9. Even if those findings had been
challenged, we could not overturn them unless we were
“definitely and firmly convinced that a mistake [had] been
committed.” Bessemer City, 470 U.S. at 573. Because there are
no grounds for such a conclusion, the District Court’s factual
findings constitute the record by which appellate review is
bound. Consequently, they are, in pertinent part, reproduced
below.
 7

On the night of December 19, 2003, around 11:00 p.m.,
a radio run alerted Officer Anthony Bowman of the
Metropolitan Police Department to a report of an armed
robbery in the 700 block of 9th Street, S.E., in Washington,
D.C. Officer Bowman canvassed the area in his patrol car,
looking for individuals matching the description of the
perpetrator: a black male, approximately six-feet tall,
wearing a blue sweatshirt and blue jeans. The radio report
reflected that the perpetrator had been last seen moving on
9th Street, S.E., in an unknown direction.
Within two minutes of the radio report, and within
approximately ten minutes of the robbery, Officer Bowman
spotted defendant Paul Askew walking in the 200 block of
9th Street, S.E., five blocks from the scene of the robbery.
Upon seeing Officer Bowman, the defendant turned and
walked in a different direction, but Officer Bowman
continued to follow the defendant in the patrol car.
Defendant is a black male, six-feet, three-inches tall, and at
the time was wearing clothing quite similar – but not
identical – to the description broadcast over the police
radio. While the description of the perpetrator mentioned
a blue sweatshirt and blue jeans, Officer Bowman testified
that the defendant was wearing blue sweatpants, “a navy
blue jacket[, and] a darker blue fleece type jacket
underneath. He had on two jackets.” Officer Bowman
reported to the dispatcher that Askew “vaguely match[ed]
th[e] description.” After noticing that the defendant had a
moustache, Officer Bowman checked with the dispatcher to
determine whether the robber also had a moustache. When
the dispatcher responded affirmatively, Officer Bowman
stopped the defendant.
Officer Bowman asked the defendant to come to the
patrol car, and he complied. The defendant also complied
with Officer Bowman’s further requests that he produce
 8

some identification, take his hands out of his pockets, and
place his hands on the top of his head. Officer Bowman
then told the defendant that he was being stopped because
of his physical similarity to the description of a robber.
When back-up units arrived, Officer Bowman returned to
the interior of his car to check whether the police
department computer returned any information on the
defendant. Officer Bowman’s back was turned for the next
couple of minutes and he did not see the pat-down of the
defendant that followed.
Officer James Koenig conducted a pat-down of the
defendant and found nothing. Shortly afterwards, another
officer, Officer Benton, drove the robbery victim to the
place where the defendant was being detained, for the
purpose of conducting a show-up. The victim remained in
the car while Officer Koenig and Officer Anthony Willis
brought the defendant to a place where he could be seen by
the victim. The defendant was not in handcuffs at that time.
Preparatory to the show-up, Officer Willis attempted to
unzip the defendant’s outer jacket to reveal the sweatshirt
underneath so the victim could better determine if the
defendant was the robber. Officer Willis testified that he
remembered the “blue hooded sweatshirt” described in the
radio run and “wanted the complainant to see what [the
defendant] had on to make sure that he wasn’t zipping
nothing up to cover up. So I went to unzip it down so that
. . . they could see what he had on.” Officer Willis had
difficulty, however, in unzipping the jacket when the zipper
hit what he described as a “hard” or “solid” object and
“didn’t go past [the object]. It stopped there. And at that
time, that’s when [the defendant] knocked my hand down,”
away from the zipper.
After the show-up, Officer Willis and Officer Edward
Snead walked the defendant backwards toward the car,
 9

placed him on the hood of the car, and unzipped his jacket.
Visible once the jacket was unzipped was an open black
waist pouch, or “fanny pack,” with a silver object sticking
out. On further inspection, the silver object was identified
as a gun, and the defendant was handcuffed and arrested.
Askew, 313 F. Supp. 2d at 2-3 (alterations in original) (footnotes
and transcript citations omitted).
In the course of its legal analysis, the District Court
concluded that the disputed unzipping was undertaken to
facilitate the show-up. Id. at 4. This conclusion is consistent
with its factual findings that Officer Koenig found “nothing”
during the frisk and that the complainant was brought to where
appellant was detained “shortly afterwards.” Id. at 3; see also
id. at 4. In a footnote, the District Court notes that Officer
Koenig did not testify at the suppression hearing. Rather, the
testimony regarding the pat down was provided by Officer
Willis. Id. at 3 n.2. (In fact, the motions hearing transcript
reveals that the Government, as part of its trial strategy, chose
not to put on Officer Koenig. See Mar. 26 Tr. at 21, 24, 38.) In
this same footnote, the Court describes Officer Willis’s
“suggest[ion] that Officer Koenig had not completed the pat-
down . . . when Officer Benton arrived with the robbery victim
for a show-up.” Askew, 313 F. Supp. 2d at 3 n.2 (emphasis
added). The Court also notes Officer Willis’s testimony that
“perhaps” the pat down had not been completed “because of
some resistance by the defendant.” Id. (emphasis added). The
Court then points to “[t]he government acknowledge[ment] that
when Officer Koenig patted the defendant down, he did not find
anything” and reiterates its own finding that “[t]he subsequent
discovery of the gun at issue was not the result of this pat-
down.” Id.; see also id. at 4 (“[t]he initial pat-down by Officer
Koenig did not reveal the presence of any weapon”). In other
words, the District Court does not credit Officer Willis’s
suggestion that the pat down may have been incomplete or
 10

Officer Willis’s speculation regarding why that may have been
the case.
II. OVERVIEW
“Time and again” the Supreme Court “has observed that
searches and seizures 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.”
Dickerson, 508 U.S. at 372 (quotation marks omitted)
(collecting cases). And the Court has made it clear that the
“inestimable right of personal security” embodied in the Fourth
Amendment “belongs as much to the citizen on the streets . . . as
to the homeowner closeted in his study . . . . For, as [the
Supreme] Court has always recognized, ‘No right is held more
sacred, or is more carefully guarded, by the common law, than
the right of every individual to the possession and control of his
own person, free from all restraint or interference of others,
unless by clear and unquestionable authority of law.’” Terry v.
Ohio, 392 U.S. 1, 8-9 (1968) (quoting Union Pac. Ry. Co. v.
Botsford, 141 U.S. 250, 251 (1891)).
In Terry v. Ohio, the Supreme Court defined one of the few
exceptions to the prohibition against warrantless searches of a
person. The Court held that in the context of a properly justified
on-the-street stop, if a police officer has a reasonable articulable
suspicion “that the individual whose suspicious behavior he is
investigating at close range is armed and presently dangerous to
the officer or to others,” that officer may conduct a limited
protective search “to determine whether the person is in fact
carrying a weapon.” 392 U.S. at 24; see also id. at 30-31. In
Sibron v. New York, 392 U.S. 40 (1968), an opinion issued on
the same day as Terry, the Court confirmed the limited nature of
the Terry search exception, explicitly stating that the “only goal
which might conceivably” justify a search in the context of a
Terry stop is a search for weapons. Id. at 65.
 11

In subsequent cases, the Court has been unequivocal in
explaining that “[t]he purpose of this limited search is not to
discover evidence of crime, but to allow the officer to pursue his
investigation without fear of violence.” Adams v. Williams, 407
U.S. 143, 146 (1972) (emphasis added). Thus, the fruit of a
search that “goes beyond what is necessary to determine if [a]
suspect is armed . . . will be suppressed.” Dickerson, 508 U.S.
at 373 (citing Sibron, 392 U.S. at 65-66). This is so, the Court
explained in Minnesota v. Dickerson, because searches that
exceed what is necessary to determine if an individual is armed
“amount[] to the sort of evidentiary search that Terry expressly
refused to authorize” and that the Court “condemned” in Sibron
v. New York, 392 U.S. 40, and Michigan v. Long, 463 U.S. 1032
(1983). Dickerson, 508 U.S. at 378.
Appellant argues that pursuant to this well-established body
of law, the unzipping of his jacket was unlawful, because it was
not undertaken for protective purposes, but rather amounted to
an impermissible evidentiary search unsupported by probable
cause or a warrant. The Government, in contrast, asserts that
this question is not controlled by Dickerson and the precedent on
which Dickerson rests, but rather should be decided through
application of the reasonableness balancing test. Under this test,
the permissibility of a Government action is determined by
“balancing the [governmental] need to search [or seize] against
the invasion which the search [or seizure] entails.” Terry, 392
U.S. at 21 (second and third alteration in original) (quotation
marks omitted). Pursuant to this test, the Government argues
that because the unzipping of appellant’s jacket “was a
reasonable, de minimis investigative measure that appropriately
facilitated the show-up procedure,” it need not have been
supported by a warrant or probable cause. Gov’t En Banc Br. at
13.
When the Supreme Court has weighed the interests relevant
to determining whether a certain type of official conduct is
 12

reasonable under the Fourth Amendment, lower courts are not
free to strike a new and different balance. With respect to
searches of individuals detained during on-the-street encounters
on less than probable cause, the balance was struck in Terry, 392
U.S. 1, and Sibron, 392 U.S. 40. Employing the reasonableness
test to which the Government refers, the Court in Terry
authorized a strictly circumscribed search for weapons when an
officer has reasonable articulable suspicion to believe that a
properly stopped individual is armed and dangerous to the
officers or others nearby. 392 U.S. at 30-31; see also New York
v. Class, 475 U.S. 106, 117 (1986) (“When a search or seizure
has as its immediate object a search for a weapon, . . . we have
struck the balance to allow the weighty interest in the safety of
police officers to justify warrantless searches based only on a
reasonable suspicion of criminal activity.”). The Sibron Court,
applying Terry’s holding, made clear that Terry did not permit
police officers to undertake searches not justified on a safety
rationale. See 392 U.S. at 63-64. It also established that
protective searches that extend beyond the strictly circumscribed
bounds of Terry are impermissible evidentiary searches. Id. at
65-66. Most important, for our purposes, the Court’s opinion in
Minnesota v. Dickerson, 508 U.S. 366, confirms that when an
on-the-street search of an individual who has been stopped on
reasonable articulable suspicion extends beyond what is
authorized in Terry, there is no reweighing to be undertaken on
the grounds that the search might be described as minimally
intrusive or that the evidence sought might provide probable
cause to believe that the suspect committed the crime in
question.
Because there is no principled way to distinguish the initial
unzipping in this case from the search in Dickerson, this court
is not free to reweigh the interests at issue to create the new and
wholly unprecedented identification exception to the warrant
and probable cause requirements that the Government urges
upon us. Rather, applying the balance that the Supreme Court
 13

struck in Terry and Sibron to the uncontested factual findings of
the District Court, we are convinced that the unzipping of
appellant’s jacket was a non-protective evidentiary search that
violated the Fourth Amendment.
III. ANALYSIS OF THE LEGAL ISSUES
A. The Partial Unzipping and Opening of Appellant’s Jacket
Was a Search
Before turning to the issues before the en banc court, we
must determine whether the unzipping of appellant’s jacket was,
in fact, a search. Clearly it was. By zipping up his jacket,
appellant unquestionably evidenced an intent to keep private
whatever lay under it. The only question, then, is whether
society is prepared to recognize such an expectation as
reasonable. See Katz v. United States, 389 U.S. 347, 361 (1967)
(Harlan, J., concurring).
This question was unequivocally answered by the Supreme
Court in Terry. At issue there was the touching of the outer
surface of the defendant’s overcoat. Terry, 392 U.S. at 7. As
the Court explained, even that limited action rose to the level of
a “search” within the purview of the Fourth Amendment. Id. at
16. Because the opening of a fastened coat, like the opening of
most other clothing, renders visible whatever lies underneath,
such an action involves an even greater intrusion in precisely the
same socially recognized expectation of privacy. And such an
intrusion is particularly great when, as here, the opening takes
place on a public street. In describing the level of personal
intrusion occasioned by a public frisk, the Terry Court stated:
[I]t is simply fantastic to urge that such a procedure
performed in public by a policeman while the citizen stands
helpless, perhaps facing a wall with his hands raised, is a
“petty indignity.” It is a serious intrusion upon the sanctity
of the person, which may inflict great indignity and arouse
strong resentment.
 14

Id. at 16-17. The undoing of clothing to reveal whatever is
underneath to whomever happens to be on the street necessarily
involves an even more serious intrusion upon the sanctity of the
person. The involuntary opening of someone’s clothing reveals
to the world at large (not just to the searching police officer)
what an individual obviously intends to keep private.
As noted above, the Government did not dispute the
characterization of the unzipping as a search during arguments
before the panel. Before the en banc court, however, the
Government refused to concede the point. Stating that it was
only “assum[ing], arguendo, that the unzipping of appellant’s
jacket was a ‘search,’” Gov’t En Banc Br. at 23 n.11, the
Government maintained that the police action did not actually
amount to a search because the sweatshirt that the police
expected to reveal “presumably was widely visible when
appellant was in indoor settings.” Id. This argument is flawed
in both its legal and factual premises.
Relying primarily on United States v. Dionisio, 410 U.S. 1
(1973), the Government likens appellant’s sweatshirt to a
“physical characteristic . . . constantly exposed to the public.”
Gov’t En Banc Br. at 23 n.11 (omissions in original). This
analogy is inapt. In Dionisio, the Supreme Court held that
production of a voice exemplar pursuant to a grand jury
subpoena did not constitute a search, because “[t]he physical
characteristics of a person’s voice, its tone and manner, as
opposed to the content of a specific conversation, are constantly
exposed to the public. Like a man’s facial characteristics, or
handwriting, his voice is repeatedly produced for others to
hear.” 410 U.S. at 14. As the Court explained, “while the
content of a communication is entitled to Fourth Amendment
protection . . . the underlying identifying characteristics – the
constant factor throughout both public and private
communications – are open for all to see or hear.” Id. (quotation
marks omitted) (omission in original). Consequently, “[n]o
 15

person can have a reasonable expectation that others will not
know the sound of his voice, any more than he can reasonably
expect that his face will be a mystery to the world.” Id.
The same cannot be said of a piece of clothing when the
only information that the police have about that clothing is that
the wearer has chosen to shield most of it from public view.
Contrary to the Government’s assertion, there is nothing about
a sweatshirt that – like the characteristics of an individual’s
voice, handwriting, or face – must necessarily be revealed to the
public in the course of daily life. An individual may choose to
expose all or part of an article of clothing to the public or he
may choose to keep all or part of that clothing covered.
The only evidence presented by the Government regarding
appellant’s sweatshirt was that appellant had demonstrated an
intent to shield most of it from public view. When a
government agent unfastens, lifts, pulls down, pats, or otherwise
manipulates clothing to reveal or determine what lies
underneath, that manipulation necessarily involves the sort of
“‘probing into an individual’s private life’” that the Court in
Davis v. Mississippi, 394 U.S. 721 (1969), characterized as the
mark of a search or interrogation. Dionisio, 410 U.S. at 15
(quoting Davis v. Mississippi). Citing Terry, the Dionisio Court
reiterated that even the minimal intrusion involved in a Terry
frisk of outer clothing necessarily amounts to a Fourth
Amendment search. Contrasting the seizure of voice exemplars
to permissible Terry pat downs, the Court explained that the
former “does not involve the severe, though brief, intrusion
upon cherished personal security, effected by” the latter.
Dionisio, 410 U.S. at 15 (quotation marks omitted).
The Government’s argument that the unzipping of
appellant’s jacket was not a search is also based on
fundamentally flawed factual premises. First, the Government
assumes that the unzipping of appellant’s jacket would reveal
only appellant’s already partially visible sweatshirt. One need
 16

only consider the special medical needs of certain individuals,
including those who are forced to use colostomy bags and heart
monitors (to name a few) to recognize the fallacy of this
assumption. Such devices frequently are attached to an
individual’s abdominal area and often require those wearing
them to hike up the clothing around their midsection to
accommodate the device. Second, many individuals wear
clothing that is comfortable to them but would appear unseemly
to others. In such circumstances, the individual who does not
want the unseemly portions of his clothing publicly exposed will
cover or partially cover that clothing with another garment.
Appellant’s fastening of his jacket effectively expressed a
recognized and reasonable expectation of privacy. And its
unfastening cannot be characterized as a non-search given the
inability of the police to know what other information pertaining
to appellant’s private life that unfastening would reveal.
B. Minnesota v. Dickerson Dictates the Conclusion That the
Unzipping of Appellant’s Jacket Was an Impermissible
Search for Evidence
The legal principles controlling the disposition of this case
were largely established in three post-Terry Supreme Court
cases, all of which were relied on by the Court in Minnesota v.
Dickerson. Those cases are Sibron, 392 U.S. 40; Ybarra v.
Illinois, 444 U.S. 85 (1979); and Long, 463 U.S. 1032.
Discussing and applying Terry, each makes clear that in the
context of an on-the-street seizure based on less than probable
cause, there is no balancing of interests to be undertaken in
determining whether a particular search of a stopped suspect is
reasonable and therefore permissible under the Fourth
Amendment. Rather, applying the balance struck in Terry,
courts are constrained to suppress evidence obtained during such
a stop if it is the fruit of a search that was not necessary to
protect the investigating officers or others nearby. Put another
way, these decisions establish that individuals stopped on no
 17

more than reasonable articulable suspicion may not, consonant
with the protections of the Fourth Amendment, be searched for
the purpose of revealing and gathering evidence pertaining to
the illegal behavior of which they are suspected.
Sibron, a companion case to Terry, clarified the limits of the
search exception authorized in Terry, holding that a non-
protective evidentiary search of an individual detained on less
than probable cause, would not pass Fourth Amendment muster.
In that case, a police officer stopped Sibron on suspicion that he
was selling narcotics. See 392 U.S. at 45. When the officer said
to Sibron, “You know what I am after,” Sibron “mumbled
something and reached into his pocket. Simultaneously, [the
officer] thrust his hand into the same pocket” discovering
several envelopes of heroin. Id. (quotation marks omitted).
After first concluding that the officer lacked probable cause to
arrest Sibron when he stopped him, id. at 62-63, the Court held
that the search of Sibron’s pocket for evidence of narcotics was
unconstitutional. Relying on its decision in Terry, the Court
concluded that the officer’s action was not predicated on any
concern that Sibron was armed, but rather constituted a search
for evidence of the crime that the officer suspected Sibron of
committing. See id. at 63-64. As the Court described it, the
officer’s “opening statement to Sibron . . . made it abundantly
clear that he sought narcotics, and his testimony at the hearing
left no doubt that he thought there were narcotics in Sibron’s
pocket.” Id. at 64. Moreover, as the Court explained, even if
the search had been predicated on a reasonable articulable
suspicion that Sibron was armed, the search would have been
impermissible because it “was not reasonably limited in scope
to the accomplishment of the only goal which might conceivably
have justified its inception – the protection of the officer by
disarming a potentially dangerous man.” Id. at 65 (emphasis
added).
 18

