(delivered the opinion of the Court)
We consider whether a law enforcement official can, consistent with the Fourth Amendment , attempt to stop a fleeing motorist from continuing his public-endangering flight by ramming the motorist’s car from behind. Put another way: Can an officer take actions that place a fleeing motorist at risk of serious injury or death in order to stop the motorist’s flight from endangering the lives of innocent bystanders?
In March 2001, a Georgia county deputy clocked respondent’s vehicle travelling at 73 miles per hour on a road with a 55-mile-per-hour speed limit. The deputy activated his blue flashing lights indicating that respondent should pull over. Instead, respondent sped away, initiating a chase down what is in most portions a two-lane road, at speeds exceeding 85 miles per hour. The deputy radioed his dispatch to report that he was pursuing a fleeing vehicle, and broadcast its license plate number. Petitioner, Deputy Timothy Scott, heard the radio communication and joined the pursuit along with other officers. In the midst of the chase, respondent pulled into the parking lot of a shopping centre and was nearly boxed in by the various police vehicles. Respondent evaded the trap by making a sharp turn, colliding with Scott’s police car, exiting the parking lot, and speeding off once again down a two-lane highway.
Following respondent’s shopping centre manoeuvring, which resulted in slight damage to Scott’s police car, Scott took over as the lead pursuit vehicle. Six minutes and nearly 10 miles after the chase had begun, Scott decided to attempt to terminate the episode by employing a “Precision Intervention Technique (‘PIT’) manoeuvre, which causes the fleeing vehicle to spin to a stop.” Brief for Petitioner 4. Having radioed his supervisor for permission, Scott was told to “ ‘[g]o ahead and take him out.’ ” Harris v Coweta County, 433 F. 3d 807, 811 (CA11 2005). Instead, Scott applied his push bumper to the rear of respondent’s vehicle. As a result, respondent lost control of his vehicle, which left the roadway, ran down an embankment, overturned, and crashed. Respondent was badly injured and was rendered a quadriplegic.
Respondent filed suit against Deputy Scott and others under Rev. Stat. §1979, 42 U. S. C. §1983, alleging, inter alia, a violation of his federal constitutional rights, viz. use of excessive force resulting in an unreasonable seizure under the Fourth Amendment. In response, Scott filed a motion for summary judgment based on an assertion of qualified immunity. The District Court denied the motion, finding that “there are material issues of fact on which the issue of qualified immunity turns which present sufficient disagreement to require submission to a jury.” Harris v Coweta County, No. 3:01–CV–148–WBH (ND Ga., Sept. 23, 2003), App. to Pet. for Cert. 41a–42a. On interlocutory appeal, the United States Court of Appeals for the Eleventh Circuit affirmed the District Court’s decision to allow respondent’s Fourth Amendment claim against Scott to proceed to trial. Taking respondent’s view of the facts as given, the Court of Appeals concluded that Scott’s actions could constitute “deadly force” under Tennessee v Garner, 471 U. S. 1 (1985) , and that the use of such force in this context “would violate [respondent’s] constitutional right to be free from excessive force during a seizure. Accordingly, a reasonable jury could find that Scott violated [respondent’s] Fourth Amendment rights.” 433 F. 3d, at 816. The Court of Appeals further concluded that “the law as it existed [at the time of the incident], was sufficiently clear to give reasonable law enforcement officers ‘fair notice’ that ramming a vehicle under these circumstances was unlawful.” Id., at 817. The Court of Appeals thus concluded that Scott was not entitled to qualified immunity. We granted certiorari, 549 U. S. __ (2006), and now reverse.
In resolving questions of qualified immunity, courts are required to resolve a “threshold question: Taken in the light most favourable to the party asserting the injury, do the facts alleged show the officer’s conduct violated a constitutional right? This must be the initial inquiry.” Saucier v Katz, 533 U. S. 194, 201 (2001). If, and only if, the court finds a violation of a constitutional right, “the next, sequential step is to ask whether the right was clearly established .... in light of the specific context of the case.” Ibid. Although this ordering contradicts “[o]ur policy of avoiding unnecessary adjudication of constitutional issues,” United States v Treasury Employees, 513 U. S. 454, 478 (1995) (citing Ashwander v TVA, 297 U. S. 288, 346–347 (1936) (Brandeis, J., concurring)), we have said that such a departure from practice is “necessary to set forth principles which will become the basis for a [future] holding that a right is clearly established.” Saucier, supra, at 201. We therefore turn to the threshold inquiry: whether Deputy Scott’s actions violated the Fourth Amendment.
The first step in assessing the constitutionality of Scott’s actions is to determine the relevant facts. As this case was decided on summary judgment, there have not yet been factual findings by a judge or jury, and respondent’s version of events (unsurprisingly) differs substantially from Scott’s version. When things are in such a posture, courts are required to view the facts and draw reasonable inferences “in the light most favourable to the party opposing the [summary judgment] motion.” United States v Diebold, Inc., 369 U. S. 654, 655 (1962) (per curiam); Saucier, supra, at 201. In qualified immunity cases, this usually means adopting (as the Court of Appeals did here) the plaintiff’s version of the facts.
