with whom Justice Brennan, Justice Marshall, and Justice Blackmun join, dissenting.
Correct analysis of the Fourth Amendment issues raised by this case requires, first, a precise identification of the two constitutional violations that occurred, and, second, an explanation of why a remedy for both is appropriate. While I do not believe that the current record justifies suppression of the challenged evidence, neither does it justify affirmance of petitioners’ convictions. We must consider the substantial contention, supported by the findings of the District Court and left unaddressed by the opinion of this Court, that the authorities’ access to the evidence introduced against petitioners at trial was made possible only through exploitation of both constitutional violations. Because I believe that contention must be addressed before petitioners’ convictions are finally affirmed, I would remand for further proceedings. The Court’s disposition, I fear, will provide government agents with an affirmative incentive to engage in unconstitutional violations of the privacy of the home. The Court’s disposition is, therefore, inconsistent with a primary purpose of the Fourth Amendment’s exclusionary rule — to ensure that all private citizens — not just these petitioners — have some meaningful protection against future violations of their rights.
I
The events that occurred on February 12 and 13, 1981, were the culmination of an investigation of petitioners that had been under way for over two weeks. On the evening of February 12, agents of the New York Drug Enforcement Task Force arrested Rivudalla-Vidal and Parra, who told them that Segura probably had cocaine in his apartment. At that point, the agents concluded that they had probable cause to search petitioners’ apartment, and contacted the United States Attorney’s office. An Assistant United States Attorney informed the agents that at that hour, 6:30 p. m., it was too late to obtain a search warrant, and advised them instead to go to the apartment, arrest Segura, and “secure the *818premises” pending the issuance of a warrant.1 The agents arrived at the apartment about an hour later and positioned themselves on a fire escape, where they could observe anyone entering or leaving the apartment. They also put their ears to the door, but heard nothing.2 After three hours of waiting, the agents left their perch and went outside the building, where they continued waiting for Segura to show up. The District Court described what followed:
“Around 11:15 p. m. Segura appeared, and as he began to enter the locked door at the lobby, he was apprehended, and placed in handcuffs under arrest. The agents, led by Shea, informed him that they wanted to go upstairs to 3D, to which Segura replied that he did not live in the building or in that apartment. Forcibly bringing him to the third floor, the agents began down the hallway, at which point Segura again resisted. Shea again forced him down the hallway to the door of 3D, an *819apartment which is located in the rear of the building, with no view of the front of the building where the arrest took place. Shea knocked on the door of 3D, with Se-gura standing, handcuffed, in front of him. Luz Colon, unknown to Shea at the time as such, opened the door. Detective Shea, without more, walked into the apartment with Segura in custody. He was then followed by two other agents, and five minutes later, by Palumbo. Neither Shea nor any other agent had an arrest warrant, or a search warrant. Nor did any of the officers ask for or receive consent to enter apartment 3D.” App. 10-11.
The agents arrested Colon and three other persons found in the apartment. Colon was unknown to the agents at the time.3 The agents made a cursory search of the apartment and saw various items of narcotics paraphernalia in plain view.4 The agents left that evidence — the “prewarrant evidence” — in the apartment, but they took the arrestees to headquarters.
At least two of the agents spent the night in the apartment and remained in it thoughout the following day while their colleagues booked the arrestees and presumably persevered in their efforts to obtain a warrant to search the apartment. Finally, at 6 p. m. on February 13, the remaining agents were informed that a search warrant had just been issued, and at that point they conducted a thorough search. The District Court concluded: “There was thus a lapse of some 18-20 hours from the entry into the apartment to the execution of the search warrant, during which time the officers remained inside the apartment and in complete control of it.” Id., at 11. Upon searching the apartment the agents found one kilo of cocaine, over $50,000, several rounds of .38-caliber ammunition, and records of narcotics transactions.
*820II
The Court frames the appropriate inquiry in this case as whether the evidence obtained when the search warrant was executed was a “fruit” of illegal conduct. Ante, at 804. As a predicate to that inquiry, the illegal conduct must, of course, be identified.
The District Court found that no exigent circumstances justified the agents’ initial warrantless entry into petitioners’ apartment. App. 11-13. The Court of Appeals affirmed this finding, and the Government did not seek review of it by this Court. Thus, it is uncontested that the warrantless entry of petitioners’ apartment was unconstitutional.5 It is equally clear that the subsequent 18-20-hour occupation of the apartment was independently unconstitutional for two separate reasons.