In Ybarra, the Court reiterated the limits of its Terry search
rationale. The search at issue there took place in the Aurora Tap
Tavern. 444 U.S. at 88. The police had obtained a warrant
authorizing the search of the tavern and an individual named
Greg. Id. Upon entering the tavern, the officers frisked each of
its patrons for weapons. Id. During the frisk of Ybarra, an
“officer felt what he described as a cigarette pack with objects
in it.” Id. at 88 (quotation marks omitted). After completing the
pat down of a number of other patrons, the officer returned to
Ybarra, frisked him again, and removed from his shirt pocket a
cigarette pack containing what turned out to be packets of
heroin. Id. at 89. The State argued, inter alia, that the first pat
down of Ybarra was a permissible frisk for weapons under
Terry. Id. at 92. The Court rejected the State’s argument,
concluding that because “[t]he initial frisk of Ybarra was simply
not supported by a reasonable belief that he was armed and
presently dangerous,” id. at 92-93, the evidence must be
suppressed. As the Court explained, “[t]he Terry case created
an exception to the requirement of probable cause, an exception
whose narrow scope this Court has been careful to maintain.
Under that doctrine a law enforcement officer, for his own
protection and safety, may conduct a patdown to find weapons
that he reasonably believes or suspects are then in the possession
of the person he has accosted.” Id. at 93 (emphasis added)
(quotation marks omitted). However, the Court was careful to
point out that “[n]othing in Terry can be understood to allow a
generalized ‘cursory search for weapons’ or, indeed, any search
whatever for anything but weapons.” Id. at 93-94 (emphasis
added).
In Long, which extended Terry’s protective search doctrine
to the interior of cars, the Court confirmed and explained its
unwillingness to allow non-protective evidentiary searches
based on less than probable cause during run-of-the-mill
investigatory stops. The Court there addressed the
constitutionality of the search of the interior compartment of an
 19

automobile during a lawful investigatory stop of its occupant.
463 U.S. at 1037. Quoting Terry, the Court concluded “that the
search of the passenger compartment of an automobile, limited
to those areas in which a weapon may be placed or hidden, is
permissible if the police officer possesses a reasonable belief
based on ‘specific and articulable facts which, taken together
with the rational inferences from those facts, reasonably
warrant’ the officer in believing that the suspect is dangerous
and the suspect may gain immediate control of weapons.” Id. at
1049. The Court “stress[ed],” however, that its decision did
“not mean that the police may conduct automobile searches
whenever they conduct an investigative stop.” Id. at n.14
(emphasis in original). As the Court explained, this is because
the justification for allowing automatic warrantless searches for
evidence incident to an arrest does not support an automatic
warrantless evidentiary search during a Terry stop. See id.
An additional interest exists in the arrest context, i.e.,
preservation of evidence, and this justifies an “automatic”
search. However, that additional interest does not exist in
the Terry context. A Terry search, “unlike a search without
a warrant incident to a lawful arrest, is not justified by any
need to prevent the disappearance or destruction of
evidence of crime . . . . The sole justification of the search
. . . is the protection of police officers and others nearby.”
Id. (quoting Terry, 392 U.S. at 21) (alterations in original).
Relying on the rationale of these cases, the Court in
Minnesota v. Dickerson again confirmed that non-protective
evidentiary searches are not permissible during Terry stops.
There, on facts essentially indistinguishable from those
presented here, the Supreme Court held that a police officer
violated the Fourth Amendment proscription against
unreasonable searches when, following a protective frisk that
produced no evidence of a weapon, he manipulated the outside
of a suspect’s jacket pocket in an effort to identify as crack
 20

cocaine a small lump detected during the frisk. 508 U.S. at 378-
79. Dickerson first came to the attention of the searching officer
and his partner when he left what was a well-known crack
house. Id. at 368. “According to testimony credited by the trial
court, [Dickerson] began walking toward the police but, upon
spotting the squad car and making eye contact with one of the
officers, abruptly halted and began walking in the opposite
direction.” Id. at 368-69. Based on these “evasive actions and
the fact that he had just left a building known for cocaine traffic,
the officers decided to . . . investigate further.” Id. at 369. They
stopped Dickerson and patted him down for weapons. Id.
Although the pat down revealed no weapons, “the officer
conducting the search did take an interest in a small lump in
respondent’s nylon jacket.” Id. The officer, who did not
immediately recognize the lump as crack cocaine, determined
what it was only after “squeezing, sliding and otherwise
manipulating the contents of the defendant’s pocket.” Id. at 378
(quotation marks omitted).
The Supreme Court granted certiorari in Dickerson “to
resolve a conflict among the state and federal courts over
whether contraband detected through the sense of touch during
a patdown search may be admitted into evidence.” Id. at 371.
In essence, the issue before the Court was whether and when the
plain view doctrine of Arizona v. Hicks, 480 U.S. 321 (1987),
could justify a plain feel exception for nonthreatening evidence
detected during a Terry pat down.
Citing and discussing Sibron, Ybarra, and Long, the Court
first reviewed the limits on searches conducted during Terry
stops, Dickerson, 508 U.S. at 372-73, concluding with the long
“settled” principle that if a “protective search goes beyond what
is necessary to determine if the suspect is armed, it is no longer
valid under Terry and its fruits will be suppressed,” id. at 373.
Turning to the plain view doctrine, the Court explained that “if
police are lawfully in a position from which they view an object,
 21

if its incriminating character is immediately apparent, and if the
officers have a lawful right of access to the object, they may
seize it without a warrant.” Id. at 375. However, the Court
continued, if “the police lack probable cause to believe that an
object in plain view is contraband without conducting some
further search of the object – i.e., if ‘its incriminating character
[is not] immediately apparent,’ the plain-view doctrine cannot
justify its seizure.” Id. (alteration in original) (citations
omitted).
Applying these principles to nonthreatening contraband
discovered during Terry pat downs, the Court concluded that a
warrantless seizure would be permissible “so long as the
officers’ search stays within the bounds marked by Terry,” id. at
373 (emphasis added), and the object’s “contour or mass makes
its identity immediately apparent,” id. at 375 (emphasis added).
In addressing the facts before it, the Dickerson Court
described the “dispositive question” as “whether the officer who
conducted the search was acting within the lawful bounds
marked by Terry at the time he gained probable cause to believe
that the lump in respondent’s jacket was contraband.” Id. at
377. Answering that question in the negative, the Court
concluded that the “officer’s continued exploration of
respondent’s pocket after having concluded that it contained no
weapon was unrelated to ‘[t]he sole justification of the search
[under Terry:] . . . the protection of the police officer and others
nearby.’” Id. at 378 (quoting Terry, 392 U.S. at 29) (alterations
in original). Thus, the Court found that the manipulation of
appellant’s pocket “amounted to the sort of evidentiary search
that Terry expressly refused to authorize, see [392 U.S.] at 26,
and that we have condemned in” Sibron, 392 U.S. at 65-66, and
Long, 463 U.S. at 1049 n.14. Dickerson, 508 U.S. at 378.
“Once again,” the Court explained, “analogy to the plain-
view doctrine is apt.” Id.
 22

In Arizona v. Hicks, 480 U.S. 321 (1987), this Court held
invalid the seizure of stolen stereo equipment found by
police while executing a valid search for other evidence.
Although the police were lawfully on the premises, they
obtained probable cause to believe that the stereo
equipment was contraband only after moving the equipment
to permit officers to read its serial numbers. The
subsequent seizure of the equipment could not be justified
by the plain-view doctrine, this Court explained, because
the incriminating character of the stereo equipment was not
immediately apparent; rather, probable cause to believe that
the equipment was stolen arose only as a result of a further
search – the moving of the equipment – that was not
authorized by a search warrant or by any exception to the
warrant requirement.
Dickerson, 508 U.S. at 378-79. Concluding that the search of
Dickerson was “very similar” to the search of the stereo
equipment, id. at 379, the Court held that the disputed search of
Dickerson’s pocket was unconstitutional.
Although the officer was lawfully in a position to feel the
lump in respondent’s pocket, because Terry entitled him to
place his hands upon respondent’s jacket, the court below
determined that the incriminating character of the object
was not immediately apparent to him. Rather, the officer
determined that the item was contraband only after
conducting a further search, one not authorized by Terry or
by any other exception to the warrant requirement. Because
this further search of respondent’s pocket was
constitutionally invalid, the seizure of the cocaine that
followed is likewise unconstitutional.
Id.
The disputed search of appellant in this case cannot be
distinguished in any meaningful way from the impermissible
 23

search in Dickerson. Here, as in Dickerson, neither the
constitutionality of the initial stop nor the protective frisk was at
issue. In addition, neither the frisk of Dickerson nor the frisk of
appellant produced any evidence of a weapon. Nevertheless, in
each case, the police undertook a further search aimed at
determining whether certain physical evidence would identify
the stopped individual as the perpetrator of the crime in
question. In Dickerson, the police officer manipulated the
outside of the suspect’s jacket pocket to determine whether the
small lump in that pocket felt like crack cocaine. Here, the
police manipulated appellant’s jacket, partially unzipping and
opening it, so that the complainant could see the sweatshirt
underneath appellant’s jacket. In each case, the goal of the
officer was to obtain information about a physical object in the
suspect’s possession that the officer believed might identify the
suspect as having committed the crime in question. In other
words, the officer’s goal in each case was to determine whether
the object of the search, the lump in Dickerson and the
sweatshirt here, had particular incriminating characteristics that
would contribute to a probable cause determination. And in
each case, the officer sought to accomplish this goal pursuant to
a search that exceeded the bounds of Terry.
The Government’s attempt to distinguish Dickerson is
entirely unavailing. The sum and total of the Government’s
argument consists of a single sentence:
This case is critically different from Dickerson, because this
case does not involve an intrusive search of the person for
evidence, but rather involves the very minimal intrusion of
partially unzipping a coat to reveal a sweatshirt, not as part
of a general search for evidence, but rather as a reasonable
incident of an entirely permissible show-up identification
procedure.
 24

Gov’t En Banc Br. at 38-39. This bald assertion is unsupported
by any further explanation and is quite wrong in what it
suggests.
The assertion that the impermissible search of Dickerson
was more intrusive than the search of appellant is specious. If
anything, the search of appellant was more intrusive. First, as
the Supreme Court pointed out, when the officer in Dickerson
undertook the impermissible search, he was at least “lawfully in
a position to feel the lump in [Dickerson’s] pocket,” having just
finished a pat down. 508 U.S. at 379. Moreover, throughout the
search in Dickerson, the officer never strayed from the outside
surface of the suspect’s jacket pocket. He did not open the
pocket. He did not look into it. And he did not reach inside the
jacket to feel the pocket’s contents. Thus, while the officer in
Dickerson felt the lump through the jacket pocket, he did not
physically penetrate the outer surface of the jacket. Rather, he
simply “squeez[ed], slid[], and otherwise manipulat[ed] the
contents of the defendant’s pocket.” Id. at 378 (quotation marks
omitted). And, significantly, his actions did not reveal the
contents of the pocket to the public at large.
Here, when Officer Willis unzipped to the waist appellant’s
fastened jacket so that the complainant could see what was
underneath, he not only penetrated the outer layer of appellant’s
clothing, he actually physically peeled a portion of it back. In
contrast to the officer in Dickerson, Officer Willis also exposed
what lay under that outer layer to the public at large.
Consequently, in addition to exposing the sweatshirt to the
complainant, the search of appellant’s jacket necessarily
exposed whatever else appellant had under that portion of his
jacket to whomever was present on or looking at the street when
his jacket was unzipped. The search here thus involved a greater
invasion of the person than the manipulation of the outer surface
of Dickerson’s jacket pocket.
 25

The Government’s second distinction, that the search of
Dickerson was a “general search for evidence,” while the
unzipping of appellant’s jacket was “a reasonable incident of an
entirely permissible show-up identification procedure,” Gov’t
En Banc Br. at 39, is, at best, puzzling. When pressed at oral
argument regarding this alleged distinction between what the
police were searching for in Dickerson and what they were
searching for here, Government counsel asserted that the search
of appellant was different from the search of Dickerson, because
the search of Dickerson was a “full-blown evidentiary search,”
Tr. of En Banc Argument (Oct. 11, 2007) at 44, “a pure
evidentiary search for contraband,” id. at 43. In contrast,
according to counsel, the officers here “weren’t trying to recover
physical evidence.” Id. at 43; see also id. at 45.
The Government’s argument is unpersuasive. Certainly if
the complainant had identified the appellant as her assailant on
the basis of his sweatshirt, that sweatshirt would have been
seized as physical evidence of appellant’s guilt. Moreover, it is
clear that the search here was intended to reveal evidence – both
physical evidence (the sweatshirt) and testimonial evidence (the
complaint’s identification of it) – that would either support the
probable cause necessary to arrest appellant or dispel the
officer’s reasonable suspicion that appellant was the robber. It
is simply impossible for us to ascertain how that differs from the
Dickerson officer’s attempt to uncover tactile evidence
regarding the lump in the suspect’s pocket that would support
the probable cause necessary to effectuate an arrest or dispel the
officer’s suspicion that Dickerson was committing a narcotics
offense.
C. The Government’s Arguments for an Investigative
Identification Search Exception Are Not Supported by
Precedent
There is no Supreme Court or federal appellate case law
supporting the search of an individual stopped only on
 26

reasonable articulable suspicion after a pat down of that
individual has produced no evidence of a weapon. At oral
argument, Government counsel conceded that there is no such
precedent. Tr. of En Banc Argument at 51-52.
Absent such precedent, the Government attempts a three-
part argument in support of a wholly new investigative
identification search exception to the warrant and probable cause
requirements. First, it points to cases in which the Supreme
Court has used the balancing test to assess the Fourth
Amendment permissibility of what the Government
characterizes as “a wide array” of warrantless “governmental
intrusions based on something less than probable cause.” Gov’t
En Banc Br. at 19-21. Second, it seeks support in several
Supreme Court cases indicating that “police officers may take
reasonable steps necessary to facilitate a brief investigation
during a Terry stop.” Id. at 21-22. Third, it asserts that the
Supreme Court has approved investigative measures, similar to
the unzipping of appellant’s jacket, that are related to
identification issues. Id. at 24-28. None of the cited cases,
either singularly or in combination, support the permissibility of
the non-protective evidentiary search at issue here.
1. The “Balancing-Test” Cases Do Not Support an
Investigative Identification Search Exception
Unsurprisingly, given the Court’s pronouncements in
Dickerson and Sibron, none of the balancing-test cases cited by
the Government involve the permissibility of a search during a
run-of-the-mill Terry stop. Rather, the cited cases fall into one
of three categories, each of which is irrelevant to the issue here.
In one category, the Court relies on the balancing test to
extend Terry’s safety rationale to new and limited settings. See,
e.g., Maryland v. Buie, 494 U.S. 325, 327 (1990) (a “protective
sweep . . . incident to an arrest” does not violate the Fourth
Amendment “if the searching officer possesse[s] a reasonable
 27