There is, however, an added wrinkle in this case: existence in the record of a videotape capturing the events in question. There are no allegations or indications that this videotape was doctored or altered in any way, nor any contention that what it depicts differs from what actually happened. The videotape quite clearly contradicts the version of the story told by respondent and adopted by the Court of Appeals. For example, the Court of Appeals adopted respondent’s assertions that, during the chase, “there was little, if any, actual threat to pedestrians or other motorists, as the roads were mostly empty and [respondent] remained in control of his vehicle.” 433 F. 3d, at 815. Indeed, reading the lower court’s opinion, one gets the impression that respondent, rather than fleeing from police, was attempting to pass his driving test, id., at 815–816:
[T]aking the facts from the non-movant’s viewpoint, [respondent] remained in control of his vehicle, slowed for turns and intersections, and typically used his indicators for turns. He did not run any motorists off the road. Nor was he a threat to pedestrians in the shopping centre parking lot, which was free from pedestrian and vehicular traffic as the centre was closed. Significantly, by the time the parties were back on the highway and Scott rammed [respondent], the motorway had been cleared of motorists and pedestrians allegedly because of police blockades of the nearby intersections.
The videotape tells quite a different story. There we see respondent’s vehicle racing down narrow, two-lane roads in the dead of night at speeds that are shockingly fast. We see it swerve around more than a dozen other cars, cross the double-yellow line, and force cars travelling in both directions to their respective shoulders to avoid being hit. We see it run multiple red lights and travel for considerable periods of time in the occasional centre left-turn-only lane, chased by numerous police cars forced to engage in the same hazardous manoeuvres just to keep up. Far from being the cautious and controlled driver the lower court depicts, what we see on the video more closely resembles a Hollywood-style car chase of the most frightening sort, placing police officers and innocent bystanders alike at great risk of serious injury.
At the summary judgment stage, facts must be viewed in the light most favourable to the non-moving party only if there is a “genuine” dispute as to those facts. Fed. Rule Civ. Proc. 56(c). As we have emphasized, “[w]hen the moving party has carried its burden under Rule 56(c), its opponent must do more than simply show that there is some metaphysical doubt as to the material facts.... Where the record taken as a whole could not lead a rational trier of fact to find for the non-moving party, there is no ‘genuine issue for trial.’ ” Matsushita Elec. Industrial Co. v Zenith Radio Corp., 475 U. S. 574, 586–587 (1986) (footnote omitted). “[T]he mere existence of some alleged factual dispute between the parties will not defeat an otherwise properly supported motion for summary judgment; the requirement is that there be no genuine issue of material fact.” Anderson v Liberty Lobby, Inc., 477 U. S. 242, 247–248 (1986). When opposing parties tell two different stories, one of which is blatantly contradicted by the record, so that no reasonable jury could believe it, a court should not adopt that version of the facts for purposes of ruling on a motion for summary judgment.
That was the case here with regard to the factual issue whether respondent was driving in such fashion as to endanger human life. Respondent’s version of events is so utterly discredited by the record that no reasonable jury could have believed him. The Court of Appeals should not have relied on such visible fiction; it should have viewed the facts in the light depicted by the videotape.
Judging the matter on that basis, we think it is quite clear that Deputy Scott did not violate the Fourth Amendment. Scott does not contest that his decision to terminate the car chase by ramming his bumper into respondent’s vehicle constituted a “seizure.” “[A] Fourth Amendment seizure [occurs] .... when there is a governmental termination of freedom of movement through means intentionally applied.” Brower v County of Inyo, 489 U. S. 593, 596–597 (1989) (emphasis deleted). See also id., at 597 (“If .... the police cruiser had pulled alongside the fleeing car and sideswiped it, producing the crash, then the termination of the suspect’s freedom of movement would have been a seizure”). It is also conceded, by both sides, that a claim of “excessive force in the course of making [a] ....‘seizure’ of [the] person .... [is] properly analyzed under the Fourth Amendment’s ‘objective reasonableness’ standard.” Graham v Connor, 490 U. S. 386, 388 (1989). The question we need to answer is whether Scott’s actions were objectively reasonable.
Respondent urges us to analyze this case as we analyzed Garner, 471 U. S. 1. See Brief for Respondent 16–29. We must first decide, he says, whether the actions Scott took constituted “deadly force.” (He defines “deadly force” as “any use of force which creates a substantial likelihood of causing death or serious bodily injury,” id., at 19.) If so, respondent claims that Garner prescribes certain preconditions that must be met before Scott’s actions can survive Fourth Amendment scrutiny:
The suspect must have posed an immediate threat of serious physical harm to the officer or others;
deadly force must have been necessary to prevent escape; and
where feasible, the officer must have given the suspect some warning. See Brief for Respondent 17–18 (citing Garner, supra, at 9–12).
Since these Garner preconditions for using deadly force were not met in this case, Scott’s actions were per se unreasonable.
