dissenting.
Because the strongest advocates of Fourth Amendment rights are frequently criminals, it is easy to forget that our interpretations of such rights apply to the innocent and the guilty alike. Illinois v. Gates, 462 U. S. 213, 290 (1983) (Brennan, J., dissenting). In the present case, the chain of events set in motion when respondent Andrew Sokolow was stopped by Drug Enforcement Administration (DEA) agents at Honolulu International Airport led to the discovery of cocaine and, ultimately, to Sokolow’s conviction for drug trafficking. But in sustaining this conviction on the ground that the agents reasonably suspected Sokolow of ongoing criminal activity, the Court diminishes the rights of all citizens “to be secure in their persons,” U. S. Const., Arndt. 4, as they *12traverse the Nation’s airports. Finding this result constitutionally impermissible, I dissent.
The Fourth Amendment cabins government’s authority to intrude on personal privacy and security by requiring that searches and seizures usually be supported by a showing of probable cause. The reasonable-suspicion standard is a derivation of the probable-cause command, applicable only to those brief detentions which fall short of being full-scale searches and seizures and which are necessitated by law enforcement exigencies such as the need to stop ongoing crimes, to prevent imminent crimes, and to protect law enforcement officers in highly charged situations. Terry v. Ohio, 392 U. S. 1, 30 (1968). By requiring reasonable suspicion as a prerequisite to such seizures, the Fourth Amendment protects innocent persons from being subjected to “overbearing or harassing” police conduct carried out solely on the basis of imprecise stereotypes of what criminals look like, or on the basis of irrelevant personal characteristics such as race. Id., at 14-15, and n. 11 (citation omitted).
To deter such egregious police behavior, we have held that a suspicion is not reasonable unless officers have based it on “specific and articulable facts.” Id., at 21; see also United States v. Brignoni-Ponce, 422 U. S. 873, 880 (1975). It is not enough to suspect that an individual has committed crimes in the past, harbors unconsummated criminal designs, or has the propensity to commit crimes. On the contrary, before detaining an individual, law enforcement officers must reasonably suspect that he is engaged in, or poised to commit, a criminal act at that moment. See, e. g., Brown v. Texas, 443 U. S. 47, 51 (1979) (to detain, officers must “have a reasonable suspicion, based on objective facts, that the individual is involved in criminal activity”); Terry, supra, at 30 (reasonable suspicion exists only where policeman reasonably concludes, inter alia, “that criminal activity may be afoot”). The rationale for permitting brief, warrantless seizures is, after all, that it is impractical to demand strict compliance *13with the Fourth Amendment’s ordinary probable-cause requirement in the face of ongoing or imminent criminal activity demanding “swift action predicated upon the on-the-spot observations of the officer on the beat.” Terry, supra, at 20. Observations raising suspicions of past criminality demand no such immediate action, but instead should appropriately trigger routine police investigation, which may ultimately generate sufficient information to blossom into probable cause.
Evaluated against this standard, the facts about Andrew Sokolow known to the DEA agents at the time they stopped him fall short of reasonably indicating that he was engaged at the time in criminal activity. It is highly significant that the DEA agents stopped Sokolow because he matched one of the DEA’s “profiles” of a paradigmatic drug courier. In my view, a law enforcement officer’s mechanistic application of a formula of personal and behavioral traits in deciding whom to detain can only dull the officer’s ability and determination to make sensitive and fact-specific inferences “in light of his experience,” Terry, supra, at 27, particularly in ambiguous or borderline cases. Reflexive reliance on a profile of drug courier characteristics runs a far greater risk than does ordinary, case-by-case police work of subjecting innocent individuals to unwarranted police harassment and detention. This risk is enhanced by the profile’s “chameleon-like way of adapting to any particular set of observations.” 