belief . . . that the area swept harbor[s] an individual posing a
danger to the officer or others”); Pennsylvania v. Mimms, 434
U.S. 106, 110-11 (1977) (“inordinate risk confronting an
officer” justifies requiring a driver who is stopped for traffic
violation to step out of the car).
In the second category, the Court uses the balancing test to
assess the permissibility of searches of individuals who have a
lessened expectation of privacy as a result of governmental
supervision to which they were legitimately subject at the time
of the search. See, e.g., United States v. Knights, 534 U.S. 112,
119-22 (2001) (defendant’s “status as a probationer subject to a
search condition” results in a diminished expectation of privacy
that does not outweigh governmental interest in crime
prevention and rehabilitation); O’Connor v. Ortega, 480 U.S.
709, 725-26 (1987) (“Balanced against the substantial
government interests” of public employers, “the privacy
interests of government employees in their place of work . . . are
far less than those found at home or in some other contexts.”);
New Jersey v. T.L.O., 469 U.S. 325, 341 (1985) (“[T]he
accommodation of the privacy interests of school children with
the substantial need of teachers and administrators for freedom
to maintain order in the schools does not require strict adherence
to the requirement that searches be based on probable cause to
believe that the subject of the search has violated or is violating
the law. Rather, the legality of a search of a student [in a public
school] should depend simply on the reasonableness, under all
the circumstances, of the search.”).
The remainder of the cases relied on by the Government to
support its assertion that the reasonableness balancing test
provides the means by which we should assess the unzipping of
appellant’s jacket are inapt in that they do not involve searches.
Rather, these cases simply stand for the proposition that, in
certain situations, a seizure – if limited enough in scope – may
be found reasonable though based on something other than
 28

probable cause to believe that criminal activity is afoot. See,
e.g., Illinois v. Lidster, 540 U.S. 419 (2004) (highway
checkpoint to locate a hit and run driver found reasonable
because contact with police – which consisted of a request for
information and the distribution of a flyer – lasted no more than
a few seconds and did not involve a search for evidence); United
States v. Place, 462 U.S. 696 (1983) (seizure of luggage based
on reasonable articulable suspicion that it contains narcotics may
be permissible if detention is limited in duration and
investigative measures used to confirm or dispel suspicion do
not include a search of contents); United States v. Martinez-
Fuerte, 428 U.S. 543 (1976) (highway checkpoints at border
requiring vehicle occupants to respond to brief questions and
possibly produce documents evidencing a right to be in the
United States found reasonable where neither the vehicle nor its
occupants were searched).
Clearly none of these cases support the Government’s
argument that the unzipping of appellant’s jacket was “a
reasonable, de minimis investigative measure that appropriately
facilitated the show-up procedure.” Gov’t En Banc Br. at 13.
The first set of cases merely allows for limited intrusions
carefully tailored to reveal or eliminate an officer’s reasonable
articulable suspicion that a person or place poses a danger to
himself or others nearby. The inapplicability of the second set
of cases is easily demonstrated by considering how any one of
the defendants in those cases would have been treated had they
not legitimately been subject to increased governmental
supervision. For example, had the fourteen-year-old public
school student in T.L.O. left school property, it is clear that no
police officer could have permissibly searched her purse merely
because she had been seen smoking in the school lavatory. The
third set of cases is similarly wholly irrelevant to the issue
before us in that none of the cases involve a search of any kind.
 29

2. Cases Indicating that Police Officers May Take Certain
Steps to Facilitate a Brief Investigation During a Terry
Stop Do Not Support an Identification Search
Exception
None of the cases on which the Government relies to
support its second premise – that police officers may take
reasonable steps necessary to facilitate a brief investigation
during a Terry stop – lend support to its ultimate conclusion that
the unzipping here was a permissible investigative step. This is
because none of the permissible investigatory measures in the
cases cited by the Government involved an evidentiary search.
As described in Michigan v. Summers, 452 U.S. 692 (1981),
permissible investigative measures in the context of a Terry stop
consist of questioning (which may include a request for
identification or inquiry concerning the suspicious conduct of
the person detained); communication with police or private
citizens to verify explanations, confirm identification, or
determine whether a person of the proffered identity is otherwise
wanted; a short detention while police check premises and talk
to others to determine whether, in fact, an offense has occurred;
the examination of objects abandoned by the suspect; and, if it
is known that an offense was committed in the area, the viewing
of the suspect by witnesses to the crime. Id. at 700 n.12.
The only additional investigative steps referenced by the
Government are those related to the seizure of a traveler’s
baggage permitted when an officer has reasonable articulable
suspicion to believe that it contains narcotics. See Gov’t En
Banc Br. at 22 (citing Place, 462 U.S. at 703-04). In Place,
though, the Court was careful to point out that if the
investigative procedure for which luggage is seized is “itself a
search requiring probable cause, the initial seizure of [the]
luggage . . . – no matter how brief – [can]not be justified on less
than probable cause.” 462 U.S. at 706. The Court went on to
explain that the dog sniff test for narcotics, to which Place’s
 30

luggage was subject, was not a search, because it neither
required that the luggage be opened, nor exposed non-
contraband items that otherwise would have remained shielded
from public view. Id. at 707. Because the unzipping here was
a search, Place supports the conclusion that the unzipping of
appellant’s jacket was not a permissible investigative measure.
3. The Investigative Measures Cases Pertaining to
Identification Issues Provide No Support for the
Government’s Identification Search Exception
The Government principally relies on four cases in support
of its argument that searches narrowly tailored to facilitate a
show-up procedure are permissible investigative measures. One
of those cases, New York v. Class, 475 U.S. 106 (1986), is not
only distinguishable, but actually supports the opposite of what
the Government claims. Two others, United States v. Dionisio,
410 U.S. 1 (1973), and United States v. Mara, 410 U.S. 19
(1973), lend no support whatsoever to the Government’s
argument, because they involved neither a search nor an
investigative step taken in the context of a Terry stop. That
leaves the dicta in Hayes v. Florida, 470 U.S. 811, 816-17
(1985). And the language in Hayes to which the Government
points simply cannot take the Government where it needs to go
to prevail on the facts presented here.
a. Class Supports the Exclusion of the Gun Evidence
In Class, the Supreme Court determined that a police officer
conducted a search when, during a traffic stop, he reached into
the defendant’s vehicle to move some papers that obscured the
vehicle identification number (VIN) on the dashboard. 475 U.S.
at 114-15. In so doing, the officer saw a gun under the driver’s
seat. Id. at 108. As described by the Government in its en banc
brief, “[t]he Supreme Court concluded that the officer’s actions
were reasonable, after applying a ‘balancing’ test that
considered, among other things, that ‘the search was focused on
 31

its objective [of revealing the VIN] and no more intrusive than
necessary to fulfill that objective;’ and that the search was ‘far
less intrusive than a formal arrest.’” Gov’t En Banc Br. at 27-28
(quoting Class, 475 U.S. at 118) (alterations in original).
“Thus,” according to the Government, Class approved a limited
investigative search “similar” to the unzipping of appellant’s
jacket, “without requiring probable cause.” Gov’t En Banc Br.
at 28. The Government’s characterization badly distorts the
Court’s reasoning and is simply wrong in its conclusion.
The Class opinion, in fact, rests on three factors – the
impossibility of having a reasonable expectation of privacy in a
VIN, officer safety, and the existence of probable cause – that
not only render the balance struck there inapplicable to the case
here, but actually support the conclusion that the unzipping of
appellant’s jacket was impermissible. Because there is no
expectation of privacy in VINs, the intrusion at issue in Class
was not the moving of the papers covering the VIN, but rather
the officer’s reach into the car to move those papers. See Class,
475 U.S. at 114-15. The inquiry was made necessary by the fact
that the officer had prevented the respondent, who had alighted
from his car when the police pulled him over, from getting back
into the car to move the papers. See id. at 115. Noting that
under Pennsylvania v. Mimms, 434 U.S. 106, safety concerns
permitted the officer to detain the driver outside of the car, the
Court explained that the question at issue was whether the
officer additionally could “effect a search for the VIN that may
have been necessary only because of that detention.” Class, 475
U.S. at 116.
The Court concluded that the search was permissible. It
said: “In light of the danger to the officers’ safety that would
have been presented by returning respondent immediately to his
car, we think the search to obtain the VIN was not prohibited by
the Fourth Amendment.” Id. However, in describing the
balancing by which it reached this conclusion, the Court made
 32

clear that because the safety concern at issue arose not from a
particularized belief that the respondent had a weapon, but rather
from the more generalized concern for officer safety that
motivated the Court in Mimms, id. at 115-16, the Government
was required to demonstrate some probable cause focusing
suspicion on the individual affected by the search. Id. at 117-18.
The Court explained:
When a search or seizure has as its immediate object a
search for a weapon, . . . we have struck the balance to
allow the weighty interest in the safety of police officers to
justify warrantless searches based only on a reasonable
suspicion of criminal activity. See Terry v. Ohio, [392 U.S.
at 21]; Adams v. Williams, 407 U.S. 143 (1972). Such
searches are permissible despite their substantial
intrusiveness. See Terry v. Ohio, [392 U.S. at] 24-25
(search was “a severe, though brief, intrusion upon
cherished personal security, and . . . must surely [have]
b[een] an annoying, frightening, and perhaps humiliating
experience”).
When the officer’s safety is less directly served by the
detention, something more than objectively justifiable
suspicion is necessary to justify the intrusion if the balance
is to tip in favor of the legality of the governmental
intrusion. In Pennsylvania v. Mimms, [434 U.S.] at 107, the
officers had personally observed the seized individual in the
commission of a traffic offense before requesting that he
exit his vehicle. In Michigan v. Summers, 452 U.S. 692,
693 (1981), the officers had obtained a warrant to search the
house that the person seized was leaving when they came
upon him.
Id. at 117 (parenthetical alterations in original).
Although the Class Court acknowledged the difference
between the facts before it and the facts in Mimms and Summers,
 33

it concluded that the balance struck in those cases supported the
permissibility of the officer’s intrusion into the car in the case
before it. Id. at 117-18. As the Court explained: “All three of
the factors involved in Mimms and Summers are present in this
case: [1] the safety of the officers was served by the
governmental intrusion; [2] the intrusion was minimal; and [3]
the search stemmed from some probable cause focusing
suspicion on the individual affected by the search. Indeed, here
the officers’ probable cause stemmed from directly observing
respondent commit a violation of the law,” which, the Court
noted, would have justified his arrest. Id. at 118 (emphasis
added).
The Class Court thus held that the officer’s search for the
VIN number “was sufficiently unintrusive to be constitutionally
permissible in light of the lack of a reasonable expectation of
privacy in the VIN and the fact that the officers observed
respondent commit two traffic violations. Any other
conclusion,” the Court found, “would expose police officers to
potentially grave risks without significantly reducing the
intrusiveness of the ultimate conduct – viewing the VIN – which
. . . the officers were entitled to do as part of an undoubtedly
justified traffic stop.” Id. at 119 (emphasis added).
Here, in contrast, there was no officer safety interest served
by the unzipping. The officers did not have probable cause
focusing suspicion on appellant, but rather only a reasonable
suspicion of criminal activity, which the Class opinion makes
clear cannot justify a search that does not have a weapon as its
“immediate object.” See id. at 117. And, perhaps most
important, appellant clearly had a reasonable expectation of
privacy in what lay underneath his jacket. The objective of the
search was not, in other words, to reveal an object in which no
one can have a reasonable expectation of privacy.
 34

b. Dionisio and Mara, Which Involved Grand Jury
Subpoenas for Evidence in Which There Was No
Reasonable Expectation of Privacy, Are
Inapposite
Dionisio and Mara also lend no support to the
Government’s claim, because neither case involved a search or
a police action during an on-the-street Terry stop. Rather, both
decisions addressed the permissibility of grand jury subpoenas
seeking evidence in which there can be no expectation of
privacy. In Dionisio, 410 U.S. 1, the Court considered the
permissibility of a subpoena requiring an individual to produce
a voice exemplar for comparison to certain incriminating tapes.
In Mara, 410 U.S. 19, it examined the permissibility of a
subpoena to compel production of a potentially incriminating
handwriting exemplar. In Dionisio, the Court first concluded
that a grand jury subpoena, unlike a Terry detention, is not a
Fourth Amendment seizure. 410 U.S. at 9. The Court held that
this was so, at least in part, because grand jury subpoenas are
subject to judicial supervision and do not involve the same type
of often forceful and demeaning compulsion as investigative
stops and arrests. Id. at 9-10. The Court then explained that
production of a voice exemplar is not a search, since no
individual can have a reasonable expectation of privacy in the
physical characteristics of his tone of voice. Id. at 14. “Like a
man’s facial characteristics, or handwriting, his voice is
repeatedly produced for others to hear. No person can have a
reasonable expectation that others will not know the sound of his
voice, any more than he can reasonably expect that his face will
be a mystery to the world.” Id.; see also Mara, 410 U.S. at 21
(“Handwriting, like speech, is repeatedly shown to the public,
and there is no more expectation of privacy in the physical
characteristics of a person’s script than there is in the tone of his
voice.”).
 35

The same cannot be said of a piece of clothing when the
only information that the police have about that clothing is that
the wearer has chosen to shield most of it from public view. A
sweatshirt, unlike an individual’s tone of voice, handwriting, or
facial characteristics, is not invariably revealed to the public in
the course of a person’s daily life.
c. The Hayes v. Florida Dicta Does Not Extend to
Investigative Measures Involving Searches
The dicta in Hayes v. Florida provides no support for the
unzipping of appellant’s jacket. In Hayes, the Court held that
the police may not transport a suspect to the police station for
fingerprinting without probable cause. The dicta to which the
Government points merely suggests that “a brief detention in the
field for the purpose of fingerprinting” may be permissible
“where there is only reasonable suspicion not amounting to
probable cause.” 470 U.S. at 816. The Court made it clear,
however, that such an investigative action might be permissible
only so long as (1) “there is reasonable suspicion that the
[stopped] suspect has committed a criminal act” (in other words,
so long as there is reasonable articulable suspicion justifying the
stop of the suspect), (2) “there is a reasonable basis for believing
that fingerprinting will establish or negate the suspect’s
connection with that crime,” and (3) “the procedure is carried
out with dispatch.” Id. at 817. In support of these criteria, the
Court cited to United States v. Place, 462 U.S. 696 (1983).
In Place, the Court determined that the seizure of a
traveler’s luggage on less than probable cause may be
reasonable so long as the “observations [of the seizing officer]
lead him reasonably to believe that . . . [the] luggage . . .
contains narcotics,” 462 U.S. at 706, and the detention is brief,
id. at 707-10. Most important, for purposes of the issue before
this court, the Place opinion made it clear that when a seizure is
based on less than probable cause, the accompanying
 36

investigative methods may not include a search. As the Court
explained:
Obviously, if [the] investigative procedure [at issue here –
a dog sniff –] is itself a search requiring probable cause, the
initial seizure [of the luggage] for the purpose of subjecting
it to the sniff test – no matter how brief – could not be
justified on less than probable cause.
Id. at 706. The Court went on to find that dog sniff tests are not
searches because they (1) do not require the opening of luggage
and thus do not expose to public view non-contraband items that
would otherwise remain hidden and (2) do not reveal to
authorities any information about the contents of a piece of
luggage beyond the presence or absence of illegal narcotics. Id.
at 707. “In these respects,” the Court concluded, “the canine
sniff is sui generis. We are aware of no other investigative
procedure that is so limited both in the manner in which the
information is obtained and in the content of the information
revealed by the procedure.” Id.
In Cupp v. Murphy, 412 U.S. 291 (1973), the Supreme
Court had earlier indicated that fingerprinting is another
investigative measure that, like a dog sniff, is not a search
requiring probable cause. The Court there distinguished
fingerprinting from the taking of fingernail scrapings:
Unlike the fingerprinting in Davis, the voice exemplar
obtained in United States v. Dionisio, [410 U.S. 1], or the
handwriting exemplar obtained in United States v. Mara,
410 U.S. 19, 93, the search of the respondent’s fingernails
went beyond mere “physical characteristics . . . constantly
exposed to the public.”
Id. at 295 (quoting United States v. Dionisio, 410 U.S. at 14)
(alterations in original). Cupp thus relied on the Court’s
suggestion in Davis v. Mississippi that fingerprinting is not a
search, because it “involves none of the probing into an
 37

individual’s private life and thoughts that marks an interrogation
or search.” Davis, 394 U.S. at 727. Describing the holding in
Davis, the Hayes Court used virtually the same language. See
470 U.S. at 814 (“fingerprinting, because it involves neither
repeated harassment nor any of the probing into private life and
thoughts that often marks interrogation and search, represents a
much less serious intrusion upon personal security than other
types of searches and detentions”).
Clearly, because the investigative method employed in this
case – the unzipping of appellant’s jacket – did constitute a
search, it cannot be permissible under the criteria set forth in
Hayes, Place, and Cupp.
D. Even if the Hayes Dicta Could Be Read to Allow Certain
Searches, It Cannot Support the Unzipping of Appellant’s
Jacket Because There Was No Reasonable Basis for
Believing That the Unzipping Would Establish Or Negate
Appellant’s Connection to the Robbery Under
Investigation
Even assuming, arguendo, that the Hayes dicta might
permit the police to unzip a suspect’s jacket to facilitate a show-
up if, inter alia, “there is a reasonable basis for believing” that
doing so “will establish or negate the suspect’s connection with
th[e] crime” under investigation, Hayes, 470 U.S. at 817, the
unzipping in this case would not qualify. There is nothing in the
record to suggest that the police had a reasonable basis for
believing the complainant’s viewing of the generic blue
sweatshirt worn by appellant would establish or negate his
connection with the robbery.2 Indeed, there is no finding – and