Respondent’s argument falters at its first step; Garner did not establish a magical on/off switch that triggers rigid preconditions whenever an officer’s actions constitute “deadly force.” Garner was simply an application of the Fourth Amendment’s “reasonableness” test, Graham, supra, at 388, to the use of a particular type of force in a particular situation. Garner held that it was unreasonable to kill a “young, slight, and unarmed” burglary suspect, 471 U. S., at 21, by shooting him “in the back of the head” while he was running away on foot, id., at 4, and when the officer “could not reasonably have believed that [the suspect] .... posed any threat,” and “never attempted to justify his actions on any basis other than the need to prevent an escape,” id., at 21. Whatever Garner said about the factors that might have justified shooting the suspect in that case, such “preconditions” have scant applicability to this case, which has vastly different facts. “Garner had nothing to do with one car striking another or even with car chases in general .... A police car’s bumping a fleeing car is, in fact, not much like a policeman’s shooting a gun so as to hit a person.” Adams v St. Lucie County Sheriff’s Dept., 962 F. 2d 1563, 1577 (CA11 1992) (Edmondson, J., dissenting), adopted by 998 F. 2d 923 (CA11 1993) (en banc) (per curiam). Nor is the threat posed by the flight on foot of an unarmed suspect even remotely comparable to the extreme danger to human life posed by respondent in this case. Although respondent’s attempt to craft an easy-to-apply legal test in the Fourth Amendment context is admirable, in the end we must still slosh our way through the fact bound morass of “reasonableness.” Whether or not Scott’s actions constituted application of “deadly force,” all that matters is whether Scott’s actions were reasonable.
In determining the reasonableness of the manner in which a seizure is effected, “[w]e must balance the nature and quality of the intrusion on the individual’s Fourth Amendment interests against the importance of the governmental interests alleged to justify the intrusion.” United States v Place, 462 U. S. 696, 703 (1983). Scott defends his actions by pointing to the paramount governmental interest in ensuring public safety, and respondent nowhere suggests this was not the purpose motivating Scott’s behaviour. Thus, in judging whether Scott’s actions were reasonable, we must consider the risk of bodily harm that Scott’s actions posed to respondent in light of the threat to the public that Scott was trying to eliminate. Although there is no obvious way to quantify the risks on either side, it is clear from the videotape that respondent posed an actual and imminent threat to the lives of any pedestrians who might have been present, to other civilian motorists, and to the officers involved in the chase. See Part III–A, supra. It is equally clear that Scott’s actions posed a high likelihood of serious injury or death to respondent – though not the near certainty of death posed by, say, shooting a fleeing felon in the back of the head, see Garner, supra, at 4, or pulling alongside a fleeing motorist’s car and shooting the motorist, cf. Vaughan v Cox, 343 F. 3d 1323, 1326–1327 (CA11 2003). So how does a court go about weighing the perhaps lesser probability of injuring or killing numerous bystanders against the perhaps larger probability of injuring or killing a single person? We think it appropriate in this process to take into account not only the number of lives at risk, but also their relative culpability. It was respondent, after all, who intentionally placed himself and the public in danger by unlawfully engaging in the reckless, high-speed flight that ultimately produced the choice between two evils that Scott confronted. Multiple police cars, with blue lights flashing and sirens blaring, had been chasing respondent for nearly 10 miles, but he ignored their warning to stop. By contrast, those who might have been harmed had Scott not taken the action he did were entirely innocent. We have little difficulty in concluding it was reasonable for Scott to take the action that he did.
But wait, says respondent: Couldn’t the innocent public equally have been protected, and the tragic accident entirely avoided, if the police had simply ceased their pursuit? We think the police need not have taken that chance and hoped for the best. Whereas Scott’s action – ramming respondent off the road – was certain to eliminate the risk that respondent posed to the public, ceasing pursuit was not. First of all, there would have been no way to convey convincingly to respondent that the chase was off, and that he was free to go. Had respondent looked in his rear-view mirror and seen the police cars deactivate their flashing lights and turn around, he would have had no idea whether they were truly letting him get away, or simply devising a new strategy for capture. Perhaps the police knew a shortcut he didn’t know, and would reappear down the road to intercept him; or perhaps they were setting up a roadblock in his path. Cf. Brower, 489 U. S., at 594. Given such uncertainty, respondent might have been just as likely to respond by continuing to drive recklessly as by slowing down and wiping his brow.
Second, we are loath to lay down a rule requiring the police to allow fleeing suspects to get away whenever they drive so recklessly that they put other people’s lives in danger. It is obvious the perverse incentives such a rule would create: Every fleeing motorist would know that escape is within his grasp, if only he accelerates to 90 miles per hour, crosses the double-yellow line a few times, and runs a few red lights. The Constitution assuredly does not impose this invitation to impunity-earned-by-recklessness. Instead, we lay down a more sensible rule: A police officer’s attempt to terminate a dangerous high-speed car chase that threatens the lives of innocent bystanders does not violate the Fourth Amendment , even when it places the fleeing motorist at risk of serious injury or death.