831 F. 2d 1413, 1418 (CA9 1987). Compare, e. g., United States v. Moore, 675 F. 2d 802, 803 (CA6 1982) (suspect was first to deplane), cert. denied, 460 U. S. 1068 (1983), with United States v. Mendenhall, 446 U. S. 544, 564 (1980) (last to deplane), with United States v. Buenaventura-Ariza, 615 F. 2d 29, 31 (CA2 1980) (deplaned from middle); United States v. Sullivan, 625 F. 2d 9, 12 (CA4 1980) (one-way tickets), with United States v. Craemer, 555 F. 2d 594, 595 (CA6 1977) (round-trip tickets), with United States v. McCaleb, 552 F. 2d 717, 720 (CA6 1977) (nonstop flight), with United States v. Sokolow, 808 F. 2d 1366, 1370 (CA9), vacated, 831 F. 2d 1413 *14(1987) (case below) (changed planes); Craemer, supra, at 595 (no luggage), with United States v. Sanford, 658 F. 2d 342, 343 (CA5 1981) (gym bag), cert. denied, 455 U. S. 991 (1982), with Sullivan, supra, at 12 (new suitcases); United States v. Smith, 574 F. 2d 882, 883 (CA6 1978) (traveling alone), with United States v. Fry, 622 F. 2d 1218, 1219 (CA5 1980) (traveling with companion); United States v. Andrews, 600 F. 2d 563, 566 (CA6 1979) (acted nervously), cert. denied sub nom. Brooks v. United States, 444 U. S. 878 (1979), with United States v. Himmelwright, 551 F. 2d 991, 992 (CA5) (acted too calmly), cert. denied, 434 U. S. 902 (1977). In asserting that it is not “somehow” relevant that the agents who stopped Sokolow did so in reliance on a prefabricated profile of criminal characteristics, ante, at 10, the majority thus ducks serious issues relating to a questionable law enforcement practice, to address the validity of which we granted certiorari in this case.1
That the factors comprising the drug courier profile relied on in this case are especially dubious indices of ongoing criminal activity is underscored by Reid v. Georgia, 448 U. S. 438 (1980), a strikingly similar case. There, four facts, encoded in a drug courier profile, were alleged in support of the DEA’s detention of a suspect at the Atlanta Airport. First, Reid had arrived from Fort Lauderdale, Florida, a source city for cocaine. Second, he arrived in the early morning, when law enforcement activity is diminished. Third, he and his companion appeared to have no luggage other than their shoulder bags. And fourth, he and his companion appeared to be trying to conceal the fact that they were traveling together. Id., at 440-441.
This collection of facts, we held, was inadequate to support a finding of reasonable suspicion. All but the last of these facts, we observed, “describe a very large category of pre*15sumably innocent travelers, who would be subject to virtually random seizures were the Court to conclude that as little foundation as there was in this case could justify a seizure.” Id., at 441. The sole fact that suggested criminal activity was that Reid “preceded another person and occasionally looked backward at him as they proceeded through the concourse.” Ibid. This observation did not of itself provide a reasonable basis for suspecting wrongdoing, for inferring criminal activity from such evidence reflected no more than an “‘inchoate and unparticularized suspicion or “hunch.”’” Ibid., quoting Terry, 392 U. S., at 27.2
The facts known to the DEA agents at the time they detained the traveler in this case are scarcely more suggestive of ongoing criminal activity than those in Reid. Unlike traveler Reid, who sought to conceal the fact that he was traveling with a companion, and who even attempted to run away after being approached by a DEA agent, 448 U. S., at 439, traveler Sokolow gave no indications of evasive activity. On the contrary, the sole behavioral detail about Sokolow noted by the DEA agents was that he was nervous. With news accounts proliferating of plane crashes, near collisions, and air terrorism, there are manifold and good reasons for being agitated while awaiting a flight, reasons that have nothing to do with one’s involvement in a criminal endeavor.
The remaining circumstantial facts known about Sokolow, considered either singly or together, are scarcely indicative of criminal activity. Like the information disavowed in Reid as nonprobative, the fact that Sokolow took a brief trip to a *16resort city for which he brought only carry-on luggage also “describe[s] a very large category of presumably innocent travelers.” Id., at 441. That Sokolow embarked from Miami, “a source city for illicit drugs,” ante, at 3, is no more suggestive of illegality; thousands of innocent persons travel from “source cities” every day and, judging from the DEA’s testimony in past cases, nearly every major city in the country may be characterized as a source or distribution city. See, e. g., Buenaventura-Ariza, 615 F. 2d, at 31, n. 5. That Sokolow had his phone listed in another person’s name also does not support the majority’s assertion that the DEA agents reasonably believed Sokolow was using an alias; it is commonplace to have one’s phone registered in the name of a roommate, which, it later turned out, was precisely what Sokolow had done.3 That Sokolow was dressed in a black jumpsuit and wore gold jewelry also provides no grounds for suspecting wrongdoing, the majority’s repeated and unexplained allusions to Sokolow’s style of dress notwithstanding. Ante, at 4, 5. For law enforcement officers to base a search, even in part, on a “pop” guess that persons dressed in a particular fashion are likely to commit crimes not only stretches the concept of reasonable suspicion beyond recognition, but also is inimical to the self-expression which the choice of wardrobe may provide.
