dissenting:
In America, people who are peaceably and lawfully minding their own business (or who seem to be) have the right to be free from arbitrary police interference. That is the explicit premise of Terry v. Ohio, 392 U.S. 1, 9, 88 S.Ct. 1868, 20 L.Ed.2d 889 (1968) (recognizing the right of every person to be free from unreasonable governmental restraint, interference, *463or intrusion). Forty years later, Terry’s, evolution raises troubling questions. In the crime-plagued, violence-prone, drug-ridden reality of our urban centers, is reasonable suspicion too high a threshold; is it too much to ask?
I
The facts, in brief, are as follows. Police were pursuing a motor vehicle. The driver pulled over and immediately fled on foot in the dark. Police caught only a fleeting glimpse of the suspect: a black man, average height and build, wearing blue jeans and a dark jacket or coat. Not much to go on in a largely black neighborhood, but police broadcast a “lookout” giving the description. They reported the man’s height as 5'8" or 5'10", and his weight as 180 to 190 pounds. About five blocks away, Officer Israel James and three other officers were patrolling in an unmarked Crown Victoria, a car well-known in the area as a police car. When they heard the report of a black man fleeing, they drove closer to the site.
Two blocks from where the suspect had fled, the officers spotted four black men— ranging in height from 5'6" to 6'4" — conversing peaceably in front of a gas station, at least one of them wearing a black coat.1 These were neighborhood residents, who quickly recognized the police car, and one of the men, Vaughan Walker, wanted nothing to do with the police, though he was not involved in any criminal activity. Even before the police stopped, Walker began walking away — but he didn’t get far. The police pulled their car part way into the gas station, blocking the entrance, and all four officers “automatically jumped out,” Tr. of June 21, 2004 Hr’g at 58, wearing jackets bearing the “MPD” logo, holstered weapons, and handcuffs. One of the officers told Mr. Walker to stop; the statement was directed at Mr. Walker, but the message to the others, including appellant, was clear enough. No one attempted to flee. As Officer Figgeroa approached appellant, appellant said, “I have a gun,” or words to that effect. Officer Figgeroa immediately yelled out “gun,” alerting the other officers, who restrained appellant without difficulty and then lifted his shirt, revealing a handgun tucked into his pants. However, the officer who had earlier broadcast the “lookout” could not identify appellant as the fleeing suspect.
Appellant was arrested, and charged with unlawful possession of a firearm, in violation of 18 U.S.C. § 922(g)(1). He moved to suppress the physical evidence and his statements, claiming an illegal stop in violation of his Fourth Amendment rights. The district court held a suppression hearing and denied the motion. The court’s analysis makes clear that, in the court’s view, a Terry stop occurred as soon as the officers arrived on the scene. But the court concluded the officers had sufficient grounds for a stop and frisk based on the broadcast description of the fleeing suspect. Appellant then pled guilty, preserving his right to appeal the Fourth Amendment issue.
We heard argument in appellant’s appeal on October 16, 2006, and on November 9, 2006, we remanded the record to the district court, asking the court to clarify *464the exact “sequence of events surrounding appellant’s stop and seizure, the factors establishing when the police contact became a stop, and the facts known to the police officers at those points.” United States v. Goddard, No. 05-3080 (D.C.Cir. Nov. 9, 2006). The district court responded with an express finding that the Terry stop occurred “as soon as the police officers drove up to the gas station at 2830 Sherman Avenue to investigate.” United States v. Goddard, No. 04-214, slip. op. at 2 (D.D.C. Dec. 18, 2006).
II
Before Terry, warrantless seizures were deemed reasonable only if based on probable cause. Terry carved out a narrow exception to the probable cause requirement, allowing police officers to make limited intrusions if the officer reasonably suspects criminal activity is afoot or public safety is at risk. Thus, before a police officer initiates an investigative stop, the officer must, based on an assessment of the whole picture, “have a particularized and objective basis for suspecting the particular person stopped of criminal activity.” United States v. Cortez, 449 U.S. 411, 417-18, 101 S.Ct. 690, 66 L.Ed.2d 621 (1981). The Court adopted a two-part inquiry for evaluating the reasonableness of an investigative stop involving something less than probable cause: (1) Was the officer’s action justified at its inception? and (2) Was it reasonably related in scope to the circumstances that justified it in the first place. Terry, 392 U.S. at 20-21, 88 S.Ct. 1868; United States v. Sharpe, 470 U.S. 675, 682, 105 S.Ct. 1568, 84 L.Ed.2d 605 (1985).