* * *
The car chase that respondent initiated in this case posed a substantial and immediate risk of serious physical injury to others; no reasonable jury could conclude otherwise. Scott’s attempt to terminate the chase by forcing respondent off the road was reasonable, and Scott is entitled to summary judgment. The Court of Appeals’ decision to the contrary is reversed.
It is so ordered.
I join the Court’s opinion and would underscore two points. First, I do not read today’s decision as articulating a mechanical, per se rule. Cf. post, at 3 (Breyer, J., concurring). The inquiry described by the Court, ante, at 10–13, is situation specific. Among relevant considerations: Were the lives and well-being of others (motorists, pedestrians, police officers) at risk? Was there a safer way, given the time, place, and circumstances, to stop the fleeing vehicle? “[A]dmirable” as “[an] attempt to craft an easy-to-apply legal test in the Fourth Amendment context [may be],” the Court explains, “in the end we must still slosh our way through the fact bound morass of ‘reasonableness.’ ” Ante, at 10.
Second, were this case suitable for resolution on qualified immunity grounds, without reaching the constitutional question, Justice Breyer’s discussion would be engaging. See post, at 1–3 (urging the Court to overrule Saucier v Katz, 533 U. S. 194 (2001) ). In joining the Court’s opinion, however, Justice Breyer apparently shares the view that, in the appeal before us, the constitutional question warrants an answer. The video footage of the car chase, he agrees, demonstrates that the officer’s conduct did not transgress Fourth Amendment limitations. See post, at 1. Confronting Saucier, therefore, is properly reserved for another day and case. See ante, at 4, n. 4.
I join the Court’s opinion with one suggestion and two qualifications. Because watching the video footage of the car chase made a difference to my own view of the case, I suggest that the interested reader take advantage of the link in the Court’s opinion, ante, at 5, n. 5, and watch it. Having done so, I do not believe a reasonable jury could, in this instance, find that Officer Timothy Scott (who joined the chase late in the day and did not know the specific reason why the respondent was being pursued) acted in violation of the Constitution.
Second, the video makes clear the highly fact-dependent nature of this constitutional determination. And that fact-dependency supports the argument that we should overrule the requirement, announced in Saucier v Katz, 533 U. S. 194 (2001) , that lower courts must first decide the “constitutional question” before they turn to the “qualified immunity question.” See id., at 200 (“[T]he first inquiry must be whether a constitutional right would have been violated on the facts alleged”). Instead, lower courts should be free to decide the two questions in whatever order makes sense in the context of a particular case. Although I do not object to our deciding the constitutional question in this particular case, I believe that in order to lift the burden from lower courts we can and should reconsider Saucier’s requirement as well.
Sometimes (e.g., where a defendant is clearly entitled to qualified immunity) Saucier’s fixed order-of-battle rule wastes judicial resources in that it may require courts to answer a difficult constitutional question unnecessarily. Sometimes (e.g., where the defendant loses the constitutional question but wins on qualified immunity) that order-of-battle rule may immunize an incorrect constitutional ruling from review. Sometimes, as here, the order-of-battle rule will spawn constitutional rulings in areas of law so fact dependent that the result will be confusion rather than clarity. And frequently the order-of-battle rule violates that older, wiser judicial counsel “not to pass on questions of constitutionality .... unless such adjudication is unavoidable.” Spector Motor Service, Inc. v McLaughlin, 323 U. S. 101, 105 (1944); see Ashwander v TVA, 297 U. S. 288, 347 (1936) (Brandeis, J., concurring) (“The Court will not pass upon a constitutional question although properly presented by the record, if there is also present some other ground upon which the case may be disposed of ”). In a sharp departure from this counsel, Saucier requires courts to embrace unnecessary constitutional questions not to avoid them.
It is not surprising that commentators, judges, and, in this case, 28 States in an amicus brief, have invited us to reconsider Saucier’s requirement. See Leval, Judging Under the Constitution: Dicta About Dicta, 81 N. Y. U. L. Rev. 1249, 1275 (2006) (calling the requirement “a puzzling misadventure in constitutional dictum”); Dirrane v Brookline Police Dept., 315 F. 3d 65, 69–70 (CA1 2002) (referring to the requirement as “an uncomfortable exercise” when “the answer whether there was a violation may depend on a kaleidoscope of facts not yet fully developed”); Lyons v Xenia, 417 F. 3d 565, 580–584 (CA6 2005) (Sutton, J., concurring); Brief for State of Illinois et al. as Amici Curiae. I would accept that invitation.
While this Court should generally be reluctant to overturn precedents, stare decisis concerns are at their weakest here. See, e.g., Payne v Tennessee, 501 U. S. 808, 828 (1991) (“Considerations in favour of stare decisis” are at their weakest in cases “involving procedural and evidentiary rules”). The order-of-battle rule is relatively novel, it primarily affects judges, and there has been little reliance upon it.