Finally, that Sokolow paid for his tickets in cash indicates no imminent or ongoing criminal activity. The majority “feel[s] confident” that “[m]ost business travelers . . . purchase airline tickets by credit card or check.” Ante, at 8. Why the majority confines its focus only to “business travelers” I do not know, but I would not so lightly infer ongoing crime from the use of legal tender. Making major cash purchases, while surely less common today, may simply reflect the traveler’s aversion to, or inability to obtain, plastic *17money. Conceivably, a person who spends large amounts of cash may be trying to launder his proceeds from past criminal enterprises by converting them into goods and services. But, as I have noted, investigating completed episodes of crime goes beyond the appropriately limited purview of the brief, Terry-style seizure. Moreover, it is unreasonable to suggest that, had Sokolow left the airport, he would have been gone forever and thus immune from subsequent investigation. Ante, at 11. Sokolow, after all, had given the airline his phone number, and the DEA, having ascertained that it was indeed Sokolow’s voice on the answering machine at that number, could have learned from that information where Sokolow resided.
The fact is that, unlike the taking of patently evasive action, Florida v. Rodriguez, 469 U. S. 1, 6 (1984), the use of an alias, Florida v. Royer, 460 U. S. 491, 502 (1983), the casing of a store, Terry, supra, at 6, or the provision of a reliable report from an informant that wrongdoing is imminent, Illinois v. Gates, 462 U. S., at 225-227, nothing about the characteristics shown by airport traveler Sokolow reasonably suggests that criminal activity is afoot. The majority’s hasty conclusion to the contrary serves only to indicate its willingness, when drug crimes or antidrug policies are at issue, to give short shrift to constitutional rights. See, e. g., Skinner v. Railway Labor Executives’ Assn., 489 U. S. 602, 636 (1989) (Marshall, J., dissenting).4 In requiring that seizures be based on at least some evidence of criminal conduct, 831 F. 2d, at 1419, the Court of Appeals was faithful to the Fourth Amendment principle that law enforcement offi*18cers must reasonably suspect a person of criminal activity before they can detain him. Because today’s decision, though limited to its facts, ante, at 11, disobeys this important constitutional command, I dissent.
Even if such profiles had reliable predictive value, their utility would be short lived, for drug couriers will adapt their behavior to sidestep detection from profile-focused officers.
Nor was Reid a close case: eight Members of the Court found the challenged detention insupportable, five of whom saw fit to dispose of the case by reversing the court below in a per curiam opinion. In a separate concurrence, Justice Powell, joined by Chief Justice Burger and Justice Blackmun, agreed that “the fragmentary facts apparently relied on by the DEA agents” provided “no justification” for Reid’s detention. 448 U. S., at 442, n. 1. Only then-JusTiCE Rehnquist, the author of today’s majority opinion, dissented, on the ground that the police conduct involved did not implicate Reid’s constitutional rights. Id., at 442.
That Sokolow was, in fact, using an alias was not known to the DEA agents until after they detained him. Thus, it cannot legitimately be considered as a basis for the seizure in this case.
The majority also contends that it is not relevant that the DEA agents, in forcibly stopping Sokolow rather than simply speaking with him, did not “use the least intrusive means available.” Ante, at 10. On the contrary, the manner in which a search is carried out — and particularly whether law enforcement officers have taken needlessly intrusive steps — is a highly important index of reasonableness under Fourth Amendment doctrine. See, e. g., Winston v. Lee, 470 U. S. 753, 760-761 (1985).