It is true, of course, that as a standard “reasonable suspicion” is necessarily imprecise. But no matter how low the bar is set, generic racial descriptions devoid of distinctive individualized details cannot, without more, provide police adequate justification for a Terry stop. It is not enough to share the same racial characteristics as a suspect and be in the vicinity.
The “lookout” broadcast at issue here described only a black male, 5'8" to 5'10" in height, about 180 to 190 pounds, wearing a black jacket or coat and blue jeans. The broadcast went out after nightfall, and the officer broadcasting the description made clear the suspect had fled before officers got a good look at him. Therefore, the arresting officers were on notice that the description might be unreliable, a point which they readily acknowledged. At best, one can speak only in the most approximate terms about the height and weight of a man seen running away in the dark, and in those circumstances, any dark-colored coat or jacket might appear to be black. In any case, the officers were definitely not looking for four young men, ranging in height from 5'6" to 6'4", conversing peaceably in front of a gas station. The police can articulate no basis for targeting these men except they were black, they were casually clothed, and they were in the general vicinity of the fleeing suspect. The officers conceded that none of these men seemed agitated, nor did any show signs of recent physical exertion, such as labored breathing or sweating. Apparently, a “lookout” broadcast encompassing virtually any casually dressed black man in the vicinity made all black males fair game. Such a generic description should never be a sufficient basis for a Terry stop.
Though the cases in this circuit are very deferential to the police, as is appropriate, none stands on as thin a record as this one. Our cases establish, as they should, the general rule that when a police officer decides to initiate a Terry stop based solely on a third-party description, that description needs to be specific enough to *465differentiate the suspect from other people who are in the vicinity. Hence, in a case involving a black fugitive in a predominantly black neighborhood, the officer must have something more to go on than race, gender, and standard street-clothes. See United States v. Davis, 235 F.3d 584 (D.C.Cir.2000); United States v. Smart, 98 F.3d 1379 (D.C.Cir.1996); United States v. Simpson, 992 F.2d 1224 (D.C.Cir.1993). That rule is not satisfied in our case, where the facts are extreme enough to debase the rule and considerably extend our holding in Davis.
Ill
A
The majority argues the Terry stop did not occur here until after appellant admitted he had a gun. Usually, of course, a Terry stop occurs only when police actually physically restrain a person or make some verbal statement indicating the person is not free to leave. A verbal statement, however, is not the only way to communicate that message; a “show of authority” can also effect a detention. Terry, 392 U.S. at 19 n. 16, 88 S.Ct. 1868. In the latter case, the applicable standard “is an objective one: not whether the citizen perceived that he was being ordered to restrict his movement, but whether the officer’s words and actions would have conveyed that to a reasonable person.” California v. Hodari D., 499 U.S. 621, 628, 111 S.Ct. 1547, 113 L.Ed.2d 690 (1991). “[A] person has been ‘seized’ within the meaning of the Fourth Amendment only if, in view of all of the circumstances surrounding the incident, a reasonable person would have believed that he was not free to leave.” United States v. Mendenhall, 446 U.S. 544, 554, 100 S.Ct. 1870, 64 L.Ed.2d 497 (1980). “[W]hat constitutes a restraint on liberty prompting a person to conclude that he is not free to ‘leave’ will vary, not only with the particular police conduct at issue, but also with the setting in which the conduct occurs.” Michigan v. Chesternut, 486 U.S. 567, 573, 108 S.Ct. 1975, 100 L.Ed.2d 565 (1988).