Third, I disagree with the Court insofar as it articulates a per se rule. The majority states: “A police officer’s attempt to terminate a dangerous high-speed car chase that threatens the lives of innocent bystanders does not violate the Fourth Amendment , even when it places the fleeing motorist at risk of serious injury or death.” Ante, at 13. This statement is too absolute. As Justice Ginsburg points out, ante, at 1, whether a high-speed chase violates the Fourth Amendment may well depend upon more circumstances than the majority’s rule reflects. With these qualifications, I join the Court’s opinion.
Today, the Court asks whether an officer may “take actions that place a fleeing motorist at risk of serious injury or death in order to stop the motorist’s flight from endangering the lives of innocent bystanders.” Ante, at 1. Depending on the circumstances, the answer may be an obvious “yes,” an obvious “no,” or sufficiently doubtful that the question of the reasonableness of the officer’s actions should be decided by a jury, after a review of the degree of danger and the alternatives available to the officer. A high speed chase in a desert in Nevada is, after all, quite different from one that travels through the heart of Las Vegas.
Relying on a de novo review of a videotape of a portion of a night-time chase on a lightly travelled road in Georgia where no pedestrians or other “bystanders” were present, buttressed by uninformed speculation about the possible consequences of discontinuing the chase, eight of the jurors on this Court reach a verdict that differs from the views of the judges on both the District Court and the Court of Appeals who are surely more familiar with the hazards of driving on Georgia roads than we are. The Court’s justification for this unprecedented departure from our well-settled standard of review of factual determinations made by a district court and affirmed by a court of appeals is based on its mistaken view that the Court of Appeals’ description of the facts was “blatantly contradicted by the record” and that respondent’s version of the events was “so utterly discredited by the record that no reasonable jury could have believed him.” Ante, at 7–8.
Rather than supporting the conclusion that what we see on the video “resembles a Hollywood-style car chase of the most frightening sort,” ante, at 7, the tape actually confirms, rather than contradicts, the lower courts’ appraisal of the factual questions at issue. More important, it surely does not provide a principled basis for depriving the respondent of his right to have a jury evaluate the question whether the police officers’ decision to use deadly force to bring the chase to an end was reasonable.
Omitted from the Court’s description of the initial speeding violation is the fact that respondent was on a four-lane portion of Highway 34 when the officer clocked his speed at 73 miles per hour and initiated the chase. More significant – and contrary to the Court’s assumption that respondent’s vehicle “force[d] cars travelling in both directions to their respective shoulders to avoid being hit” ante, at 6 – a fact unmentioned in the text of the opinion explains why those cars pulled over prior to being passed by respondent. The sirens and flashing lights on the police cars following respondent gave the same warning that a speeding ambulance or fire engine would have provided. The 13 cars that respondent passed on his side of the road before entering the shopping centre, and both of the cars that he passed on the right after leaving the centre, no doubt had already pulled to the side of the road or were driving along the shoulder because they heard the police sirens or saw the flashing lights before respondent or the police cruisers approached. A jury could certainly conclude that those motorists were exposed to no greater risk than persons who take the same action in response to a speeding ambulance, and that their reactions were fully consistent with the evidence that respondent, though speeding, retained full control of his vehicle.
The police sirens also minimized any risk that may have arisen from running “multiple red lights,” ibid. In fact, respondent and his pursuers went through only two intersections with stop lights and in both cases all other vehicles in sight were stationary, presumably because they had been warned of the approaching speeders. Incidentally, the videos do show that the lights were red when the police cars passed through them but, because the cameras were farther away when respondent did so and it is difficult to discern the colour of the signal at that point, it is not entirely clear that he ran either or both of the red lights. In any event, the risk of harm to the stationary vehicles was minimized by the sirens, and there is no reason to believe that respondent would have disobeyed the signals if he were not being pursued.
My colleagues on the jury saw respondent “swerve around more than a dozen other cars,” and “force cars travelling in both directions to their respective shoulders,” ante, at 6, but they apparently discounted the possibility that those cars were already out of the pursuit’s path as a result of hearing the sirens. Even if that were not so, passing a slower vehicle on a two-lane road always involves some degree of swerving and is not especially dangerous if there are no cars coming from the opposite direction. At no point during the chase did respondent pull into the opposite lane other than to pass a car in front of him; he did the latter no more than five times and, on most of those occasions, used his turn signal. On none of these occasions was there a car travelling in the opposite direction. In fact, at one point, when respondent found himself behind a car in his own lane and there were cars travelling in the other direction, he slowed and waited for the cars travelling in the other direction to pass before overtaking the car in front of him while using his turn signal to do so. This is hardly the stuff of Hollywood. To the contrary, the video does not reveal any incidents that could even be remotely characterized as “close calls.”
In sum, the factual statements by the Court of Appeals quoted by the Court, ante, at 5–6, were entirely accurate. That court did not describe respondent as a “cautious” driver as my colleagues imply, ante, at 7, but it did correctly conclude that there is no evidence that he ever lost control of his vehicle. That court also correctly pointed out that the incident in the shopping centre parking lot did not create any risk to pedestrians or other vehicles because the chase occurred just before 11 p.m. on a weekday night and the centre was closed. It is apparent from the record (including the videotape) that local police had blocked off intersections to keep respondent from entering residential neighbourhoods and possibly endangering other motorists. I would add that the videos also show that no pedestrians, parked cars, sidewalks, or residences were visible at any time during the chase. The only “innocent bystanders” who were placed “at great risk of serious injury,” ante, at 7, were the drivers who either pulled off the road in response to the sirens or passed respondent in the opposite direction when he was driving on his side of the road.