2
The dissent notwithstanding, this is a point that appellant
expressly made to the panel on appeal. See Appellant Panel Reply Br.
at 5 & n.1 (arguing that “justification for creating a new exception to
the Constitution’s prohibition against warrantless searches” is
“particularly lacking in this case in which the clothing that the officer
 38

there was no suggestion – that the complainant said the “blue
sweatshirt” worn by the robber was distinctive in any way that
might have distinguished it from others, let alone in a way that
would have established or negated a wearer’s connection to the
crime. Cf. Hayes, 470 U.S. at 812 (police sought suspect’s
fingerprints to compare with latent prints found at rape scene);
Class, 475 U.S. at 111 (police sought to view vehicle
identification number). Nor did the witness even suggest that
seeing whether appellant was wearing such a sweatshirt would
enable her to make an identification when seeing appellant’s
face had not, although we need not decide here whether that
would have made a difference. Hence, even if the Government
is correct that Hayes can be read to permit a search to facilitate
a show-up under certain circumstances, the factual predicate for
application of the Hayes dicta is absent here.3

sought to reveal was a blue sweatshirt under a dark jacket, not
distinctive clothing that might have distinguished appellant from
others”).
3
Part III.D should not be read to suggest that if the police
officers in this case had a reasonable basis for believing that unzipping
Askew’s jacket could merely help facilitate the witness’ identification,
the Hayes dicta would be satisfied. That is not the holding of the
court. Rather, what we have said is that, assuming the Hayes dicta
applies, the police in this case did not have a reasonable basis for
believing that unzipping appellant’s jacket would establish or negate
appellant’s connection with the robbery for which he was stopped.
Six members of the court – a majority – join in Part III.D, equally
convinced that the Government’s argument resting on the Hayes dicta
holds no water in this case.
 39

E. The Government’s Alternative Argument, That the
Unzipping Can Be Justified as an Objectively Reasonable
Continuation of the Protective Frisk, Is Unsupported by
the District Court’s Factual Findings
The Government’s alternative argument, that the unzipping
can be justified as an objectively reasonable response to
appellant’s alleged resistance during the pat down, is without
support in the District Court’s factual findings. The
Government never challenged the District Court’s findings as
clearly erroneous. Quite the contrary, it described those findings
as “consistent with the government’s evidence at the suppression
hearing.” Gov’t Panel Br. at 9. And pursuant to those findings,
we can ascertain no basis for concluding, as the Government
urges, that “it was objectively reasonable for Officer Willis to
complete the protective frisk of appellant . . . by unzipping
appellant’s jacket.” Gov’t Supplemental Br. at 9.
The essence of the Government’s argument is that when the
complaining witness was brought to the scene, Officer Koenig
broke off the pat down of appellant so that appellant could be
presented to the complaining witness for identification. Id. at 7.
According to the Government, prior to the show-up, appellant
had resisted Officer Koenig’s attempt to pat him down. Id. at 7-
8, 11. The Government thus contends that because appellant
“thwarted” Officer Koenig’s pat down of his outer clothing, “it
was objectively reasonable for [Officer Willis] to unzip
appellant’s jacket to determine whether he had a weapon in his
waist area.” Id. at 11. In other words, the Government argues
that while Officer Willis’s subjective intent in unzipping
appellant’s jacket during the show-up was to reveal to the
complaining witness what appellant wore underneath the jacket,
that unzipping was an objectively reasonable continuation of
Officer Koenig’s allegedly thwarted pat down for weapons.
Tellingly, the Government does not cite to the District
Court’s factual findings in support of this argument. Rather, it
 40

cites to testimony that the District Court noted, but did not
credit. See id. at 7. Thus, contrary to what the Government
suggests, the District Court did not find that appellant resisted or
thwarted Officer Koenig’s pat down. Rather, referring in a
footnote to the testimony on which the Government rests its
argument, the District Court merely noted that Officer Willis
had “suggested that Officer Koenig had not completed the pat-
down, perhaps because of some resistance by the defendant.”
Askew, 313 F. Supp. 2d at 3 n.2 (emphasis added). In other
words, the District Court declined to find either that the pat
down was incomplete or that appellant had in any way resisted
the pat down. Indeed, the court explicitly characterizes Officer
Willis’s testimony on these two points as a suggestion. Then,
pointing to “[t]he government acknowledge[ment] that when
Officer Koenig patted the defendant down, he did not find
anything,” the court reiterates its own finding that “[t]he
subsequent discovery of the gun at issue was not the result of
this pat-down.” Id.; see also id. at 4 (“The initial pat-down of
Officer Koenig did not reveal the presence of any
weapon . . . .”). Thus, while we agree that the standard for
determining reasonableness in the Fourth Amendment context
is an objective one, see Whren v. United States, 517 U.S. 806
(1996); United States v. (Rocky Lee) Brown, 334 F.3d 1161,
1166-67 (D.C. Cir. 2004), we find that the District Court’s
unchallenged factual findings provide no support for the
Government’s argument that the unzipping here was objectively
reasonable.4

4
Our dissenting colleagues disagree with our reading of the
District Court’s opinion. But whatever the District Court did, it
certainly did not make findings to support a safety-unzipping version
of events, and we have no authority to make our own findings of fact.
See United States v. Burke, 888 F.2d 862, 868-69 (D.C. Cir. 1989)
(“The United States suggests that we might infer from the record that
Burke was aware of the gun found in his tote bag. To engage in this
 41

Even if the Government had challenged the District Court’s
factual findings as clearly erroneous, it could not succeed on its
argument that the unzipping was an objectively reasonable
continuation of the pat down, since that argument is based on a
wholly implausible reading of the record. Consider the illogic
of the Government’s proposed characterization of Officer
Willis’s testimony: Officers Koenig and Willis initiate a frisk,
appellant places his hands on a police cruiser, and Officer
Koenig pats down appellant’s outer garments. But when Officer
Koenig reaches appellant’s waist, appellant leans repeatedly
against the cruiser. Appellant thus “thwarts” Officer Koenig’s
efforts to frisk appellant’s waist area and consequently provides
what the Government describes as an objectively reasonable
basis for undertaking a protective search for a weapon. But, at
this point, because the pat down is “interrupted” by the arrival
of another officer with the complaining witness, the officers do
not undertake a protective search of appellant’s waist area,
complete the frisk, or otherwise act to ensure their safety and the
safety of the complainant. Rather, they walk toward the
complainant with an unhandcuffed robbery suspect who has
successfully prevented them from completing a frisk of his
person for weapons. See Gov’t Supplemental Br. at 7-8. This
scenario makes no sense. Consequently, even if the
Government had framed the issue pursuant to the correct
standard of review, it could not have “definitely and firmly”
convinced us that the District Court made a mistake when it
chose not to credit Officer Willis’s “suggest[ion] that Officer
Koenig had not completed the pat-down, perhaps because of
some resistance by the defendant, when Officer Benton arrived
with the robbery victim for a show-up.” Askew, 313 F. Supp. 2d
at 3. n.2 (emphasis added).

sort of de novo factfinding, however, would be wholly inconsistent
with the function of an appellate court.”).
 42

The total implausibility of the Government’s reading of the
record may explain why it abandoned any safety rationale before
the District Court. In arguing its case at the close of the
suppression hearing, the Government never suggested that there
was anything about the pat down that provided objectively
reasonable grounds for continuing the frisk or undertaking a
protective search. Rather, dismissing the possibility of treating
the initial unzipping as anything other than an act to facilitate
the show-up, the Government confirmed for the District Court
that Officer Koenig “didn’t feel anything during the traditional
pat-down that would have permitted [the officers] to go into the
jacket at that point, but rather they then had the show-up.” Mar.
26 Tr. at 38. The impossibility of reading the record as the
Government now suggests may also explain why before the
panel, as well as in its principal brief before the en banc court,
the Government argued that the “sole” justification for the
unzipping was to facilitate the show-up. See Gov’t Panel Br. at
18; Gov’t En Banc Br. at 22.5

5
In asserting that the Government “pressed” the safety rationale
in the District Court, our dissenting colleagues ignore the fact that the
Government abandoned the safety rationale before the District Court.
The Government’s safety rationale was proffered before the
completion of the two-day suppression hearing at a time when, by its
own admission, the Government was unsure of what its officers would
say regarding the circumstances surrounding the pat down. See, e.g.,
Additional Case Law in Support of Government’s Opposition at 2 &
n.1(Mar. 19, 2004). Once Officer Willis testified to explain what
transpired, see Mar. 26 Tr. at 8 (“I went to unzip [the jacket] down so
that the show-up, they could see what he had on.”), the Government
consistently eschewed a safety rationale argument. Thus, in its closing
arguments and supplemental filing before the District Court, its brief
to the three-judge appellate panel, and its en banc brief, the
Government relied solely on the argument that the initial partial
unzipping of appellant’s jacket was reasonable under the Fourth
Amendment because it was a “minimal intrusion” that “facilitated a
 43

Although we do not dispute that a protective search may be
lawful when a suspect prevents the police from performing a
Terry frisk, we conclude that neither the District Court’s factual
findings nor the evidentiary record support the conclusion that
such was the situation here.
IV. CONCLUSION
On the record of this case, the police officers violated
appellant’s Fourth Amendment rights by unzipping his jacket
without his permission and without probable cause or a warrant.
The judgment of the District Court must therefore be reversed
and the case remanded.

So ordered.

show-up identification procedure by exposing clothing that matched
the lookout description of a robbery suspect,” Gov’t Panel Br. at v;
see also id. at 12, 18; Gov’t En Banc Br. at vii, 13-15, 22-28; Gov’t
Panel Br. at 8 (summarizing the Government’s argument before the
District Court); Gov’t En Banc Br. at 7 (same); Second Filing of
Additional Case Law in Support of Government’s Opposition at 4
(Mar. 31, 2004); Mar. 26 Tr. at 24, 26-27, 38. Not until after the en
banc argument, when this court issued a supplemental briefing order
asking the parties to address the possibility of justifying the initial
unzipping on a safety rationale, did the Government again make the
argument.
 GRIFFITH, Circuit Judge, with whom Circuit Judges
ROGERS and TATEL join except as to footnote 2, concurring: I
agree that police officers violated Paul Askew’s Fourth
Amendment rights and so join the majority opinion in almost
every respect. Given the importance of this appeal, I write
separately to clarify my rationale for rejecting the government’s
principal argument before the en banc court. Hayes v. Florida
noted in dicta that a brief detention for the purpose of
fingerprinting a suspect is not necessarily an unconstitutional
seizure. 470 U.S. 811, 816–17 (1985). The government argues
that this dicta created a new exception to the probable cause
requirement for evidentiary searches. Nothing the Supreme
Court said in Hayes or its progeny supports this argument, and it
is not our prerogative to create a new exception to the probable
cause requirement of the Fourth Amendment. We are bound by
Supreme Court precedent to subject the unzipping of Askew’s
jacket to “the textual and traditional standard of probable
cause.” Arizona v. Hicks, 480 U.S. 321, 329 (1987). Because the
unzipping cannot satisfy that standard, I agree that we should
reverse the judgment of the district court.

* * *

When police officers unzipped Askew’s jacket, they
conducted a search of his person. Where zipping up creates a
“reasonable expectation of privacy,” Katz v. United States, 389
U.S. 347, 361 (1967) (Harlan, J., concurring), as it does in this
case, to unzip is to search. See, e.g., New Jersey v. T.L.O., 469
U.S. 325, 347 (1985) (unzipping purse compartment); New York
v. Belton, 453 U.S. 454, 456, 462–63 (1981) (unzipping jacket
pocket); United States v. Brown, 671 F.2d 585, 586 (D.C. Cir.
1982) (per curiam) (unzipping pouch); United States v. Waller,
426 F.3d 838, 844 (6th Cir. 2005) (unzipping luggage). That the
unzipping may have been minimally intrusive is immaterial. See
Hicks, 480 U.S. at 328–29. The Fourth Amendment prohibits
such a search in the absence of probable cause. See Minnesota v.
 2
Dickerson, 508 U.S. 366, 373 (1993); Hicks, 480 U.S. at 325–
29; Almeida-Sanchez v. United States, 413 U.S. 266, 269–70
(1973); Chambers v. Maroney, 399 U.S. 42, 51 (1970); Sibron v.
New York, 392 U.S. 40, 62–66 (1968); Henry v. United States,
361 U.S. 98, 100–02, 104 (1959); Carroll v. United States, 267
U.S. 132, 149, 156 (1925).

The unzipping of Askew’s jacket does not fit within any of
the exceptions to the probable cause requirement that have been
created by the Supreme Court. As I read the cases, these
exceptions can be divided into two general categories. The
officer safety cases allow a minimally intrusive search that
protects officers from harm. See, e.g., Terry v. Ohio, 392 U.S. 1,
30 (1968) (protective frisk of outer clothing for weapons);
Maryland v. Buie, 494 U.S. 325, 334 (1990) (protective sweep
during home raid to prevent ambush); New York v. Class, 475
U.S. 106, 115–18 (1986) (officer not required to permit
suspect’s re-entry into car where weapon may be hidden);
Pennsylvania v. Mimms, 434 U.S. 106, 109–11 (1977) (per
curiam) (compelling suspect to alight from car during traffic
stop); United States v. Robinson, 414 U.S. 218, 234–35 (1973)
(search incident to arrest to disarm suspect before taking into
custody). The special needs cases allow a search that advances a
governmental end deemed by the Supreme Court to be weightier
than the typical law enforcement interest. See, e.g., United
States v. Knights, 534 U.S. 112, 120–22 (2001) (monitoring
probationers); Skinner v. Ry. Labor Executives’ Ass’n, 489 U.S.
602, 619–20 (1989) (preventing intoxicated employees from
operating railroads); T.L.O., 469 U.S. at 340–41 (protecting
schoolchildren); Camara v. Municipal Court, 387 U.S. 523,
534–39 (1967) (inspecting buildings for compliance with
municipal code). These two types of exceptions do not apply to
garden-variety searches for evidence, which are governed by the
probable cause requirement. See United States v. Colyer, 878
F.2d 469, 478 (D.C. Cir. 1989).
 3

The government concedes that probable cause to justify the
unzipping was absent but makes no effort to fit the search of
Askew within either category of established exceptions. Instead,
the government argues that the search fits within a new
exception created by Hayes for identification searches.1 I find no
basis in Hayes for that conclusion. The hunt for an
identification-search exception begins, like so many efforts to
weaken the probable cause requirement, with Terry v. Ohio. The
government asserts that the search of Askew’s person is not
limited by probable cause because the purpose of the unzipping,
identification of a suspect, was “reasonably related to the
purpose of the Terry stop.” Government’s En Banc Br. at 14.
Terry allows an officer lacking probable cause to make a “brief
stop [(i.e., seizure)] of a suspicious individual, in order to
determine his identity,” Adams v. Williams, 407 U.S. 143, 145–
46 (1972), so the government argues that a search calculated to
identify a suspect should also be permissible on less than
probable cause. The unspoken premise is that the power to
search a suspect is as broad as the power to seize him. But this
asks us to assume that whatever Terry said about “seizures”
applies equally to “searches.” In fact, Terry’s regime of stops
and frisks distinguishes between the two. An officer lacking
probable cause may seize a suspect to determine his identity, but
must limit his search to a protective frisk for weapons. See

1
The government has only recently come to regard Hayes as
significant, having relegated the case to a single “cf.” citation in its
brief to the three-judge panel. Government’s Br. at 20. Now that two
members of this court have opined that Hayes governs this matter,
United States v. Askew, 482 F.3d 532, 540–45 (D.C. Cir. 2007),
vacated and reh’g en banc granted, No. 04-3092 (D.C. Cir. July 12,
2007), the government has placed greater emphasis on the case. See
Government’s En Banc Br. at 12, 24, 25–26, 34 (citing Hayes); id. at
ii (identifying Hayes as a “[c]ase[] chiefly relied upon”).
 4
Minnesota v. Dickerson, 508 U.S. 366, 373 (1993); Michigan v.
Long, 463 U.S. 1032, 1052 n.16 (1983); Ybarra v. Illinois, 444
U.S. 85, 93–94 (1979); Dunaway v. New York, 442 U.S. 200,
210 (1979); Sibron v. New York, 392 U.S. 40, 65–66 (1968);
Terry, 392 U.S. at 30; see also 4 WAYNE R. LAFAVE, SEARCH
AND SEIZURE: A TREATISE ON THE FOURTH AMENDMENT
§ 9.6(g) (4th ed. 2004) (“[T]here is no search-for-evidence
counterpart to the Terry weapons search, permissible on only a
reasonable suspicion that such evidence would be found.”)
(quotation marks omitted). Terry allows an identification
seizure. It does not permit an identification search.