Here, as noted, the district court made a determination that the Terry stop occurred “as soon as the police officers drove up to the gas station.” In other words, in the district court’s view, the circumstances at that moment were such that a reasonable person would not have felt free to leave. Generally, we accept the district court’s findings of fact as true unless clearly erroneous, but the question of when a person is seized for Fourth Amendment purposes is a mixed question of fact and law. The historical facts are subject to the clearly erroneous standard of review, Ornelas v. United States, 517 U.S. 690, 699, 116 S.Ct. 1657, 134 L.Ed.2d 911 (1996), but the ultimate conclusion as to whether the facts constitute a Fourth Amendment seizure is a legal question, id. In some cases, however, the district court’s factual findings present only a general sketch, and other findings remain implicit in the court’s ultimate conclusion. For this reason, we must “give due weight to inferences drawn from [the findings] by resident judges.” Id. Therefore, it does not suffice in this case for us to consider only the court’s express findings; we must also consider the entire record, reading the evidence in a way that favors the court’s conclusion as to when the seizure occurred.
B
According to the record before us, four officers pulled up in a car known in the neighborhood to be a police car. They brought the car only part way into the gas station, parking so as to block the entrance, and all four officers jumped out, wearing clothing marked “MPD” and bearing visible weapons. Walker testified he *466had seen police officers “jump out” of the same car on numerous prior occasions. Tr. of June 21, 2004 Hr’g at 57. His use of the slang expression “jump out” conveys his impression that these four officers did not calmly exit the vehicle in order to ask a few questions. Rather, Walker’s phrasing indicates a coordinated police action— or, to put the point in street terms, a bust was going down. This was not a casual contact, and these men were not free to walk away.
The fact that Walker tried to leave and the police instructed him to return only confirms this conclusion. See United States v. Alarcon-Gonzalez, 73 F.3d 289, 292 (10th Cir.1996) (ordering a person to “freeze” effects a seizure of a nearby associate). Walker testified that he began to walk away while the police were still in the car, Tr. of June 21, 2004 Hr’g at 67-68, and he did not get very far before being told to stop, id. at 69-70. Also, the sequence of Walker’s narrative indicates police told him to stop before they confronted appellant. Id. at 58-64. The most natural conclusion, then, is that the police told Walker to stop as they were emerging from the car or immediately thereafter.
Contrary to the majority’s assertion, see maj. op. at 459, the record contains no explicit finding by the district court that Officer Figgeroa had already yelled “gun” when police instructed Walker to stop and return. The court does state that, at the time Officer Figgeroa yelled “gun,” Walker was “walking away” with “his back towards” the others, Tr. of June 21, 2004 Hr’g at 84, but the court may have been referring to the fact that, after police told Walker to return, they then instructed him to go over “[tjowards the rail, in between the pay phone and the air pump,” id. at 60-61. Moreover, in the passage the majority references, the district court was not making a finding but only summarizing Walker’s testimony. The court introduced this passage by stating that Officer James’s testimony “was corroborated by the testimony of Mr. Walker.” Id. at 84 (emphasis added). The court then expressly referred to “notes” the court had taken of Walker’s testimony and, in that context, made the comment the majority calls a finding. Id. Significantly, the court’s notes of Walker’s testimony were apparently inaccurate, for Walker never actually stated that the gun was found while he was walking away. See id. at 68-69. In short, the court’s asserted “finding” is ambiguous at best, and in any case, the record contains no evidence to support it.
In addition, the fact that appellant blurted out that he had a gun evidences his subjective impression that the police were going to search him, and the officers confirmed that they did not stop merely to ask these individuals what, if anything, they might have seen. Rather, Officer Israel James testified that the purpose in stopping was “to attempt to determine whether or not ... one of the individuals was the one who actually fled the scene.” Id. at 7. The subjective intent of the police officer is perhaps not determinative in this context, Mendenhall, 446 U.S. at 554 n. 6, 100 S.Ct. 1870, but it does tend to support the district court’s conclusion that the police conveyed the message, through their actions and demeanor, that no one was free to leave. Thus, I find the evidence easily supports the court’s ultimate conclusion as to when the seizure occurred.
True, the district court concluded the seizure occurred “as soon as the police officers drove up,” whereas it more accurately occurred when the police drove up, jumped out, and told Walker not to leave, but the court’s statement can reasonably be taken in a general sense. The court’s meaning was only that the seizure did not *467occur later, when appellant admitted to having a gun. This interpretation of the court’s statement is consistent with the court’s -application of Terry, in which the court focused solely on events as they existed prior to appellant’s admission.