I recognize, of course, that even though respondent’s original speeding violation on a four-lane highway was rather ordinary, his refusal to stop and subsequent flight was a serious offense that merited severe punishment. It was not, however, a capital offense, or even an offense that justified the use of deadly force rather than an abandonment of the chase. The Court’s concern about the “imminent threat to the lives of any pedestrians who might have been present,” ante, at 11, while surely valid in an appropriate case, should be discounted in a case involving a night-time chase in an area where no pedestrians were present.
What would have happened if the police had decided to abandon the chase? We now know that they could have apprehended respondent later because they had his license plate number. Even if that were not true, and even if he would have escaped any punishment at all, the use of deadly force in this case was no more appropriate than the use of a deadly weapon against a fleeing felon in Tennessee v Garner, 471 U. S. 1 (1985). In any event, any uncertainty about the result of abandoning the pursuit has not prevented the Court from basing its conclusions on its own factual assumptions. The Court attempts to avoid the conclusion that deadly force was unnecessary by speculating that if the officers had let him go, respondent might have been “just as likely” to continue to drive recklessly as to slow down and wipe his brow. Ante, at 12. That speculation is unconvincing as a matter of common sense and improper as a matter of law. Our duty to view the evidence in the light most favourable to the non-moving party would foreclose such speculation if the Court had not used its observation of the video as an excuse for replacing the rule of law with its ad hoc judgment. There is no evidentiary basis for an assumption that dangers caused by flight from a police pursuit will continue after the pursuit ends. Indeed, rules adopted by countless police departments throughout the country are based on a judgment that differs from the Court’s. See, e.g., App. to Brief for Georgia Association of Chiefs of Police, Inc., as Amicus Curiae A–52 (“During a pursuit, the need to apprehend the suspect should always outweigh the level of danger created by the pursuit. When the immediate danger to the public created by the pursuit is greater than the immediate or potential danger to the public should the suspect remain at large, then the pursuit should be discontinued or terminated.... [P]ursuits should usually be discontinued when the violator’s identity has been established to the point that later apprehension can be accomplished without danger to the public”).
Although Garner may not, as the Court suggests, “establish a magical on/off switch that triggers rigid preconditions” for the use of deadly force, ante, at 9, it did set a threshold under which the use of deadly force would be considered constitutionally unreasonable, 471 U. S., at 11–12:
Where the officer has probable cause to believe that the suspect poses a threat of serious physical harm, either to the officer or to others, it is not constitutionally unreasonable to prevent escape by using deadly force. Thus, if the suspect threatens the officer with a weapon or there is probable cause to believe that he has committed a crime involving the infliction or threatened infliction of serious physical harm, deadly force may be used if necessary to prevent escape, and if, where feasible, some warning has been given.
Whether a person’s actions have risen to a level warranting deadly force is a question of fact best reserved for a jury. Here, the Court has usurped the jury’s fact-finding function and, in doing so, implicitly labelled the four other judges to review the case unreasonable. It chastises the Court of Appeals for failing to “vie[w] the facts in the light depicted by the videotape” and implies that no reasonable person could view the videotape and come to the conclusion that deadly force was unjustified. Ante, at 8. However, the three judges on the Court of Appeals panel apparently did view the videotapes entered into evidence and described a very different version of events:
At the time of the ramming, apart from speeding and running two red lights, Harris was driving in a non-aggressive fashion (i.e., without trying to ram or run into the officers). Moreover, .... Scott’s path on the open highway was largely clear. The videos introduced into evidence show little to no vehicular (or pedestrian) traffic, allegedly because of the late hour and the police blockade of the nearby intersections. Finally, Scott issued absolutely no warning (e.g., over the loudspeaker or otherwise) prior to using deadly force.
Harris v Coweta County, 433 F. 3d 807, 819, n. 14 (CA11 2005)
If two groups of judges can disagree so vehemently about the nature of the pursuit and the circumstances surrounding that pursuit, it seems eminently likely that a reasonable juror could disagree with this Court’s characterization of events. Moreover, under the standard set forth in Garner, it is certainly possible that “a jury could conclude that Scott unreasonably used deadly force to seize Harris by ramming him off the road under the instant circumstances.” 433 F. 3d, at 821.