The government, forced to look beyond Terry for its new
exception, claims to have found it in the following language
from Hayes, 470 U.S. at 816–17 (citations omitted):

None of the foregoing implies that a brief detention in the
field for the purpose of fingerprinting, where there is only
reasonable suspicion not amounting to probable cause, is
necessarily impermissible under the Fourth Amendment. In
addressing the reach of a Terry stop in Adams v. Williams,
we observed that “[a] brief stop of a suspicious individual,
in order to determine his identity or to maintain the status
quo momentarily while obtaining more information, may be
most reasonable in light of the facts known to the officer at
the time.” Also, just this Term, we concluded that if there
are articulable facts supporting a reasonable suspicion that a
person has committed a criminal offense, that person may
be stopped in order to identify him, to question him briefly,
or to detain him briefly while attempting to obtain
additional information. There is thus support in our cases
for the view that the Fourth Amendment would permit
seizures for the purpose of fingerprinting, if there is
reasonable suspicion that the suspect has committed a
criminal act, if there is a reasonable basis for believing that
 5
fingerprinting will establish or negate the suspect’s
connection with that crime, and if the procedure is carried
out with dispatch. Of course, neither reasonable suspicion
nor probable cause would suffice to permit the officers to
make a warrantless entry into a person’s house for the
purpose of obtaining fingerprint identification.

The government contends that this passage, which nowhere
mentions “searches,” creates a new exception to the probable
cause requirement for searches that facilitate identification of a
suspect. See Government’s En Banc Br. at 26. I reject that
view.2

At issue in Hayes was the permissibility of seizing a person
and transporting him to the stationhouse to obtain his
fingerprints. The Supreme Court found the practice
unconstitutional in light of another case about seizures, Davis v.
Mississippi, 394 U.S. 721 (1969), which condemned a similar
instance of stationhouse fingerprinting. Subsequent citations to
Hayes confirm that the Court has only regarded it as a case
about seizures. See Hiibel v. Sixth Judicial Dist. Court, 542 U.S.
177, 187–89 (2004); Kaupp v. Texas, 538 U.S. 626, 630 n.2
(2003); Tennessee v. Garner, 471 U.S. 1, 8 (1985).

Hayes’s suggestion that fingerprinting might be permissible
during a Terry stop does not reach searches, because the
Supreme Court has never held that fingerprinting is a search.
See 1 LAFAVE § 2.2(d) (noting that “it is to be doubted that the

2
A majority of the court concludes that even if Hayes created an
identification-search exception, the government cannot prevail
because the police did not have a reasonable basis for believing that a
view of the sweatshirt beneath Askew’s jacket would establish or
negate Askew’s connection with the robbery. See supra Majority
Opinion, Part III.D. I decline to join this portion of the opinion.
 6
question of whether fingerprinting is a search can be taken as
settled”). The Supreme Court offered conflicting views of the
issue within Hayes itself. Compare Hayes, 470 U.S. at 814
(noting that fingerprinting does not involve “any of the probing
into private life and thoughts that often marks interrogation and
search”), with id. (implying that fingerprinting is a search by
contrasting the practice to “other types of searches”); see also
Cupp v. Murphy, 412 U.S. 291, 295 (1973) (contrasting
fingernail scraping, a search, to fingerprinting, not a search).
Hayes’s fingerprinting dicta dealt with what was then, and is
still, a non-search investigative step.

Hayes is about a seizure, not a search, so it cannot have
created a new exception that would permit the search of
Askew’s person in the absence of probable cause. The same is
true of other cases the government cites in support of its
purported identification-search exception. See Hiibel, 542 U.S.
at 187–89 (upholding state stop-and-identify law as authorizing
reasonable seizures); Illinois v. Lidster, 540 U.S. 419, 426–28
(2004) (allowing brief seizure at highway checkpoint to identify
witnesses to hit-and-run accident); United States v. Place, 462
U.S. 696, 703–04 (1983) (allowing brief seizure of luggage to
conduct dog sniff); Michigan v. Summers, 452 U.S. 692, 705
(1981) (allowing seizure of home occupants during execution of
search warrant).

My adherence to the textual distinction between “searches”
and “seizures” is consistent with the Supreme Court’s Fourth
Amendment jurisprudence. For example, the Court has justified
a judgment about seizures by noting that it was not dealing with
a search. United States v. Martinez-Fuerte, 428 U.S. 543, 561
(1976). The Court has also held that a search and a seizure
require different types of proof before they are permissible.
Consider Terry v. Ohio itself: A Terry stop (i.e., seizure)
requires reasonable suspicion that criminal activity is afoot,
 7
United States v. Cortez, 449 U.S. 411, 417–18 (1981), while a
Terry frisk (i.e., search) requires reasonable suspicion that the
stopped individual may have a weapon, Terry, 392 U.S. at 30. If
there were no meaningful distinction between searches and
seizures, then police would be justified in frisking every person
they stopped, yet this is not what the law allows. See Ybarra v.
Illinois, 444 U.S. 85, 92–93 (1979) (rejecting routine frisks
unsupported by “a reasonable belief that [the person searched]
was armed and presently dangerous”) (citing Adams, 407 U.S. at
146, and Terry, 392 U.S. at 21–24). Finally, as we noted in
United States v. Colyer, 878 F.2d 469, 479 (D.C. Cir. 1989), the
Supreme Court has created analytical frameworks for seizures
that it has never applied to searches:

Although there may be no compelling reason to
differentiate between seizures on the basis of their
intrusiveness and failing to likewise differentiate between
types of searches, the fact remains that we are unable to
point to a single Supreme Court case that has upheld a
search on reasonable suspicion merely because it was
minimally intrusive.

This differing treatment is sensible, as the separate protections
against unreasonable searches and unreasonable seizures
safeguard different interests. See Horton v. California, 496 U.S.
128, 133 (1990) (“The right to security in person and property
protected by the Fourth Amendment may be invaded in quite
different ways by searches and seizures. A search compromises
the individual interest in privacy; a seizure deprives the
individual of dominion over his or her person or property.”); see
also United States v. Jacobsen, 466 U.S. 109, 113 (1984).

* * *
 8
Hayes held that police cannot transport a suspect to the
stationhouse for fingerprinting without probable cause, consent,
or prior judicial authorization. 470 U.S. at 813–16. Its
observation that “a brief detention in the field for the purpose of
fingerprinting . . . is [not] necessarily impermissible under the
Fourth Amendment,” id. at 816, was unnecessary to decision of
the case and so was dicta. See Gersman v. Group Health Ass’n,
Inc., 975 F.2d 886, 897 (D.C. Cir. 1992); Cross v. Harris, 418
F.2d 1095, 1105 n.64 (D.C. Cir. 1969); Noel v. Olds, 138 F.2d
581, 586 (D.C. Cir. 1943). The government concedes this point,
Government’s En Banc Br. at 26, but urges us to accord
controlling significance to this dicta all the same. It is true that
“ ‘[c]arefully considered language of the Supreme Court, even if
technically dictum, generally must be treated as authoritative.’ ”
United States v. Oakar, 111 F.3d 146, 153 (D.C. Cir. 1997)
(quoting Doughty v. Underwriters at Lloyd’s, London, 6 F.3d
856, 861 n.3 (1st Cir. 1993)) (alteration in original). But if
anything more firmly commands the attention of the courts of
appeals than Supreme Court dicta, it is Supreme Court holdings.
By the time Hayes issued in 1985, the Supreme Court had
already established the probable cause requirement for searches
and its two categories of exceptions through a series of carefully
articulated holdings.

The government now claims that dicta have overtaken
holdings. I reject this argument because it reverses the accepted
hierarchy of legal authority.3 If there is to be an identification-

3
See, e.g., Kokkonen v. Guardian Life Ins. Co., 511 U.S. 375, 379
(1994) (“It is to the holdings of our cases, rather than their dicta, that
we must attend . . . .”); St. Mary’s Honor Ctr. v. Hicks, 509 U.S. 502,
515 (1993) (“[W]e think it generally undesirable, where holdings of
the Court are not at issue, to dissect the sentences of the United States
Reports as though they were the United States Code.”); Hawks v.
 9
search exception to the probable cause requirement of the
Fourth Amendment, we must leave it for the Supreme Court to
create. Cf. Rodriguez de Quijas v. Shearson/Am. Express, Inc.,
490 U.S. 477, 484 (1989) (“If a precedent of [the Supreme
Court] has direct application in a case, yet appears to rest on
reasons rejected in some other line of decisions, the Court of
Appeals should follow the case which directly controls, leaving
to [the Supreme Court] the prerogative of overruling its own
decisions.”). The government would have us believe that Hayes
pierced a new hole in the probable cause requirement by way of
an aside about fingerprinting. But just as Congress does not
“hide elephants in mouseholes,” Whitman v. Am. Trucking
Ass’ns, 531 U.S. 457, 468 (2001), the Supreme Court “does not
decide important questions of law by cursory dicta inserted in
unrelated cases,” In re Permian Basin Area Rate Cases, 390
U.S. 747, 775 (1968). The officer safety and special needs
exceptions to the probable cause requirement were announced in
cases that made clear the magnitude of the Fourth Amendment
moment and that altered the course of the law through abundant
progeny. The supposed Hayes identification-search exception,
by contrast, came into this world unannounced and lay idle until
two judges on a three-judge panel of this court employed it over
twenty years later. I think the Supreme Court would be surprised
to learn that Hayes created an exception to the probable cause
requirement, for it “does not normally overturn, or so
dramatically limit, earlier authority sub silentio.” Shalala v. Ill.
Council on Long Term Care, Inc., 529 U.S. 1, 18 (2000).

Judicial creation of a constitutional code of criminal
procedure has given the Supreme Court primary responsibility
for regulating police behavior. See generally Henry J. Friendly,
The Bill of Rights as a Code of Criminal Procedure, 53 CAL. L.

Hamill, 288 U.S. 52, 58–59 (1938) (noting that “an authority [can] be
ranked as a definitive holding or merely a considered dictum”).
 10
REV. 929 (1965); William J. Stuntz, The Political Constitution
of Criminal Justice, 119 HARV. L. REV. 780 (2006). With regard
to evidentiary searches, the Supreme Court has drawn the line of
constitutional “unreasonab[ility]” at probable cause. See Arizona
v. Hicks, 480 U.S. 321, 329 (1987); cf. Dunaway v. New York,
442 U.S. 200, 214 (1979) (noting, in a case concerning seizures,
that “the requisite ‘balancing’ has been performed in centuries
of precedent and is embodied in the principle that seizures are
‘reasonable’ only if supported by probable cause”). Hayes
created no exception to the Supreme Court’s probable cause
requirement for searches, be they for identification or any other
investigatory purposes. As a court of appeals, we are in no
position to create a new exception that would have far-reaching
effects on how the police may properly investigate crime.
Rather, we are bound by Supreme Court precedent, which in this
case requires probable cause to support the unzipping of
Askew’s jacket. Because there was none, I agree that we should
reverse the judgment of the district court denying Askew’s
motion to suppress evidence of the firearm.
 KAVANAUGH, Circuit Judge, with whom Chief Judge
SENTELLE and Circuit Judges HENDERSON and RANDOLPH
join, dissenting:

On a December night in 2003, a D.C. police officer
stopped Paul Askew on a Washington street based on
reasonable suspicion that Askew had just committed an armed
robbery. Consistent with Terry v. Ohio, 392 U.S. 1 (1968),
the police attempted to frisk Askew to ensure he was not
carrying a weapon. After Askew resisted the frisk and the
robbery victim arrived for a show-up procedure, the police
partially unzipped Askew’s outer jacket and discovered he
was carrying a loaded .38 caliber gun. Because Askew was
already a convicted felon, he was prosecuted in U.S. District
Court for being a felon in possession of a firearm. After the
District Court denied Askew’s motion to suppress and ruled
the gun evidence admissible, Askew pled guilty, reserving his
right to appeal the Fourth Amendment issue.

In this Court, Askew accepts that the initial stop was
lawful under the Fourth Amendment. Askew also
acknowledges that the police could frisk him for purposes of
officer safety. But Askew contends that the police exceeded
the scope of a permissible Terry frisk by partially unzipping
his outer jacket, which in turn revealed his gun. He argues
that the gun evidence therefore must be excluded.

We would uphold the search and affirm Askew’s
conviction for either of two alternative reasons. First, after
Askew actively resisted and impeded the police’s initial frisk
attempt, unzipping Askew’s outer jacket to search for a
weapon around his waist area was an objectively reasonable
protective step to ensure officer safety. Second, the police
may reasonably maneuver a suspect’s outer clothing – such as
unzipping a suspect’s outer jacket – when, as here, doing so
could help facilitate a witness’s identification at a show-up
during a Terry stop.
 2
Before explaining why we would affirm Askew’s
conviction, we point out that the legal import of today’s long-
pending and badly splintered en banc decision turns out to be
zero.

On the protective search question, all 11 members of the
en banc Court agree on the settled legal principle that “a
protective search may be lawful when a suspect prevents the
police from performing a Terry frisk.” Maj. Op. at 43. But
Part III(E) of Judge Edwards’s opinion, which on this point is
for a majority, concludes as a factual matter that the police
here did not have an objective basis to unzip Askew’s jacket
as a protective step. Our difference with Part III(E) of the
majority opinion is entirely fact-bound and depends solely on
our different reading of the suppression hearing testimony and
the District Court’s opinion.

On the show-up issue, there is no majority decision either
way on the legal question that we granted en banc review to
decide: whether the police may reasonably unzip a suspect’s
outer jacket to help facilitate a witness’s identification at a
show-up during a Terry stop. Unlike the other nine judges on
the en banc panel, Judges Ginsburg and Garland have not
reached that legal question: In their view, the facts of this
case do not present it. They thus do not join Part III(A)-(C) of
Judge Edwards’s opinion on the show-up issue; but they also
do not join Part II(B) of our opinion. On the show-up issue,
they join only the single, fact-bound paragraph in Part III(D)
of Judge Edwards’s opinion, thereby making Part III(D) the
entirety of the majority’s opinion on that issue. That
paragraph assumes arguendo that the police may “unzip a
suspect’s jacket to facilitate a show-up if, inter alia, ‘there is a
reasonable basis for believing’ that doing so ‘will establish or
negate the suspect’s connection with the crime’ under
investigation.” Maj. Op. at 37 (quoting Hayes v. Florida, 470
 3
U.S. 811, 817 (1985) (alteration omitted)). But Part III(D)
finds on the facts of this case that the police did not have a
sufficient factual predicate to unzip Askew’s jacket.

The fact-based approach of Judges Ginsburg and
Garland, as reflected in their joining only Part III(D) of Judge
Edwards’s opinion on the show-up issue, is the narrowest
ground necessary for reversing the conviction and thus
constitutes the binding expression of the en banc Court on that
issue. Cf. Marks v. United States, 430 U.S. 188, 193 (1977).
As a result, there is no decision of the Court with respect to
the legal question that was decided by the divided three-judge
panel and listed without dissent as the only issue in the order
granting en banc review. That question will remain
unanswered in this Circuit and will have to be decided anew
in future cases by other district court judges and three-judge
panels of this Court, and perhaps ultimately by another en
banc panel. For purposes of future decisions on the show-up
issue, in other words, Part III(A)-(C) of Judge Edwards’s
opinion and Part II(B) of our opinion carry equal precedential
weight, which is to say they carry no precedential weight at
all.

