delivered the opinion of the Court.

In this case we consider whether the rule of Pennsylvania v. Mimms, 434 U. S. 106 (1977) (per curiam), that a police officer may as a matter of course order the driver of a lawfully stopped car to exit his vehicle, extends to passengers as well. We hold that it does.

At about 7:30 p.m. on a June evening, Maryland state trooper David Hughes observed a passenger car driving southbound on 1-95 in Baltimore County at a speed of 64 miles per hour. The posted speed limit was 55 miles per hour, and the car had no regular license tag; there was a torn piece of paper reading “Enterprise Rent-A-Car” dangling from its rear. Hughes activated his lights and sirens, signaling the car to pull over, but it continued driving for another mile and a half until it finally did so.

During the pursuit, Hughes noticed that there were three occupants in the car and that the two passengers turned to look at him several times, repeatedly ducking below sight level and then reappearing. As Hughes approached the car on foot, the driver alighted and met him halfway. The driver was trembling and appeared extremely nervous, but nonetheless produced a valid Connecticut driver’s license. Hughes instructed him to return to the car and retrieve the rental documents, and he complied. During this encounter, Hughes noticed that the front-seat passenger, respondent Jerry Lee Wilson, was sweating and also appeared extremely *411nervous. While the driver was sitting in the driver’s seat looking for the rental papers, Hughes ordered Wilson out of the car.

When Wilson exited the car, a quantity of crack cocaine fell to the ground. Wilson was then arrested and charged with possession of cocaine with intent to distribute. Before trial, Wilson moved to suppress the evidence, arguing that Hughes’ ordering him out of the ear constituted an unreasonable seizure under the Fourth Amendment. The Circuit Court for Baltimore County agreed, and granted respondent’s motion to suppress. On appeal, the Court of Special Appeals of Maryland affirmed, 106 Md. App. 24, 664 A. 2d 1 (1995), ruling that Pennsylvania v. Mimms does not apply to passengers. The Court of Appeals of Maryland denied certiorari. 340 Md. 502, 667 A. 2d 342 (1995). We granted certiorari, 518 U. S. 1003 (1996), and now reverse.

In Mimms, we considered a traffic stop much like the one before us today. There, Mimms had been stopped for driving with an expired license plate, and the officer asked him to step out of his car. When Mimms did so, the officer noticed a bulge in his jacket that proved to be a .38-caliber revolver, whereupon Mimms was arrested for carrying a concealed deadly weapon. Mimms, like Wilson, urged the suppression of the evidence on the ground that the officer’s ordering him out of the car was an unreasonable seizure, and the Pennsylvania Supreme Court, like the Court of Special Appeals of Maryland, agreed.

We reversed, explaining that “[t]he 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,’ ” 434 U. S., at 108-109 (quoting Terry v. Ohio, 392 U. S. 1, 19 (1968)), and that reasonableness “depends ‘on a balance between the public interest and the individual’s right to personal security free • from arbitrary interference by law officers,’ ” 434 U. S., at 109 (quoting United States v. Brignoni-Ponce, 422 U. S. 873, *412878 (1975)). On the public interest side of the balance, we noted that the State “freely concede[d]” that there had been nothing unusual or suspicious to justify ordering Mimms out of the car, but that it was the officer’s “practice to order all drivers [stopped in traffic stops] out of their vehicles as a matter of course” as a “precautionary measure” to protect the officer’s safety. 434 U. S., at 109-110. We thought it “too plain for argument” that this justification — officer safety — was “both legitimate and weighty.” Id., at 110. In addition, we observed that the danger to the officer of standing by the driver’s door and in the path of oncoming traffic might also be “appreciable.” Id., at 111.

On the other side of the balance, we considered the intrusion into the driver’s liberty occasioned by the officer’s ordering him out of the car. Noting that the driver’s car was already validly stopped for a traffic infraction, we deemed the additional intrusion of asking him to step outside his car “de minimis.” Ibid. Accordingly, we concluded that “once a motor vehicle has been lawfully detained for a traffic violation, the police officers may order the driver to get out of the vehicle without violating the Fourth Amendment’s proscription of unreasonable seizures.” Id., at 111, n. 6.

Respondent urges, and the lower courts agreed, that this per se rule does not apply to Wilson because he was a passenger, not the driver. Maryland, in turn, argues that we have already implicitly decided this question by our statement in Michigan v. Long, 463 U. S. 1032 (1983), that “[i]n [Mimms], we held that police may order persons out of an automobile during a stop for a traffic violation,” id., at 1047-1048 (emphasis added), and by Justice Powell’s statement in Rakas v. Illinois, 439 U. S. 128 (1978), that “this Court determined in [Mimms] that passengers in automobiles have no Fourth Amendment right not to be ordered from their vehicle, once a proper stop is made,” id., at 155, n. 4 (Powell, J., joined by Burger, C. J., concurring) (emphasis added). We agree with respondent that the former statement was dictum, and the *413latter was contained in a concurrence, so that neither constitutes binding precedent.