The Court today sets forth a per se rule that presumes its own version of the facts: “A police officer’s attempt to terminate a dangerous high-speed car chase that threatens the lives of innocent bystanders does not violate the Fourth Amendment , even when it places the fleeing motorist at risk of serious injury or death.” Ante, at 13 (emphasis added). Not only does that rule fly in the face of the flexible and case-by-case “reasonableness” approach applied in Garner and Graham v Connor, 490 U. S. 386 (1989) , but it is also arguably inapplicable to the case at hand, given that it is not clear that this chase threatened the life of any “innocent bystande[r].” In my view, the risks inherent in justifying unwarranted police conduct on the basis of unfounded assumptions are unacceptable, particularly when less drastic measures – in this case, the use of stop sticks or a simple warning issued from a loudspeaker – could have avoided such a tragic result. In my judgment, jurors in Georgia should be allowed to evaluate the reasonableness of the decision to ram respondent’s speeding vehicle in a manner that created an obvious risk of death and has in fact made him a quadriplegic at the age of 19.
I respectfully dissent.
 Scott says he decided not to employ the PIT manoeuvre because he was “concerned that the vehicles were moving too quickly to safely execute the manoeuvre.” Brief for Petitioner 4. Respondent agrees that the PIT manoeuvre could not have been safely employed. See Brief for Respondent 9. It is irrelevant to our analysis whether Scott had permission to take the precise actions he took.
 Qualified immunity is “an immunity from suit rather than a mere defence to liability; and like an absolute immunity, it is effectively lost if a case is erroneously permitted to go to trial.” Mitchell v. Forsyth, 472 U. S. 511, 526 (1985). Thus, we have held that an order denying qualified immunity is immediately appealable even though it is interlocutory; otherwise, it would be “effectively unreviewable.” Id., at 527. Further, “we repeatedly have stressed the importance of resolving immunity questions at the earliest possible stage in litigation.” Hunter v. Bryant, 502 U. S. 224, 227 (1991) (per curiam).
 None of the other claims respondent brought against Scott or any other party are before this Court.
 Prior to this Court’s announcement of Saucier’s “rigid ‘order of battle,’ ” Brosseau v. Haugen, 543 U. S. 194, 201–202 (2004) (Breyer, J., concurring), we had described this order of inquiry as the “better approach,” County of Sacramento v. Lewis, 523 U. S. 833 , n. 5 (1998), though not one that was required in all cases. See id., at 858–859 (Breyer, J., concurring); id., at 859 (Stevens, J., concurring in judgment). There has been doubt expressed regarding the wisdom of Saucier’s decision to make the threshold inquiry mandatory, especially in cases where the constitutional question is relatively difficult and the qualified immunity question relatively straightforward. See, e.g., Brosseau, supra, at 201 (Breyer, J., joined by Scalia and Ginsburg, JJ., concurring); Bunting v. Mellen, 541 U. S. 1019 (2004) (Stevens, J., joined by Ginsburg and Breyer, JJ., respecting denial of certiorari); id., at 1025 (Scalia, J., joined by Rehnquist, C.J., dissenting). See also Lyons v. Xenia, 417 F. 3d 565, 580–584 (CA6 2005) (Sutton, J., concurring). We need not address the wisdom of Saucier in this case, however, because the constitutional question with which we are presented is, as discussed in Part III–B, infra, easily decided. Deciding that question first is thus the “better approach,” Lewis, supra, at 841, n. 5, regardless of whether it is required.
 Justice Stevens suggests that our reaction to the videotape is somehow idiosyncratic, and seems to believe we are misrepresenting its contents. See post, at 4 (dissenting opinion) (“In sum, the factual statements by the Court of Appeals quoted by the Court .... were entirely accurate”). We are happy to allow the videotape to speak for itself. See Record 36, Exh. A, available at http://www.supremecourtus.gov/opinions/video/scott_v_harris.rmvb and in Clerk of Court’s case file.
 Justice Stevens hypothesizes that these cars “had already pulled to the side of the road or were driving along the shoulder because they heard the police sirens or saw the flashing lights,” so that “[a] jury could certainly conclude that those motorists were exposed to no greater risk than persons who take the same action in response to a speeding ambulance.” Post, at 3. It is not our experience that ambulances and fire engines careen down two-lane roads at 85-plus miles per hour, with an unmarked scout car out in front of them. The risk they pose to the public is vastly less than what respondent created here. But even if that were not so, it would in no way lead to the conclusion that it was unreasonable to eliminate the threat to life that respondent posed. Society accepts the risk of speeding ambulances and fire engines in order to save life and property; it need not (and assuredly does not) accept a similar risk posed by a reckless motorist fleeing the police.
 This is not to say that each and every factual statement made by the Court of Appeals is inaccurate. For example, the videotape validates the court’s statement that when Scott rammed respondent’s vehicle it was not threatening any other vehicles or pedestrians. (Undoubtedly Scott waited for the road to be clear before executing his manoeuvre.)
 Justice Stevens incorrectly declares this to be “a question of fact best reserved for a jury,” and complains we are “usurp[ing] the jury’s fact-finding function.” Post, at 7. At the summary judgment stage, however, once we have determined the relevant set of facts and drawn all inferences in favour of the non-moving party to the extent supportable by the record, see Part III–A, supra, the reasonableness of Scott’s actions – or, in Justice Stevens’ parlance, “[w]hether [respondent’s] actions have risen to a level warranting deadly force,” post, at 7 – is a pure question of law.