I

At about 11:00 p.m. on December 19, 2003, minutes after
an armed robbery in Washington, D.C., a police dispatcher
broadcast the location of the crime and the victim’s
description of the armed robber, saying among other things
that the perpetrator was wearing a blue sweatshirt.
Metropolitan Police Department Officer Bowman then
stopped Paul Askew on a street near the robbery because
Askew appeared similar to the radio description. After
Officer Bowman stopped Askew, Officers Willis and Koenig
arrived at the scene and decided to frisk Askew “for officers
 4
safety” and “to make sure he wasn’t armed or anything.”
Mar. 26 Tr. at 6. Consistent with Terry v. Ohio, 392 U.S. 1
(1968), Officer Koenig directed Askew to put his hands on
the police car and then tried to frisk him. Mar. 26 Tr. at 6-7.
But according to Officer Willis, as Officer Koenig attempted
to frisk Askew “towards his waist” and then “continued to try
to pat him down,” Askew “kept leaning up against the cruiser
that his hands was on.” Id. at 8. (The two other officers who
testified at the suppression hearing also stated that Askew was
uncooperative at various times during the encounter. Mar. 10
Tr. at 17, 40-42, 47.)

According to Officer Willis, while Officer Koenig was
attempting to conduct the frisk and Askew was resisting, the
robbery victim arrived in another police car for a show-up.
Mar. 26 Tr. at 8. Officer Willis turned Askew around so that
the robbery victim could see him. Id. Officer Willis
remembered that the police dispatcher had said the robber was
wearing a blue sweatshirt. Id. Officer Willis stated that he
“wanted to expose the blue hooded sweatshirt to the victim to
make sure that that’s what she saw.” Id. at 9. He therefore
partially unzipped Askew’s outer jacket so that the victim
could get a better view of Askew’s clothing, specifically the
blue sweatshirt underneath Askew’s outer jacket. Id. at 8-9.
At that point, Officer Willis did not know whether the show-
up witness had identified Askew as the robber. Id. at 18-19.
The unzipping of Askew’s jacket revealed his loaded .38
caliber gun. Id. at 9-11.

Askew was prosecuted for being a felon in possession of
a firearm. Askew moved to suppress the gun evidence,
arguing that the police violated the Fourth Amendment when
they unzipped his jacket during the show-up. The District
Court rejected Askew’s argument, explaining that partially
unzipping Askew’s jacket reasonably facilitated the show-up
 5
by allowing the witness to obtain a better view of Askew’s
clothing. United States v. Askew, 313 F. Supp. 2d 1 (D.D.C.
2004) (Friedman, J.).

II

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. As the Supreme
Court has explained, the text of the Fourth Amendment does
not prohibit all searches and seizures without probable cause
and a warrant. Rather, the “touchstone of the Fourth
Amendment is reasonableness.” United States v. Knights, 534
U.S. 112, 118 (2001).1

1
See also Illinois v. McArthur, 531 U.S. 326, 330 (2001)
(Fourth Amendment’s “central requirement is one of
reasonableness.”) (internal quotation marks omitted); Ohio v.
Robinette, 519 U.S. 33, 39 (1996) (“We have long held that the
touchstone of the Fourth Amendment is reasonableness.”) (internal
quotation marks omitted); Florida v. Jimeno, 500 U.S. 248, 250
(1991) (“The Fourth Amendment does not proscribe all state-
initiated searches and seizures; it merely proscribes those which are
unreasonable.”); Pennsylvania v. Mimms, 434 U.S. 106, 108-09
(1977) (“The touchstone of our analysis under the Fourth
Amendment is always the reasonableness in all the circumstances
of the particular governmental invasion of a citizen’s personal
security.”) (internal quotation marks omitted); Akhil Reed Amar,
Fourth Amendment First Principles, 107 HARV. L. REV. 757, 759
(1994) (“We need to read the Amendment’s words and take them
seriously: they do not require warrants [or] probable cause, . . . but
they do require that all searches and seizures be reasonable.”).
 6
Reasonableness “is measured in objective terms by
examining the totality of the circumstances.” Ohio v.
Robinette, 519 U.S. 33, 39 (1996). The test “balances the
nature and quality of the intrusion on personal security against
the importance of the governmental interests alleged to justify
the intrusion.” United States v. Hensley, 469 U.S. 221, 228
(1985); see also Bell v. Wolfish, 441 U.S. 520, 559 (1979);
Terry v. Ohio, 392 U.S. 1, 20-22 (1968).

The Supreme Court has conducted the Fourth
Amendment reasonableness inquiry in common-law fashion
and has developed rules and standards for different categories
of search-and-seizure situations – including “the quantum of
evidence needed for certain distinct kinds of official action.”
4 WAYNE R. LAFAVE, SEARCH AND SEIZURE § 9.1(e), at 279
(4th ed. 2004). In some categories, the Court has required
individualized probable cause, usually supported by a judicial
warrant. In situations where the Court has determined that the
government need outweighs the individual interest, the Court
has allowed warrantless searches and seizures without
probable cause – for example, with only reasonable suspicion
or in some cases even with no individualized suspicion.2

2
For cases involving warrantless searches without probable
cause, see, e.g., United States v. Knights, 534 U.S. 112, 119-22
(2001) (allowing search of probationer’s home on reasonable
suspicion); Vernonia Sch. Dist. 47J v. Acton, 515 U.S. 646, 653-65
(1995) (upholding school’s policy of suspicionless drug testing of
student athletes); Nat’l Treasury Employees Union v. Von Raab,
489 U.S. 656, 667-77 (1989) (allowing suspicionless urinalysis of
United States Customs employees who applied for promotions);
Skinner v. Ry. Labor Executives’ Ass’n, 489 U.S. 602, 620-34
(1989) (upholding suspicionless urine, blood, and breath testing of
railroad employees); O’Connor v. Ortega, 480 U.S. 709, 722-26
(1987) (allowing search of government employee’s office on
 7
A

We would hold that the gun evidence in this case is
admissible because unzipping Askew’s jacket to search for a
weapon in his waist area was an objectively reasonable
protective step to ensure officer safety after Askew’s
resistance to the initial Terry frisk attempt.

In Terry v. Ohio, the Supreme Court held that the police –
without a warrant and based on reasonable suspicion that an
individual committed, was committing, or was about to
commit a crime – may stop the suspect for further
investigation and conduct a protective frisk for weapons or
other instruments of assault. 392 U.S. at 19-27. The Supreme

reasonable suspicion that search would produce evidence of work-
related misconduct); New Jersey v. T.L.O., 469 U.S. 325, 340-42 &
n.8 (1985) (permitting school authorities to search student upon
reasonable grounds for suspecting violation of school rule); New
York v. Belton, 453 U.S. 454, 459-62 (1981) (allowing searches
incident to lawful arrest without additional justification); cf.
Camara v. Mun. Court, 387 U.S. 523, 534-40 (1967) (allowing
warrantless administrative searches by municipal health and safety
inspectors). For cases involving warrantless seizures without
probable cause, see, e.g., Illinois v. Lidster, 540 U.S. 419, 427-28
(2004) (allowing suspicionless stops of motorists to seek
information regarding recent hit-and-run accident); Michigan v.
Sitz, 496 U.S. 444, 451-55 (1990) (upholding highway sobriety
checkpoint without any individualized suspicion); Delaware v.
Prouse, 440 U.S. 648, 663 (1979) (allowing stop of vehicle to
check license and registration on reasonable suspicion that motorist
is unlicensed or automobile is unregistered); Mimms, 434 U.S. at
108-11 (allowing officers conducting traffic stop to order driver out
of car even if they lack particularized reason for believing driver
possesses a weapon); United States v. Martinez-Fuerte, 428 U.S.
543, 557-62 (1976) (allowing suspicionless stops at border-control
checkpoints).
 8
Court has allowed protective frisks during Terry stops
because police officers are at great risk during those
encounters. See, e.g., Michigan v. Long, 463 U.S. 1032, 1052
(1983). The Court has explained that the Fourth Amendment
does not require officers to choose between investigating
criminal activity and avoiding violent attack. On the contrary,
“it would be unreasonable to require that police officers take
unnecessary risks in the performance of their duties.” Terry,
392 U.S. at 23. A “policeman making a reasonable
investigatory stop should not be denied the opportunity to
protect himself from attack by a hostile suspect.” Adams v.
Williams, 407 U.S. 143, 146 (1972).

Those cases and principles have led courts to establish
what by now is a well-settled corollary to the Terry frisk
doctrine: When a suspect hinders the police from adequately
performing the initial Terry frisk, officers may protect
themselves by following up with reasonable steps to
determine whether the suspect is concealing a weapon. See
Maj. Op. at 43 (“[W]e do not dispute that a protective search
may be lawful when a suspect prevents the police from
performing a Terry frisk . . . .”).3

3
See Inouye v. Kemna, 2007 WL 2669540, at *1-2 (9th Cir.
2007) (officer justified in reaching into suspect’s pockets after
suspect evaded frisk); State v. Heitzmann, 632 N.W.2d 1, 9 (N.D.
2001) (“Courts have recognized that a more intrusive Terry search
may be constitutionally permissible when the detainee attempts to
prevent an officer from performing an effective pat-down.”); 4
WAYNE R. LAFAVE, SEARCH AND SEIZURE § 9.6(b) & nn.190, 194
(4th ed. 2004) (collecting cases holding that further search is
warranted when suspect resists frisk or makes sudden move toward
the place where weapon may be hidden); cf. Adams v. Williams,
407 U.S. 143, 147-48 (1972) (officer justified in directly retrieving
gun from suspect’s waist when suspect did not comply with
 9
In analyzing whether the police here could unzip
Askew’s jacket as a protective step to ensure officer safety,
we are not limited to examining the officer’s subjective intent
(which, according to the testimony, was to facilitate the show-
up). Rather, we consider whether the officer had objectively
reasonable grounds to unzip Askew’s jacket for purposes of
officer safety. See Terry, 392 U.S. at 21-22; cf. Whren v.
United States, 517 U.S. 806, 812-13 (1996); Scott v. United
States, 436 U.S. 128, 136-38 (1978). As then-Judge Roberts
explained, the “propriety of a search under the Fourth
Amendment depends on an objective assessment of the
officer’s actions in light of the facts and circumstances

officer’s request to exit car); United States v. Thompson, 597 F.2d
187, 191 (9th Cir. 1979) (officer justified in reaching into suspect’s
pocket after suspect repeatedly tried to reach into his pocket despite
warnings not to and bulky coat prevented officer from determining
from pat-down whether pocket contained weapon); United States v.
Vaughn, 1994 WL 119002, at *2 (D.C. Cir. 1994) (officer justified
in forcibly removing suspect’s hands from his pockets after suspect
made abrupt movement: “The whole point of a protective search
conducted under the aegis of the Terry stop is to secure the officer’s
safety while investigating suspected criminal activity. Where a
target of a Terry stop makes abrupt movement towards areas of his
body that might harbor a weapon, officers have good reason to be
wary and to take adequate precautionary measures.”); United States
v. Cherry¸ 767 F. Supp. 285, 286 (D.D.C. 1991) (“Generally, Terry
searches consist of patdowns; however, numerous cases have
approved searches without a preceding patdown where the exigent
circumstances of the stop justify immediate action by the police.
Such cases invariably involve threatening gestures or sudden
movements by the person being searched or other conditions which
courts have found to be sufficiently exigent to warrant an increased
intrusion upon the defendant’s person.”); State v. Roach, 796 A.2d
214, 219-20 (N.J. 2002) (officers justified in removing item from
suspect’s pants when suspect refused to obey the officers’ orders
and “continued to move his hands toward the unidentified bulge”).
 10
confronting him at the time and not on the officer’s own
subjective intent in executing the search.” United States v.
Holmes, 385 F.3d 786, 790 (D.C. Cir. 2004) (internal
quotation marks and citation omitted); see also United States
v. Jackson, 415 F.3d 88, 91 (D.C. Cir. 2005) (“[O]fficers’
actual motives for conducting the search are not relevant as
long as their actions were objectively reasonable.”) (internal
quotation marks and alterations omitted); United States v.
McKie, 951 F.2d 399, 402 (D.C. Cir. 1991) (“[W]e are not
limited to what the stopping officer says or to evidence of his
subjective rationale . . . .”). The majority agrees that the
officers’ subjective intent is not relevant. Maj. Op. at 40
(“[W]e agree that the standard for determining reasonableness
in the Fourth Amendment context is an objective one.”).

Applying those settled principles to this case is
straightforward. Askew concedes that, based on the victim’s
description, the police had reasonable suspicion to believe he
just committed an armed robbery. The police officers also
knew at the moment of unzipping that Askew had resisted the
initial frisk attempt and tried to prevent them from feeling his
waist area. Because of Askew’s resistance and evasive
movements, the officers had an objectively reasonable basis
to protect themselves by unzipping Askew’s jacket to
determine whether Askew was concealing a weapon at his
waist area – which, in fact, he was.

It bears emphasis that the officers did not skip the Terry
frisk and immediately unzip Askew’s jacket. Cf. Adams, 407
U.S. at 147-48; United States v. Casado, 303 F.3d 440, 447-
48 (2d Cir. 2002); United States v. Vaughn, 1994 WL 119002,
at *1-2 (D.C. Cir. 1994). On the contrary, the officers started
with the Terry frisk. After a suspect actively attempts to
impede a frisk, as Askew did here, it is entirely reasonable
and proportional for officers to conduct a targeted and limited
 11
search for weapons. Cf. Adams, 407 U.S. at 147-48. Such a
two-step approach represents textbook compliance with the
requirements of Terry. See id. On the facts of this case,
therefore, unzipping Askew’s jacket to search for a weapon in
his waist area was an objectively reasonable protective step to
ensure officer safety during the Terry stop.

Part III(E) of Judge Edwards’s opinion, which on this
point is for a majority, reads the record of this case
differently. According to the majority, the District Court
found that the officers completed the frisk after Askew’s
interference, and that Askew’s interference therefore could
not justify any follow-up protective search. But the problem
for the majority opinion is that the District Court made no
such finding. On the contrary, the District Court expressly
recounted Officer Willis’s testimony that, at the time of the
unzipping, “Officer Koenig had not completed the pat-down,
perhaps because of some resistance by the defendant.” United
States v. Askew, 313 F. Supp. 2d 1, 3 n.2 (D.D.C. 2004). The
District Court gave no indication that it did not credit Officer
Willis’s testimony. And we see no reason why the District
Court would have explicitly recited this testimony while
somehow intending to implicitly discredit it; that
interpretation of the District Court’s opinion frankly does not
make sense. The fact that Askew impeded Officer Koenig’s
attempts to pat him down is otherwise uncontroverted in the
record. Officer Willis’s testimony thus establishes that, at the
time of the unzipping, the officers’ initial frisk was not
complete, and the officers’ concern about Askew’s
interference with the initial frisk attempt had not dissipated.

The majority nonetheless refers to what it calls the “total
implausibility” of a scenario whereby the police would have
walked “toward the complainant with an unhandcuffed
robbery suspect who has successfully prevented them from
 12
completing a frisk of his person for weapons.” Maj. Op. at
42, 41. But the majority’s replay-booth-like review does not
acknowledge how quickly the events transpired; the
unzipping of the jacket occurred within seconds of Askew’s
resistance and simultaneously with the show-up – all while
the victim remained in the police cruiser. The majority has
had months to unpack and second-guess those split-second
police decisions. The officers did not have that luxury.
Events do not unfold in super slow motion in the real world in
which police officers operate. Moreover, with respect, it’s the
majority’s version of events that is implausible: The majority
claims that the police completed a full frisk of Askew after
Askew initially resisted, yet still somehow failed to discover
the loaded .38 caliber gun at his waist. The majority’s
conclusion thus necessarily rests on an assumption that the
officers here were dangerously incompetent. We see no basis,
however, for such an assumption.

In an effort to bolster their strained reading of the factual
record, the majority tries to suggest that the Government has
not pressed the protective search argument, as if the
Government somehow conceded the point. That is inaccurate.
The Government argued before the District Court that the
unzipping was a reasonable protective step to ensure officer
safety; indeed, that was the Government’s primary contention.
See Additional Case Law in Support of Government’s
Opposition to Defendant’s Motion to Suppress Tangible
Evidence at 2-5, United States v. Askew, 313 F. Supp. 2d 1
(D.D.C. 2004) (No. 1:04-cr-10). After the District Court
expressed its preliminary belief that the unzipping of Askew’s
jacket was better justified on a show-up rationale, the
Government instead focused on that alternative argument,
which ultimately was the sole basis for the District Court’s
ruling. The Government never conceded the search was not
justified as a protective search. On the contrary, the
 13
Government continues to support this officer safety rationale:
“Given that appellant had successfully resisted an officer’s
‘continued’ attempts to conduct a more traditional pat-down,
it was objectively reasonable for the police to unzip
appellant’s jacket to determine whether he had a weapon in
his waist area.” Gov’t Supp. En Banc Br. at 11.

All of that said, our fact-based disagreement with the
majority on the protective search issue simply underscores
that, at least as to that question, this is a routine, fact-bound
Fourth Amendment case where different Judges happen to
interpret the evidence and the District Court’s opinion
differently.4 In other words, the majority’s protective search
analysis works no change in and does no harm to Circuit
precedent on the permissibility of protective searches
following a suspect’s interference with a Terry frisk. See
Maj. Op. at 43 (“[W]e do not dispute that a protective search
may be lawful when a suspect prevents the police from
performing a Terry frisk . . . .”).