IV
Not all encounters with police implicate the Fourth Amendment; rather, different police-citizen interactions trigger different standards. Consensual encounters are entirely outside the scope of the Amendment. Investigative detentions — commonly known as Terry stops — must be supported by reasonable suspicion of criminal activity. Custodial arrests require probable cause.
Terry was a sensible decision. It recognized that armed and dangerous offenders present a serious threat to the safety of both the police and the public, and it allowed a greater role for street savvy and police judgment. The holding in Terry permits police officers to take reasonable steps, based on their experience and expert training, to prevent a crime before it occurs and to maintain safety, justifying their actions with only a “reasonable, artic-ulable suspicion.” Illinois v. Wardlow, 528 U.S. 119, 123, 120 S.Ct. 673, 145 L.Ed.2d 570 (2000) (characterizing Terry's holding). But Terry was never intended to upend completely the warrant and probable cause requirements, and it does so if the lower courts give police a free hand to search whenever there is the coincidence of a black man’s recent flight from the police and another black man’s presence in the vicinity. This lax standard perverts the core requirement that police have a “particularized and objective basis for suspecting the particular person ... of criminal activity.” Cortez, 449 U.S. at 417-18, 101 S.Ct. 690.
I have already stated my view that the stop in this case occurred when the police jumped out of their vehicle and told Walker not to leave. This case, however, illuminates a broader issue. The majority’s analytic approach turns almost entirely on the question of when the stop occurred. Having identified the stop (“when one of the officers yelled ‘gun,’ ” maj. op. at 462), everything preceding it passes out of the analysis, comprehended under the familiar axiom that officers are free “to make a contact — to approach individuals and interact with them — without reasonable suspicion.” Id. at 461. But Terry “emphatically reject[s]” a logic that would “isolate from constitutional scrutiny the initial stages of the contact between the policeman and the citizen.” 392 U.S. at 16-17, 88 S.Ct. 1868. Rather, Terry limits “the initiation,” as well as “the scope,” of investigative detentions. Id. at 17, 88 S.Ct. 1868. In my view, truly consensual conversation (“Have you seen this man?”) in which police are on equal terms with ordinary citizens is very different from a confrontation in which police target a particular person because they suspect him to be the criminal described in a police broadcast. It makes sense for the Fourth Amendment not to regulate casual conversation between police and citizens. But how could the Fourth Amendment stay dormant when police confront someone because they suspect him to be a criminal? Targeted police investigation activates the Amendment. In the instant case, the confrontation began as an investigative stop; the police claimed a particularized suspicion about four specific men — a claim that could not conceivably be supported by the objective information available. Not even the majority claims the initial investigation was justified. Rather, the majority employs an analytic approach that makes justifying the confrontation constitutionally unnecessary. This is certainly consistent with the *468trend of recent cases applying Terry, but I think it is an error. When police focus their investigative attention on someone and choose to confront that person, they must have a constitutionally adequate reason for doing so. Nothing in Terry suggests that the police can employ arbitrary criteria to select their targets and then artificially create the circumstances that justify application of the Terry exception.
Terry itself acknowledged the unacceptable social cost of Fourth Amendment violations, discussing “police-community tensions in the crowded centers of our Nation’s cities” and the “wholesale harassment by certain elements of the police community” of minorities, “particularly Negroes.” Id. at 12, 14, 88 S.Ct. 1868. There are two specific aspects to this social problem. First, inappropriate use of Terry in America’s minority neighborhoods offends the principle of equal justice under law. For, as we all know, courts would not approve the search of four men in business attire, conversing peaceably in front of a Starbucks, if the only basis for the search was a “lookout” broadcast specifying a white man, medium height and build, wearing a business suit. Second, excessive Terry searches set poor black communities and the police on opposing sides of pitched battle. At a minimum, these perpetual intrusions leave young black men feeling bruised and insulted. Often enough, the anger leads to confrontations with the police, sometimes ■with violent or lethal consequences. As Terry put it, when police officers use stop-and-frisk tactics against minorities to “maintain the power image” of the officers, the aim is “sometimes accomplished by humiliating anyone who attempts to undermine police control of the streets.” Id. at 15 n. 11, 88 S.Ct. 1868 (citation and internal quotation marks omitted). Is the even-handed application of the Terry standard too much to ask?