 Respondent, like the Court of Appeals, defines this second precondition as “ ‘necessary to prevent escape,’ ” Brief for Respondent 17; Harris v. Coweta County, 433 F. 3d 807, 813 (CA11 2005), quoting Garner, 471 U. S., at 11. But that quote from Garner is taken out of context. The necessity described in Garner was, in fact, the need to prevent “serious physical harm, either to the officer or to others.” Ibid. By way of example only, Garner hypothesized that deadly force may be used “if necessary to prevent escape” when the suspect is known to have “committed a crime involving the infliction or threatened infliction of serious physical harm,” ibid., so that his mere being at large poses an inherent danger to society. Respondent did not pose that type of inherent threat to society, since (prior to the car chase) he had committed only a minor traffic offense and, as far as the police were aware, had no prior criminal record. But in this case, unlike in Garner, it was respondent’s flight itself (by means of a speeding automobile) that posed the threat of “serious physical harm .... to others.” Ibid.
 The Court of Appeals cites Brower v. County of Inyo, 489 U. S. 593, 595 (1989) , for its refusal to “countenance the argument that by continuing to flee, a suspect absolves a pursuing police officer of any possible liability for all ensuing actions during the chase,” 433 F. 3d, at 816. The only question in Brower was whether a police roadblock constituted a seizure under the Fourth Amendment. In deciding that question, the relative culpability of the parties is, of course, irrelevant; a seizure occurs whenever the police are “responsib[le] for the termination of [a person’s] movement,” 433 F. 3d, at 816, regardless of the reason for the termination. Culpability is relevant, however, to the reasonableness of the seizure – to whether preventing possible harm to the innocent justifies exposing to possible harm the person threatening them.
 Contrary to Justice Stevens’ assertions, we do not “assum[e] that dangers caused by flight from a police pursuit will continue after the pursuit ends,” post, at 6, nor do we make any “factual assumptions,” post, at 5, with respect to what would have happened if the police had gone home. We simply point out the uncertainties regarding what would have happened, in response to respondent’s factual assumption that the high-speed flight would have ended.
 I can only conclude that my colleagues were unduly frightened by two or three images on the tape that looked like bursts of lightning or explosions, but were in fact merely the headlights of vehicles zooming by in the opposite lane. Had they learned to drive when most high-speed driving took place on two-lane roads rather than on superhighways – when split-second judgments about the risk of passing a slow-poke in the face of oncoming traffic were routine – they might well have reacted to the videotape more dispassionately.
 According to the District Court record, when respondent was clocked at 73 miles per hour, the deputy who recorded his speed was sitting in his patrol car on Highway 34 between Lora Smith Road and Sullivan Road in Coweta County, Georgia. At that point, as well as at the point at which Highway 34 intersects with Highway 154 – where the deputy caught up with respondent and the videotape begins – Highway 34 is a four-lane road, consisting of two lanes in each direction with a wide grass divider separating the flow of traffic.
 While still on the four-lane portion of Highway 34, the deputy who had clocked respondent’s speed turned on his blue light and siren in an attempt to get respondent to pull over. It was when the deputy turned on his blue light that the dash-mounted video camera was activated and began to record the pursuit.
 Although perhaps understandable, because their volume on the sound recording is low (possibly due to sound proofing in the officer’s vehicle), the Court appears to minimize the significance of the sirens audible throughout the tape recording of the pursuit.
 In noting that Scott’s action “was certain to eliminate the risk that respondent posed to the public” while “ceasing pursuit was not,” the Court prioritizes total elimination of the risk of harm to the public over the risk that respondent may be seriously injured or even killed. Ante, at 12 (emphasis in original). The Court is only able to make such a statement by assuming, based on its interpretation of events on the videotape, that the risk of harm posed in this case, and the type of harm involved, rose to a level warranting deadly force. These are the same types of questions that, when disputed, are typically resolved by a jury; this is why both the District Court and the Court of Appeals saw fit to have them be so decided. Although the Court claims only to have drawn factual inferences in respondent’s favour “to the extent supportable by the record,” ante, at 8, n. 8 (emphasis in original), its own view of the record has clearly precluded it from doing so to the same extent as the two courts through which this case has already travelled, see ante, at 2–3, 5–6.
 In its opinion, the Court of Appeals correctly noted: “We reject the defendants’ argument that Harris’ driving must, as a matter of law, be considered sufficiently reckless to give Scott probable cause to believe that he posed a substantial threat of imminent physical harm to motorists and pedestrians. This is a disputed issue to be resolved by a jury.” Harris v. Coweta County, 433 F. 3d 807, 815 (CA11 2005).
 In total, there are four police tapes which captured portions of the pursuit, all recorded from different officers’ vehicles.
 It is unclear whether, in referring to “innocent bystanders,” the Court is referring to the motorists driving unfazed in the opposite direction or to the drivers who pulled over to the side of the road, safely out of respondent’s and petitioner’s path.
 “Stop sticks” are a device which can be placed across the roadway and used to flatten a vehicle’s tires slowly to safely terminate a pursuit.
all rights reserved