B

As an alternative basis for upholding the search and
affirming the conviction, we would conclude that unzipping
Askew’s jacket was a reasonable investigative step to
facilitate the show-up procedure. In our judgment, the police
may reasonably maneuver a suspect’s outer clothing, such as
removing a suspect’s hat or sunglasses or unzipping a
suspect’s outer jacket, when doing so could help facilitate a
witness’s identification at a show-up during a Terry stop.

4
It bears mention that this fact-based disagreement between
the majority opinion and our opinion likely would have been
avoided had the officer who conducted the initial frisk (Officer
Koenig) been called to testify at the suppression hearing.
 14
In Terry v. Ohio and Sibron v. New York, the Supreme
Court held that the police may forcibly stop a suspect without
probable cause if they have reasonable suspicion of a crime.
See Terry, 392 U.S. at 21-23; Sibron, 392 U.S. 40, 63-64
(1968). In explaining the permissible contours of a Terry
stop, the Court stated that the officer’s actions during the stop
must be “reasonably related in scope to the circumstances
which justified the interference in the first place.” Terry, 392
U.S. at 20. Applying that standard, the Court then considered
whether police may search a suspect during a Terry stop by
conducting (i) a protective frisk for weapons or instruments of
assault that could be used to harm officers or (ii) an
“exploratory search” for contraband or evidence of crime
possessed by the suspect. See id. at 24-31. The Court
permitted the police to perform protective frisks for weapons
or other instruments of assault, finding that the government
interest in officer safety outweighed the individual privacy
interest. See id.; see also Adams, 407 U.S. at 146; cf. Sibron,
392 U.S. at 65. But the Court prohibited, in the absence of
probable cause, what the Court has variously called
“exploratory,” “evidentiary,” or “full” searches for contraband
or evidence of crime that the suspect might be carrying.5

5
See Terry, 392 U.S. at 30 (prohibiting “general exploratory
search” aimed at recovering “whatever evidence of criminal activity
[officer] might find”); Sibron, 392 U.S. at 64-66 (distinguishing
protective search for weapons from search for narcotics); see also
Minnesota v. Dickerson, 508 U.S. 366, 378 (1993) (prohibiting
“evidentiary search,” namely one raising prospect that officer will
“rummage and seize at will” beyond “specific authorization” of
Terry stop) (internal quotation marks omitted); Florida v. Royer,
460 U.S. 491, 499 (1983) (plurality opinion of White, J.) (“police
may not carry out a full search of the person or of his automobile or
other effects” during Terry stop); cf. Maryland v. Buie, 494 U.S.
325, 335 (1990) (distinguishing “cursory inspection” from “full
 15
A search that consists of unzipping a suspect’s outer
jacket to help facilitate a show-up cannot be pigeonholed into
either of the two categories of police procedures addressed in
Terry. Rather, such a search falls into a third category:
“identification procedures.” Identification procedures seek to
match a suspect to a crime or crime scene and include, for
example, fingerprints, palm prints, footprints, body
measurements, saliva samples, hair samples, fingernail
samples, fingernail scrapings, lineups and show-ups,
photographs, voice samples, and handwriting samples. See
MODEL CODE OF PRE-ARRAIGNMENT PROCEDURE § 170.1
(1975). Many such “identification procedures” constitute
“searches” for purposes of the Fourth Amendment. Because
identification procedures do not seek to uncover weapons,
they are not the protective frisks that Terry permits. Cf.
Terry, 392 U.S. at 24-31. But because identification
procedures do not seek to uncover contraband or evidence of
crime possessed by the suspect, they also are not the
exploratory searches that Terry prohibits. Cf. Minnesota v.
Dickerson, 508 U.S. 366, 378 (1993); Sibron, 392 U.S. at 64-
66.

Applying the bedrock Terry principle that the police, with
reasonable suspicion, may “stop the person for a brief time
and take additional steps to investigate further,” the Supreme

search”); Skinner, 489 U.S. at 638 (Marshall, J., dissenting)
(describing past cases: “Only where the government action in
question had a substantially less intrusive impact on privacy, and
thus clearly fell short of a full-scale search, did we relax the
probable-cause standard.”) (internal quotation marks and citation
omitted); New York v. Class, 475 U.S. 106, 118-19 (1986)
(distinguishing permissible focused search of car’s dashboard from
full search of passenger compartment); United States v. Jacobsen,
466 U.S. 109, 125 n.28 (1984) (distinguishing fingernail scrapings
from “full search”).
 16
Court has permitted a variety of identification procedures
during Terry stops. Hiibel v. Sixth Judicial Dist. Court, 542
U.S. 177, 185 (2004). The Court has held that police may
conduct show-ups, compel suspects to provide their
identification, search a suspect’s car to see the vehicle
identification number, and take fingerprints. See Michigan v.
Summers, 452 U.S. 692, 701 n.12 (1981) (show-ups); Hiibel,
542 U.S. at 185-89 (identification); Hensley, 469 U.S. at 229,
234 (identification); New York v. Class, 475 U.S. 106, 119
(1986) (vehicle identification number); Hayes v. Florida, 470
U.S. 811, 817 (1985) (fingerprinting); cf. Davis v. Mississippi,
394 U.S. 721, 727-28 (1969) (fingerprinting). The Supreme
Court has authorized those various identification procedures
during Terry stops – that is, without a warrant or probable
cause – so long as the procedures were reasonable under the
circumstances. In no case has the Supreme Court prohibited
an identification procedure during a Terry stop.

Askew argues, however, that only those identification
procedures that do not constitute Fourth Amendment
“searches” are permissible during Terry stops. We disagree.

1

In Hayes v. Florida, the Supreme Court authorized
fingerprinting during Terry stops “if there is reasonable
suspicion that the suspect has committed a criminal act, if
there is a reasonable basis for believing that fingerprinting
will establish or negate the suspect’s connection with that
crime, and if the procedure is carried out with dispatch.” 470
U.S. at 817. In reaching that conclusion, the Court recognized
that fingerprinting is a Fourth Amendment search, albeit a
“much less serious intrusion upon personal security than other
types of searches and detentions.” Id. at 814 (emphasis
added); see generally Katz v. United States, 389 U.S. 347
 17
(1967). Hayes thus stands for the proposition that an
identification procedure constituting a search is permissible
during a Terry stop if the procedure is reasonable under the
circumstances.

Askew dismisses the import of Hayes by suggesting that
fingerprinting is not a search. Askew’s position ignores the
Hayes language comparing fingerprinting to “other types of
searches and detentions.” Hayes, 470 U.S. at 814 (emphasis
added). Moreover, if Askew were correct, the Supreme Court
could not have limited fingerprinting during Terry stops to
cases where the police have a “reasonable basis for believing
that fingerprinting will establish or negate the suspect’s
connection with that crime.” Id. at 817. In other words, if
fingerprinting were not a search, the police would not have to
make any showing to take fingerprints during any – or indeed
every – Terry stop. That is obviously not what Hayes held; in
fact, Hayes said just the opposite.6

6
In a 1973 case, the Supreme Court hinted in dicta that
fingerprinting may not be a search. See United States v. Dionisio,
410 U.S. 1, 15 (1973). The leading Fourth Amendment scholar
explains that this “dictum in Dionisio might be considered suspect.”
1 LAFAVE, SEARCH AND SEIZURE § 2.6(a), at 668. The Court’s
later decision in Hayes plainly considered fingerprinting a search;
the Hayes language and requirement for some evidentiary showing
by the police to obtain fingerprints during a Terry stop cannot be
explained otherwise. Therefore, although some of the Supreme
Court’s pre-Hayes language “indicates that a person does not have a
reasonable expectation of privacy in his or her fingerprints, which
is to say that the police actions of taking a fingerprint would not be
a search within the meaning of the Fourth Amendment, in Hayes,
the Court signaled that a person has a protected interest, albeit a
diminished one, in the taking of his or her fingerprints.” Thomas
K. Clancy, What Is a “Search” Within the Meaning of the Fourth
Amendment?, 70 ALB. L. REV. 1, 8 n.39 (2006). Hayes reflects and
 18
In their separate opinion in Hayes, Justices Brennan and
Marshall similarly interpreted the Court’s majority opinion to
allow a search (fingerprinting) without probable cause during
a Terry stop. They strenuously objected to this result,
however, because they argued – as Askew does here – that the
only search permitted during a Terry stop is a protective frisk
for weapons. They complained that on-site fingerprinting
involves “a singular intrusion on the suspect’s privacy” that is
not “justifiable (as was the patdown in Terry) as necessary for
the officer’s protection.” Id. at 819 (Brennan, J., concurring
in judgment). Indeed, they labeled the majority’s conclusion
a “regrettable assault on the Fourth Amendment.” Id. Of
course, Justices Brennan and Marshall could call the Hayes
majority opinion a “regrettable assault on the Fourth

is consistent with the modern trend in Supreme Court cases, which
increasingly has been to recognize intrusive and invasive police
practices as searches – and then to determine whether they are
reasonable. See Kyllo v. United States, 533 U.S. 27, 34 (2001)
(“We think that obtaining by sense-enhancing technology any
information regarding the interior of the home that could not
otherwise have been obtained without physical intrusion into a
constitutionally protected area constitutes a search – at least where
(as here) the technology in question is not in general public use.”)
(internal quotation marks and citation omitted); Bond v. United
States, 529 U.S. 334, 337 (2000) (merely feeling or squeezing a
passenger’s bag is a search: “Physically invasive inspection is
simply more intrusive than purely visual inspection.”); Skinner, 489
U.S. at 616-17 (breath, blood, and urine collection and testing
procedures are searches). The Supreme Court ordinarily no longer
engages in the kinds of contortions it once arguably employed to
find certain police procedures not to be searches. See Akhil Reed
Amar, Fourth Amendment First Principles, 107 HARV. L. REV.
757, 769, 783-85 (1994) (“To avoid some of the absurdities created
by the so-called warrant and probable cause requirements, the
Justices have watered down the plain meaning of ‘search’ and
‘seizure.’”).
 19
Amendment” only because the decision permitted a search
(fingerprinting) on less than probable cause.

Askew also tries to discount the Supreme Court’s
reasoning in Hayes as dicta. But the Supreme Court has
treated this language from Hayes as authoritative. See Hiibel,
542 U.S. at 188-89. Moreover, the Court’s analysis in Hayes
was carefully considered; indeed, it was the point of strenuous
disagreement between the Hayes majority and Justices
Brennan and Marshall in their separate opinion. We have said
that “carefully considered language of the Supreme Court,
even if technically dictum, generally must be treated as
authoritative.” United States v. Dorcely, 454 F.3d 366, 375
(D.C. Cir. 2006) (internal quotation marks omitted); see also
Sierra Club v. EPA, 322 F.3d 718, 724 (D.C. Cir. 2003);
Natural Res. Def. Council, Inc. v. NRC, 216 F.3d 1180, 1189
(D.C. Cir. 2000); United States v. Oakar, 111 F.3d 146, 153
(D.C. Cir. 1997).

Hayes establishes that the identification procedure used
in Askew’s case was reasonable under the Fourth
Amendment. Here as in Hayes, the identification procedure
constituted a search. Here as in Hayes, the purpose of the
identification procedure was to match a suspect with a serious
crime that had just occurred and thereby protect the public
from a violent criminal at large (a rapist in Hayes; an armed
robber here). Here as in Hayes, the police had an objectively
reasonable basis for believing that the identification procedure
could help establish or negate the suspect’s connection with a
crime. Here as in Hayes, the intrusion on the suspect was
relatively minimal; indeed, many people would think having
their fingerprints taken on the street in plain view of the
public is far more intrusive than having their outer jacket
partially unzipped. Here as in Hayes, the police action was
carried out with dispatch.
 20
Contrary to Askew’s suggestion, moreover, Hayes is no
outlier. In New York v. Class, 475 U.S. 106, 118-19 (1986),
the Supreme Court allowed the police during a Terry stop to
conduct a limited search of a car so that they could see the
car’s vehicle identification number (VIN) – even though the
search was not a protective search for weapons as permitted in
Terry or Michigan v. Long, 463 U.S. at 1049. In her opinion
for the Court, Justice O’Connor succinctly described the issue
and holding:

In this case, we must decide whether, in order to
observe a Vehicle Identification Number (VIN) generally
visible from outside an automobile, a police officer may
reach into the passenger compartment of a vehicle to
move papers obscuring the VIN after its driver has been
stopped for a traffic violation and has exited the car. We
hold that, in these circumstances, the police officer’s
action does not violate the Fourth Amendment.

Class, 475 U.S. at 107; see also id. at 132 (White, J.,
dissenting) (majority opinion “in effect holds that a search of
a car for the VIN is permissible whenever there is a legal
stop”).

The reasoning of Class is murky in some places. But the
bottom line of Class is crystal clear: During a Terry car stop,
the police may conduct a limited search of the vehicle so as to
see the VIN, even though they are not searching for a
weapon. The decision in Class, like the opinion in Hayes,
refutes Askew’s position that the police may not conduct a
reasonable search for identification purposes during a Terry
stop.

Also instructive is the Supreme Court’s decision in Hiibel
v. Sixth Judicial District Court, 542 U.S. 177 (2004). Unlike
Hayes and Class, Hiibel did not directly involve a search, but
 21
it nevertheless underscored the importance of identification
procedures during Terry stops. The Court held that police
during Terry stops may compel suspects to provide their
names and, pursuant to state law, may arrest suspects who
refuse to do so, in which case the police necessarily may
conduct a full search incident to arrest that would uncover the
suspect’s license or other form of identification. Id. at 185-
88; cf. State v. Flynn, 285 N.W.2d 710, 719 (Wis. 1979)
(allowing search for wallet with identification during Terry
stop). The Hiibel Court explained that obtaining a person’s
identification during a Terry stop is a “commonsense inquiry”
that “serves important government interests”: It may “inform
an officer that a suspect is wanted” or may “help clear a
suspect and allow the police to concentrate their efforts
elsewhere.” 542 U.S. at 189, 186. The Court emphasized that
the request for identification “does not alter the nature of the
stop itself: it does not change its duration or its location.” Id.
at 188 (citations omitted). In response to concerns about
police harassment, the Court noted that the police must have a
reasonable basis for stopping the suspect and, indeed, cited
Hayes for the proposition that “Terry may permit an officer to
determine a suspect’s identity by compelling the suspect to
submit to fingerprinting only if there is ‘a reasonable basis for
believing that fingerprinting will establish or negate the
suspect’s connection with that crime.”’ Id. (quoting Hayes,
470 U.S. at 817). Hiibel thus reinforces what Hayes and
Class previously stated: Identification is a critical and
legitimate component of Terry stops.

Consistent with the Supreme Court’s precedents, a
number of state legislatures and courts have long allowed law
enforcement officers – with less than probable cause – to
conduct identification procedures that constitute Fourth
Amendment searches. See ARIZ. REV. STAT. ANN. § 13-3905;
COLO. R. CRIM. P. 41.1; IDAHO CODE ANN. § 19-625; IOWA
 22
CODE ANN. §§ 810.1 to .8; N.J. CT. R. 3:5A; N.C. GEN. STAT.
ANN. §§ 15A-271 to -274; VT. R. CRIM. P. 41.1; State v.
Rodriguez, 921 P.2d 643, 650-51 (Ariz. 1996) (en banc);
People v. Madson, 638 P.2d 18, 31-33 (Colo. 1981) (en banc);
Wise v. Murphy, 275 A.2d 205, 213-16 (D.C. 1971); Bousman
v. Iowa Dist. Court, 630 N.W.2d 789, 796-99 (Iowa 2001);
State v. Hall, 461 A.2d 1155, 1159-61 (N.J. 1983); In re
Fingerprinting of M.B., 309 A.2d 3, 6-7 (N.J. Super. Ct. App.
Div. 1973); In re Order Requiring Fingerprinting of a
Juvenile, 537 N.E.2d 1286, 1288-89 (Ohio 1989); In re
Nontestimonial Identification Order Directed to R.H., 762
A.2d 1239, 1243-47 (Vt. 2000). Most of those state rules and
decisions broadly authorize the police to obtain a court order
– based only on reasonable suspicion – to seize a suspect,
compel the suspect’s appearance at the police station or other
location, and forcibly require the suspect to submit to an
intrusive identification procedure. The Model Code of Pre-
Arraignment Procedure similarly permits police based only on
reasonable suspicion to obtain a court order, seize a suspect,
and compel the suspect to submit to an identification
procedure. MODEL CODE OF PRE-ARRAIGNMENT PROCEDURE
§ 170.2. Professor LaFave generally approves of these rules
and decisions because certain “identification searches” do not
“require the rummaging through a suspect’s personal effects
as does an ordinary full-blown search.” 4 LAFAVE, SEARCH
AND SEIZURE § 9.8(b), at 730 (internal quotation marks
omitted). For that reason, Professor LaFave says: “Taking
fingernail scrapings, for example, is a search, but yet is a very
limited intrusion, and thus should be deemed permissible”
even without probable cause. Id. (internal quotation marks
omitted and emphasis added).7

7
The Supreme Court has often looked to state laws, the
Model Code of Pre-Arraignment Procedure, and Professor
 23
In sum, the Supreme Court’s precedents, the various state
laws and decisions, the Model Code of Pre-Arraignment
Procedure, and the views of the leading Fourth Amendment
scholar all demonstrate that Askew’s extreme interpretation of
the Fourth Amendment is seriously flawed. Identification
procedures constituting searches are permitted during Terry
stops so long as the procedures are reasonable under the
circumstances. Police may therefore reasonably maneuver a
suspect’s outer clothing – such as removing a suspect’s hat or
sunglasses or unzipping a suspect’s outer jacket – when doing
so could help facilitate a witness’s identification at a show-up
during a Terry stop. For that reason, it was reasonable for the
police to unzip Askew’s outer jacket so that the robbery
victim could see his clothing and get a better view of his blue
sweatshirt.