The facts of this case lead me to wonder if Terry’s prudent constraints on police conduct have been forgotten in our frustration over city life plagued with drug trafficking and violent crime. As a result, what we are now tempted to enforce is not Tern/ but the rule that, in a high-crime neighborhood, being young, male, and black creates reasonable, articulable suspicion. See David A. Harris, Factors for Reasonable Suspicion: When Black and Poor Means Stopped and Frisked, 69 Ind. L.J. 659 (1994). Here, four men were stopped. There was no constitutionally adequate justification for the initial confrontation. Three of them were innocent of criminal activity, but nevertheless faced the indignity of being placed against a rail and searched. Vaughan Walker testified he started walking away as soon as he saw the police car because he “didn’t feel like being harassed.” The lesson of today’s decision is clear: he has no choice.
When the ostensibly neutral principles set forth in Terry are thus applied, what was created to be a carefully outlined exception to the Fourth Amendment’s warrant and probable cause requirements is transformed into a general warrant — a police license to search out crime by playing the odds, relying on hunch, intuition, street smarts, and stereotypes. The odds are good, although the crimes charged are too often unrelated to the “suspicion” that led to the stop. After all, in the instant case, a search of four almost randomly selected black men in a high-crime neighborhood turned up one with a gun — a 25% success rate. The instances where a search discloses no criminal activity are brushed aside and forgotten — except in the neighborhoods where the sense of unfairness and frustration festers and swells and then *469seeps out in little acts of defiance or mindless eruptions of rage.
Acknowledging that no opinion “can comprehend the protean variety of the street encounter,” the Terry court left intact the courts’ “traditional responsibility to guard against police conduct which is overbearing or harassing, or which trenches upon personal security without the objective evidentiary justification which the Constitution requires.” 392 U.S. at 15, 88 S.Ct. 1868. If this court is ever going to draw such a line, this is the case in which to draw it.
V
I am not advocating here a broad new rule of constitutional law that will inhibit police. Police are free to contact members of the public without every contact being construed as a seizure. See, e.g., United States v. Nurse, 916 F.2d 20, 23 (D.C.Cir.1990). Moreover, the presence of two or more officers does not, by itself, transform a contact into a seizure. See, e.g., United States v. Tavolacci, 895 F.2d 1423, 1425 (D.C.Cir.1990). The case before us presents very specific circumstances: The police targeted these men in response to a specific broadcast about a fleeing suspect. They pulled part way into the gas station, blocking the entrance; four officers jumped from the car at once, wearing clothing marked “MPD”; and they immediately told someone who was attempting to leave to return. That is not how one would expect police to proceed when they are merely making inquiries, and it is very different from the facts of cases in which police did no more than make their presence felt. See, e.g., Chesternut, 486 U.S. at 569, 575, 108 S.Ct. 1975; United States v. Johnson, 212 F.3d 1313, 1317 (D.C.Cir.2000). Most important, we have here a finding by the district court that the police conduct would have communicated to a reasonable person at the scene that he or she was not free to leave, and the best interpretation of the evidence in the record reasonably supports that legal conclusion.
We should adopt the district court’s finding that the Terry stop occurred when four police officers pulled into the entrance of the gas station, sprang simultaneously from their car, and instructed Walker not to leave. We should then reject the court’s legal conclusion that the generic description of a medium-sized black man fleeing justifies the Terry stop that occurred here. By the district court’s logic, police would have been able to stop virtually every casually dressed black man within a sixteen-block radius of the crime. Because the meager facts available to the officers did not come close to justifying the initial contact, the fruit they subsequently shook out of the poisonous tree should be excluded. Is that too much to ask? It is what the constitution requires; it is just enough.
Accordingly, I respectfully dissent.
. The testimony regarding the clothes the men were wearing is quite unhelpful. Officer James testified that their clothes ''[flor the most part” matched the description in the broadcast, Tr. of June 21, 2004 Hr’g at 17, adding that appellant (who is over six feet tall) was wearing a ''[b]lack coat and blue jeans,” id. at 49. At one point, the trial judge describes the men as wearing "jeans and a dark shirt,” id. at 83; at another point, "blue jeans and a dark jacket,” id. at 86. In neither case is the court's statement supported by specific testimony.