We must therefore now decide whether the rule of Mimms applies to passengers as well as to drivers.1 On the public interest side of the balance, the same weighty interest in officer safety is present regardless of whether the occupant of the stopped car is a driver or passenger. Regrettably, traffic stops may be dangerous encounters. In 1994 alone, there were 5,762 officer assaults and 11 officers killed during traffic pursuits and stops. Federal Bureau of Investigation, Uniform Crime Reports: Law Enforcement Officers Killed and Assaulted 71, 33 (1994). In the case of passengers, the danger of the officer’s standing in the path of oncoming traffic would not be present except in the case of a passenger in the left rear seat, but the fact that there is more than one occupant of the vehicle increases the possible sources of harm to the officer.2

On the personal liberty side of the balance, the case for the passengers is in one sense stronger than that for the driver. There is probable cause to believe that the driver has committed a minor vehicular offense, but there is no such reason to stop or detain the passengers. But as a practical *414matter, the passengers are already stopped by virtue of the stop of the vehicle. The only change in their circumstances which will result from ordering them out of the car is that they will be outside of, rather than inside of, the stopped car. Outside the car, the passengers will be denied access to any possible weapon that might be concealed in the interior of the passenger compartment. It would seem that the possibility of a violent encounter stems not from the ordinary reaction of a motorist stopped for a speeding violation, but from the fact that evidence of a more serious crime might be uncovered during the stop. And the motivation of a passenger to employ violence to prevent apprehension of such a crime is every bit as great as that of the driver. 452

We think that our opinion in Michigan v. Summers, 452 U. S. 692 (1981), offers guidance by analogy here. There the police had obtained a search warrant for contraband thought to be located in a residence, but when they arrived to execute the warrant they found Summers coming down the front steps. The question in the case depended “upon a determination whether the officers had the authority to require him to re-enter the house and to remain there while they conducted their search.” Id., at 695. In holding as it did, the Court said:

“Although no special danger to the police is suggested by the evidence in this record, the execution of a warrant to search for narcotics is the kind of transaction that may give rise to sudden violence or frantic efforts to conceal or destroy evidence. The risk of harm to both the police and the occupants is minimized if the officers routinely exercise unquestioned command of the situation.” Id., at 702-703 (footnote omitted).

In summary, danger to an officer from a traffic stop is likely to be greater when there are passengers in addition to the driver in the stopped car. While there is not the same basis for ordering the passengers out of the car as there is *415for ordering the driver out, the additional intrusion on the .* passenger is minimal. We therefore hold that an officer making a traffic stop may order passengers to get out of the car pending completion of the stop.3

The judgment of the Court of Special Appeals of Maryland is reversed, and the case is remanded for proceedings not inconsistent with this opinion.

It is so ordered.

dissenting.

In Pennsylvania v. Mimms, 434 U. S. 106 (1977) (per cu-riam), the Court answered the “narrow question” whether an “incremental intrusion” on the liberty of a person who had been lawfully seized was reasonable. Id., at 109. This case, in contrast, raises a separate and significant question concerning the power of the State to make an initial seizure of persons who are not even suspected of having violated , the law.

My concern is not with the ultimate disposition of this particular case, but rather with the literally millions of other cases that will be affected by the rule the Court announces. Though the question is not before us, I am satisfied that— under the rationale of Terry v. Ohio, 392 U. S. 1 (1968) — if a police officer conducting a traffic stop has an articulable suspicion of possible danger, the officer may order passengers to exit the vehicle as a defensive tactic without running afoul of the Fourth Amendment. Accordingly, I assume that the facts recited in the majority’s opinion provided a valid justi*416fication for this officer’s order commanding the passengers to get out of this vehicle.1 But the Court’s ruling goes much further. It applies equally to traffic stops in which there is not even a scintilla of evidence of any potential risk to the police officer. In those cases, I firmly believe that the Fourth Amendment prohibits routine and arbitrary seizures of obviously innocent citizens.

I

The majority suggests that the personal liberty interest at stake here, which is admittedly “stronger” than that at issue in Mimms, is outweighed by the need to ensure officer safety. Ante, at 413, 414-415. The Court correctly observes that “traffic stops may be dangerous encounters.” Ante, at 413. The magnitude of the danger to police officers is reflected in the statistic that, in 1994 alone, “there were 5,762 officer assaults and 11 officers killed during traffic pursuits and stops.” Ibid. There is, unquestionably, a strong public interest in minimizing the number of such assaults and fatalities. The Court’s statistics, however, provide no support for the conclusion that its ruling will have any such effect.