2

Askew argues that post-Terry decisions such as Hayes,
Class, and Hiibel, as well as the various state statutes, rules,
and opinions, do not resolve this case. For the reasons stated
above, Askew’s argument is wrong. But even absent all of
those precedents and laws – in other words, based solely on
Terry and general Fourth Amendment principles – Askew errs
in arguing that the police are prohibited during Terry stops
from conducting identification procedures constituting
searches.

LaFave’s treatise in considering the reasonableness of police
practices under the Fourth Amendment. See, e.g., Brendlin v.
California, 127 S. Ct. 2400, 2407-08 (2007) (state court decisions
and LaFave); Illinois v. Lidster, 540 U.S. 419, 425 (2004) (Model
Code of Pre-Arraignment Procedure); Atwater v. City of Lago
Vista, 532 U.S. 318, 344-45, 355-60 (2001) (state laws, Model
Code of Pre-Arraignment Procedure, and LaFave).
 24
In “evaluating the validity of an officer’s investigative or
protective conduct under Terry, the touchstone of our analysis
is always the reasonableness in all the circumstances of the
particular governmental invasion of a citizen’s personal
security.” Long, 463 U.S. at 1051 (internal quotation marks
and alterations omitted). The police action during a Terry
stop must be “reasonably related in scope to the
circumstances which justified the interference in the first
place.” Terry, 392 U.S. at 20; see also Hensley, 469 U.S. at
228 (reasonableness test “balances the nature and quality of
the intrusion on personal security against the importance of
the governmental interests alleged to justify the intrusion”);
Bell, 441 U.S. at 559. Applying that standard, Terry says it is
reasonable to frisk the suspect for weapons and unreasonable
to search the suspect for contraband or evidence. The precise
question in this case – again, assuming cases like Hayes,
Class, and Hiibel do not already resolve it – is whether it is
reasonable for the police to conduct identification procedures.

On the government interest side of the Fourth
Amendment balance, when the police seek to conduct an
identification procedure at a Terry stop, they know that a
serious crime recently has been committed – usually a violent
crime such as murder, rape, robbery, or assault – and that the
perpetrator is on the loose. In such cases, speed is often of the
essence, and tools to quickly establish or negate a suspect’s
connection with the crime are essential so that the police can
determine whether to arrest the suspect or move on to
someone else. Identification procedures at Terry stops
therefore help to protect the public from violent criminals at
large – an interest Judge Edwards’s opinion never mentions.

On the other side of the Fourth Amendment balance, the
primary intrusion on a suspect’s individual privacy during a
Terry stop results from the forcible stop itself and the initial
 25
protective frisk. The additional step of conducting a targeted
identification procedure – such as unzipping an outer jacket
during a show-up – is certainly an interference with individual
privacy. But it is a relatively minimal interference, especially
as compared to numerous other searches that the Supreme
Court has authorized without probable cause. Cf. Knights,
534 U.S. at 119-22 (search of home); Vernonia Sch. Dist. 47J
v. Acton, 515 U.S. 646, 653-65 (1995) (urinalysis); Nat’l
Treasury Employees Union v. Von Raab, 489 U.S. 656, 667-
77 (1989) (urinalysis); Skinner v. Ry. Labor Executives’
Ass’n, 489 U.S. 602, 620-34 (1989) (blood, breath, and urine
testing); O’Connor v. Ortega, 480 U.S. 709, 722-26 (1987)
(search of employee’s office); Class, 475 U.S. at 114-119
(search of part of interior of defendant’s car); New Jersey v.
T.L.O., 469 U.S. 325, 340-42 & n.8 (1985) (search of
student); Pennsylvania v. Mimms, 434 U.S. 106, 111 (1977)
(driver asked to get out of car during Terry stop “is being
asked to expose to view very little more of his person than is
already exposed”).

When balancing the competing interests, the
consequences of accepting Askew’s argument are important
to consider. Prohibiting the police during Terry stops from
conducting identification procedures that constitute searches
would lead to absurd and dangerous results. It would mean
that the police could not remove a suspect’s gloves to perform
the fingerprinting that Hayes expressly allows. It would mean
that the police could not remove a robbery suspect’s hat
during a show-up after an armed robbery. It would mean that
the police could not take off a suspected bank robber’s ski
mask. It would mean that the police could not remove a
murder suspect’s shoe to take a shoe imprint during a Terry
stop, even though a shoeprint was the key evidence at a
murder scene. Cf. State v. Moffatt, 450 N.W.2d 116, 120
(Minn. 1990). It would mean that the police could not lift a
 26
rape suspect’s sleeve to view a tattoo on the suspect’s
forearm, even though the rape victim said the perpetrator had
a distinctive tattoo on his forearm. It would mean that the
police could not take a hair sample from a rape suspect during
a Terry stop, even though the key piece of evidence at the
rape scene was a hair from the perpetrator. Cf. United States
v. Ingram, 797 F. Supp. 705, 716-18 (E.D. Ark. 1992). It
would mean that the police during a Terry stop could not take
fingernail scrapings or a saliva swab from a murder suspect in
a case where the victim was killed in a violent struggle. Cf. In
re Shabazz, 200 F. Supp. 2d 578, 584-85 (D.S.C. 2002). It
would mean that a show-up is all but useless in certain cases,
notwithstanding that the Supreme Court has long blessed
show-ups during Terry stops. It would mean that a large
number of state statutes, rules, and decisions permitting
identification procedures on less than probable cause – which
have been on the books for decades – are all unconstitutional
and wrongly decided. In short, Askew’s position would
hamstring the police and prevent them from performing
reasonable identification procedures that could solve serious
crimes and protect the community from violent criminals at
large.8

For those reasons, even if the Hayes-Class-Hiibel line of
cases did not exist, the government need for identification

8
Askew argues that his position – namely, that only those
identification procedures not constituting searches under the Fourth
Amendment are permitted during a Terry stop – provides a clear,
bright-line rule. But that claim proved hollow at oral argument
when Askew’s skilled counsel could not answer, perhaps
understandably given the state of the case law, numerous questions
about whether certain identification procedures constitute searches
under the Fourth Amendment. See Tr. of En Banc Oral Arg. at 3-
12.
 27
procedures during Terry stops outweighs the intrusion on
individual privacy.9

We would hold that the police may conduct reasonable
identification procedures at Terry stops; in particular, the
police may reasonably maneuver a suspect’s outer clothing –
such as removing a suspect’s hat or sunglasses or unzipping a
suspect’s outer jacket – when, as here, doing so could help
facilitate a witness’s identification at a show-up during a
Terry stop. But we recognize, of course, that there is not a
majority of the Court on the show-up issue for either our
position or Judge Edwards’s position as reflected in Part
III(A)-(C) of his opinion. As a result, the legal question about
the permissible scope of police activity at show-ups that we
granted en banc review to decide remains undecided for now
in this Circuit.

C

We close with a few observations about Part III(D) of
Judge Edwards’s opinion, which consists of just a single, fact-
bound paragraph that constitutes the entirety of the binding
opinion of the Court on the show-up issue.10

9
A variety of conceivable searches would be more intrusive
than the search at issue here. In such hypothetical cases, the
Government would face a heavier burden to show that the
investigative step was reasonable. See Schmerber v. California,
384 U.S. 757, 769-72 (1966); Helton v. United States, 191 F. Supp.
2d 179, 184-85 (D.D.C. 2002); see also 4 LAFAVE, SEARCH AND
SEIZURE § 9.8(b), at 730 (“there may be a serious question as to
whether the taking of a blood sample should be allowed” on less
than probable cause).
10
To reiterate: On the show-up issue, Judges Ginsburg and
Garland join only the paragraph in Part III(D) of Judge Edwards’s
 28
That paragraph assumes arguendo that the police may
“unzip a suspect’s jacket to facilitate a show-up if, inter alia,
‘there is a reasonable basis for believing’ that doing so ‘will
establish or negate the suspect’s connection with the crime’
under investigation.” Maj. Op. at 37 (quoting Hayes, 470
U.S. at 817 (alteration omitted)). But the paragraph then
decides that the police in this case did not have a sufficient
factual predicate to meet this standard. We disagree. It is
generally reasonable for police to believe that unzipping a
suspect’s outer jacket could help facilitate a robbery witness’s
identification at a show-up. Even if a witness can readily see
a suspect’s face, allowing the witness to view the suspect’s
clothing would enhance the accuracy of the identification.
That is particularly true where, as here, the witness had
identified a specific item of the perpetrator’s clothing when
first reporting the crime, and the suspect appeared to be
wearing that clothing under his outer jacket.

Because the robbery victim here specified a blue
sweatshirt when she first reported the crime, the officers
reasonably believed that the sweatshirt was a significant
identifying factor for the victim. Officer Willis testified: “I
remember they said the defendant had on a blue hooded
sweatshirt, but [Askew] had his jacket zipped up. I wanted
the complainant to see what he had on to make sure that he
wasn’t zipping nothing up to cover up. . . . [I] wanted to
expose the blue hooded sweatshirt to the victim to make sure
that that’s what she saw.” Mar. 26 Tr. at 8-9.

The police’s belief was reasonable, moreover, even
though the victim, when she first reported the crime,
apparently did not indicate whether the perpetrator’s
sweatshirt had distinctive markings. The majority suggests

opinion; they do not reach the legal question addressed in Part
III(A)-(C) of Judge Edwards’s opinion.
 29
that there was insufficient justification for allowing the victim
to see Askew’s sweatshirt because it would have been
impossible for her to distinguish a “generic blue sweatshirt”
from others. Maj. Op. at 37. But just because the victim did
not mention a distinctive characteristic of the sweatshirt in her
initial description doesn’t mean allowing her to see Askew’s
sweatshirt would have been futile for identification purposes.
The majority ignores the obvious point that there are
numerous styles of “blue sweatshirts,” and any shade,
material, logo, pattern, or feature could have helped the victim
match or distinguish Askew’s sweatshirt from the one she
recalled being worn by the robber. As the Government
correctly notes, a “blue sweatshirt” could be “cotton or fleece,
zippered or buttoned; and it could have pockets or a hood.”
Gov’t Supp. En Banc Br. at 4 n.1. A “blue sweatshirt” could
feature, among countless other things, an athletic company
logo, a college mascot, a professional sports team name, or a
corporate slogan. To be sure, a “blue sweatshirt” could also
have none of those things, presumably rendering it the kind of
plain old “generic blue sweatshirt” that the majority describes.
But what is “generic blue”? One major manufacturer of
sweatshirts lists 14 wildly variant shades of this “generic”
color: Aqua, Baltic Blue, Blue, Cadet Blue, California Blue,
Cornflower Blue, Granite Blue, Light Blue, Medium
Stonewash, North Carolina Blue, Periblue, True Navy, True
Royal, and Vintage Navy. See Jerzees Color Translation, at
http:// www.jerzees.com/ jerzeesactivewear/ ColorTranslation
.aspx. If the robber’s “generic blue” sweatshirt had been
“North Carolina Blue” but Askew’s was “True Navy,” that
would have been the end of the matter and Askew would have
walked. Or the victim alternatively might have said, “that’s
the sweatshirt and that’s the robber.”

The irony here, moreover, is that seeing the sweatshirt
may actually have helped the victim rule Askew out as the
 30
robber. Keep in mind that the end result of the show-up was
that the robbery victim said Askew was not the right guy.
The only reason unzipping the outer jacket to see the
sweatshirt proved problematic for Askew is the inconvenient
fact that he was illegally carrying a loaded .38 caliber gun
under his jacket.

The larger point is that Hayes requires only a “reasonable
basis for believing” that the identification procedure will
establish or negate the suspect’s connection with the crime.
Hayes, 470 U.S. at 817. Given that the robbery victim
initially told the police about the blue sweatshirt, how can we
say that there is not even a reasonable basis for believing that
the victim’s seeing the blue sweatshirt would have assisted
her identification? The majority offers no good answer to that
commonsense question.11

In our judgment, requiring this kind of pinpoint accuracy
and detail in a victim’s initial description of a criminal is
inconsistent with the Hayes “reasonable basis” standard and
could place a dangerous burden on law enforcement officials.
After all, victims often do not provide such richly detailed
descriptions when first reporting violent crimes to the police.

A good indication that Part III(D) rests on quicksand is
the fact that Askew himself did not raise this argument either
in his submission to the District Court or on appeal, where he
was represented by skilled and experienced counsel from the

11
No doubt because there is no good answer to that question,
Judge Griffith, notwithstanding his separate opinion’s narrower
view of what police are permitted to do under the Fourth
Amendment, does not join Part III(D) of Judge Edwards’s opinion;
he agrees with us that, if the Hayes standard applies, the facts here
meet the Hayes test. Concurring Op. of Judge Griffith at 5 n.2.
 31
Federal Public Defender’s Office. Defense counsel
presumably considered the argument untenable.12

In any event, our disagreement with Part III(D) of Judge
Edwards’s opinion (which on this point is for a majority) is a
narrow fact-bound dispute with respect to the record in
Askew’s specific case. In many future show-up cases, the
record indisputably will satisfy the Hayes reasonable-basis
requirement as described by Part III(D) of that opinion. In
such a case, of course, the Court would need to decide the
legal issue regarding show-ups that the Court today leaves
open.

One final point: The majority’s ruling in Part III(D), as
we read it, is limited to a witness’s viewing clothing at a
show-up during a Terry stop. The majority does not purport
to address or alter long-standing law about what is sufficient
to constitute “reasonable suspicion” to stop someone in the
first place. If the majority’s analysis were to migrate to that
context, then it would of course wreak havoc with Terry stops
more generally. We do not read the majority opinion to open
that can of worms. Indeed, various judges in the majority
have joined recent opinions suggesting there will be no such

12
The majority opinion notes that Askew alluded to the blue-
sweatshirt point in a footnote in his reply brief to the three-judge
panel. Maj. Op. at 37 n.2. The fact that the point appeared in one
oblique sentence in a reply-brief footnote speaks for itself in terms
of defense counsel’s confidence in the point. Moreover, contrary to
the majority’s suggestion, Askew did not make this point as an
alternative argument for reversal, but rather only to illustrate why
the Court should not allow any identification searches during Terry
stops. And even if Askew had offered this as an alternative
argument for reversal, this Court of course does not consider
alternative arguments for reversal raised for the first time in a
footnote in a reply brief.
 32
spillover effect. See United States v. Abdus-Price, 518 F.3d
926, 929-31 (D.C. Cir. 2008) (Griffith, J., joined by Ginsburg
and Rogers, JJ.); United States v. Goddard, 491 F.3d 457, 462
(D.C. Cir. 2007) (per curiam) (Ginsburg and Tatel, JJ.); see
also Douglas H. Ginsburg, Of Hunches and Mere Hunches:
Two Cheers for Terry, 4 GEO. MASON J.L. ECON. & POL’Y 79,
84 (2007) (“Under Terry, the test for determining whether a
police officer had a reasonable suspicion takes account of the
‘totality of the circumstances,’ which leads courts to review
police practices deferentially. As a result, close calls go to the
police.”) (footnote omitted).

***

We would hold that unzipping Askew’s jacket to search
for a weapon in his waist area was an objectively reasonable
protective step to ensure officer safety after Askew’s
resistance to the initial frisk attempt. In the alternative, we
would hold that the police may reasonably maneuver a
suspect’s outer clothing – such as unzipping a suspect’s outer
jacket – when, as here, doing so could help facilitate a
witness’s identification at a show-up during a Terry stop. We
would uphold the search and affirm the judgment of
conviction on either of those two alternative grounds. We
respectfully dissent.