Those statistics do not tell us how many of the incidents involved passengers. Assuming that many of the assaults were committed by passengers, we do not know how many occurred after the passenger got out of the vehicle, how many took place while the passenger remained in the vehicle, or indeed, whether any of them could have been prevented *417by an order commanding the passengers to exit.2 There is no indication that the number of assaults was smaller in jurisdictions where officers may order passengers to exit the vehicle without any suspicion than in jurisdictions where they were then prohibited from doing so. Indeed, there is > no indication that any of the assaults occurred when there •was a complete absence of any articulable basis for concern about the officer’s safety — the only condition under which I would hold that the Fourth Amendment prohibits an order commanding passengers to exit a vehicle. In short, the statistics are as consistent with the hypothesis that ordering passengers to get out of a vehicle increases the danger of assault as with the hypothesis that it reduces that risk.

Furthermore, any limited additional risk to police officers must be weighed against the unnecessary invasion that will be imposed on innocent citizens under the majority’s rule in the tremendous number of routine stops that occur each day. We have long recognized that “[bjecause of the extensive regulation of motor vehicles and traffic . . . the extent of police-citizen contact involving automobiles will be substantially greater than police-citizen contact in a home or office.” Cady v. Dombrowski, 413 U. S. 433, 441 (1973).3 Most traffic *418stops involve otherwise law-abiding citizens who have committed minor traffic offenses. A strong interest in arriving at a destination — to deliver a patient to a hospital, to witness a kickoff, or to get to work on time — will often explain a traffic violation without justifying it. In the aggregate, these stops amount to significant law enforcement activity.

Indeed, the number of stops in which an officer is actually at risk is dwarfed by the far greater number of routine stops. If Maryland’s share of the national total is about average, the State probably experiences about 100 officer assaults each year during traffic stops and pursuits. Making the unlikely assumption that passengers are responsible for one-fourth of the total assaults, it appears that the Court’s new rule would provide a potential benefit to Maryland officers in only roughly 25 stops a year.4 These stops represent a minuscule portion of the total. In Maryland alone, there are something on the order of one million traffic stops each year.5 Assuming that there are passengers in about half of the cars stopped, the majority’s rule is of some possible advantage to police in only about one out of every twenty thousand traffic stops in which there is a passenger in the car. And, any benefit is extremely marginal. In the overwhelming majority of cases posing a real threat, the officer would almost *419certainly have some ground to suspect danger that would justify ordering passengers out of the ear.

In contrast, the potential daily burden on thousands of innocent citizens is obvious. That burden may well be “minimal” in individual cases. Ante, at 415. But countless citizens who cherish individual liberty and are offended, embarrassed, and sometimes provoked by arbitrary official commands may well consider the burden to be significant.6 In all events, the aggregation of thousands upon thousands of petty indignities has an impact on freedom that I would characterize as substantial, and which in my view clearly outweighs the evanescent safety concerns pressed by the majority.

II

The Court concludes today that the balance of convenience and danger that supported its holding in Mimms applies to passengers of lawfully stopped cars as well as drivers. In Mimms itself, however, the Court emphasized the fact that the intrusion into the driver’s liberty at stake was “occasioned not by the initial stop of the vehicle, which was admittedly justified, but by the order to get out of the car.” 434 U. S., at 111. The conclusion that “this additional intrusion can only be described as de minimis” rested on the premise that the “police have already lawfully decided that the driver shall be briefly detained.” Ibid.7

*420In this case as well, the intrusion on the passengers’ liberty occasioned by the initial stop of the vehicle is not challenged. That intrusion was a necessary by-product of the lawful detention of the driver. But the passengers had not yet been seized at the time the car was pulled over, any more than a traffic jam caused by construction or other state-imposed delay not directed at a particular individual constitutes a seizure of that person. The question is whether a passenger in a lawfully stopped car may be seized, by an order to get out of the vehicle, without any evidence whatsoever that he or she poses a threat to the officer or has committed an offense.8

To order passengers about during the course of a traffic stop, insisting that they exit and remain outside the car, can hardly be classified as a de minimis intrusion. The traffic violation sufficiently justifies subjecting the driver to detention and some police control for the time necessary to conclude the business of the stop. The restraint on the liberty of blameless passengers that the majority permits is, in contrast, entirely arbitrary.9 or

In my view, wholly innocent passengers in a taxi, bus, or private car have a constitutionally protected right to decide whether to remain comfortably seated within the vehicle rather than exposing themselves to the elements and the observation of curious bystanders. The. Constitution should not be read to permit law enforcement officers to order innocent passengers about simply because they have the misfor*421tune to be seated in a car whose driver has committed a minor traffic offense.

Unfortunately, the effect of the Court’s new rule on the law may turn out to be far more significant than its immediate impact on individual liberty. Throughout most of our history the Fourth Amendment embodied a general rule requiring that official searches and seizures be authorized by a warrant, issued “upon probable cause, supported by Oath or affirmation, and particularly describing the place to be searched, and the persons or things to be seized.”10 During the prohibition era, the exceptions for warrantless searches supported by probable cause started to replace the general rule.11 In 1968, in the landmark “stop and frisk” case Terry v. Ohio, 392 U. S. 1 (1968), the Court placed its stamp of approval on seizures supported by specific and articulable facts that did not establish probable cause. The Court crafted Terry as a narrow exception to the general rule that “the police must, whenever practicable, obtain advance judicial approval of searches and seizures through the warrant procedure.” Id., at 20. The intended scope of the Court’s major departure from prior practice was reflected in its statement that the “demand for specificity in the information upon which police action is predicated is the central teaching of this Court’s Fourth Amendment jurisprudence.” Id., at 21, n. 18; see also id., at 27. In the 1970’s, the Court twice rejected attempts to justify suspicionless seizures that caused only “modest” intrusions on the liberty of passengers in automobiles. United States v. Brignoni-Ponce, 422 U. S. 873, 879-880 (1975); Delaware v. Prouse, 440 U. S. 648, 662-663 *422(1979).12 Today, however, the Court takes the unprecedented step of authorizing seizures that are unsupported by any individualized suspicion whatsoever.

The Court’s conclusion seems to rest on the assumption that the constitutional protection against “unreasonable” seizures requires nothing more than a hypothetically rational basis for intrusions on individual liberty. How' far this ground-breaking decision will take us, I do not venture to predict. I fear, however, that it may pose a more serious threat to individual liberty than the Court realizes.

I respectfully dissent.

dissenting.

I join in the dissent by Justice Stevens and add these few observations.

The distinguishing feature of our criminal justice system is its insistence on principled, accountable decisionmaking in individual cases. If a person is to be seized, a satisfactory explanation for the invasive action ought to be established by an officer who exercises reasoned judgment under all the circumstances of the case. This principle can be accommodated even where officers must make immediate decisions to ensure their own safety.

Traffic stops, even for minor violations, can take upwards of 30 minutes. When an officer commands passengers innocent of any violation to leave the vehicle and stand by the side of the road in full view of the public, the seizure is serious, not trivial. As Justice Stevens concludes, the command to exit ought not to be given unless there are objective circumstances making it reasonable for the officer to issue the order. (We do not have before us the separate question whether passengers, who, after all, are in the car by choice, *423can be ordered to remain there for a reasonable time while the police conduct their business.)

The requisite showing for commanding passengers to exit need be no more than the existence of any circumstance justifying the order -in the interests of the officer’s safety or to facilitate a lawful search or investigation. As we have acknowledged for decades, special latitude is given to the police in effecting searches and seizures involving vehicles and their occupants. See, e. g., Chambers v. Maroney, 399 U. S. 42 (1970); New York v. Class, 475 U. S. 106 (1986); New York v. Belton, 453 U. S. 454 (1981). Just last Term we adhered to a rule permitting vehicle stops if there is some objective indication thát a violation has been committed, regardless of the officer’s real motives. See Whren v. United States, 517 U. S. 806 (1996). We could discern no other, workable rule. Even so, we insisted on a reasoned explanation for the stop.

The practical effect of our holding in Whren, of course, is to allow the police to stop vehicles in almost countless circumstances. When Whren is coupled with today’s holding, the Court puts tens of millions of passengers at risk of arbitrary control by the police. If the command to exit were to become commonplace, the Constitution would be diminished in a most public way. As the standards suggested in dissent are adequate to protect the safety of the police, we ought not to suffer so great a loss.

Since a myriad of circumstances will give a cautious officer reasonable grounds for commanding passengers to leave the vehicle, it might be thought the rule the Court adopts today will be little different in its operation than the rule offered in dissent. It does no disservice to police officers, however, to insist upon exercise of reasoned judgment. Adherence to neutral principles is the very premise of the rule of law the police themselves defend with such courage and dedication.

Most officers, it might be said, will exercise their new power with discretion and restraint; and no doubt this often *424will be the case. It might also be said that if some jurisdictions use today’s ruling to require passengers to exit as a matter of routine in every stop, citizen complaints and political intervention will call for an end to the practice. These arguments, however, would miss the point. Liberty comes not from officials by grace but from the Constitution by right.

For these reasons, and with all respect for the opinion of the Court, I dissent.