First, the occupation was an unreasonable “search” within the meaning of the Fourth Amendment. A “search” for purposes of the Fourth Amendment occurs when a reasonable expectation of privacy is infringed.6 Nowhere are expectations of privacy greater than in the home. As the Court has repeatedly noted, “physical entry of the home is the chief evil against which the wording of the Fourth Amendment is directed.” United States v. United States District Court, 407 U. S. 297, 313 (1972).7 Of course, the invasion of privacy *821occasioned by a physical entry does not cease after the initial entry. In Mincey v. Arizona, 437 U. S. 385 (1978), we held that although the police lawfully entered Mineey’s home to arrest him, the Constitution forbade them to remain in the home and to search it. The Court reasoned that despite the lawful initial entry, Mincey retained a constitutionally protected privacy interest in his home that could not be infringed without a warrant. See id., at 390-391. Similarly, in Chimel v. California, 395 U. S. 752 (1969), we could “see no reason why, simply because some interference with an individual’s privacy and freedom of movement has lawfully taken place, further intrusions should automatically be allowed despite the absence of a warrant that the Fourth Amendment would otherwise require.” Id., at 766-767, n. 12.8 Here, by remaining in the home after the initial entry, the agents exacerbated the invasion of petitioners’ protected privacy interests. Even assuming the most innocent of motives, the agents’ occupation of petitioners’ living quarters inevitably involved scrutiny of a variety of personal effects throughout the apartment.9 Petitioners’ privacy interests were unreasonably infringed by the agents’ prolonged *822occupation of their home. The Chief Justice simply ignores this point, assuming that there is no constitutional distinction between surveillance of the home from the outside and physical occupation from the inside. The Chief Justice’s assumption is, of course, untenable; there is a fundamental difference when there is a
“breach of the entrance to an individual’s home. The Fourth Amendment protects the individual’s privacy in a variety of settings. In none is the zone of privacy more clearly defined than when bounded by the unambiguous physical dimensions of an individual’s home — a zone which finds its roots in clear and specific constitutional terms: ‘The right of the people to be secure in their . . . houses . . . shall not be violated.’” Payton v. New York, 445 U. S. 573, 589 (1980).
Second, the agents’ occupation was also an unreasonable “seizure” within the meaning of the Fourth Amendment. A “seizure” occurs when there is some meaningful interference with an individual’s possessory interests.10 There can be no doubt here that petitioners’ possessory interests with respect to their apartment were subject to meaningful governmental interference. The agents not only excluded petitioners from access to their own apartment, and thereby prevented them from exercising any possessory right at all to the apartment and its contents, but they also exercised complete dominion and control over the apartment and its contents.11 Our cases virtually compel the conclusion that the contents of the apart*823ment were seized. We have held that when the police take custody of a person, they concomitantly acquire lawful custody of his personal effects, see Illinois v. Lafayette, 462 U. S. 640, 648 (1983); United States v. Edwards, 415 U. S. 800 (1974); United States v. Robinson, 414 U. S. 218 (1973); and when they take custody of a car, they are also in lawful custody of its contents, see South Dakota v. Opperman, 428 U. S. 364 (1976). Surely it follows that when the authorities take custody of an apartment they also take custody of its contents.12
This seizure was constitutionally unreasonable. Even a seizure reasonable at its inception can become unreasonable because of its duration. United States v. Place, 462 U. S. 696, 709-710 (1983). Even if exigent circumstances justified the entry into and impoundment of the premises pending a warrant — and no one even argues that such circumstances existed — the duration of the seizure would nevertheless have been unreasonable. While exigent circumstances may justify police conduct that would otherwise be unreasonable if undertaken without a warrant, such conduct must be “strictly circumscribed by the exigencies which justify its initiation,” Terry v. Ohio, 392 U. S. 1, 25-26 (1968).13 The cases The Chief Justice cites, ante, at 807-810, for the proposition that the government may impound premises for the amount of time necessary to procure a warrant thus have no application to this case whatsoever.14 There is no conten*824tion that a period of 18-20 hours was even remotely necessary to procure a warrant. The contrast between the 90-minute duration of the seizure of a piece of luggage held unreasonable in Place and the 18-20-hour duration of the seizure of the apartment and its contents in this case graphically illustrates the unreasonable character of the agents’ conduct. Moreover, unlike Place, which involved a seizure lawful at its inception, this seizure was constitutionally unreasonable from the moment it began. It was conducted without a warrant and in the absence of exigent circumstances.15 It has been clear since at least Chimel v. California, 395 U. S. 752 (1969), that the police may neither search nor seize the contents of a home without a warrant.16 There is simply no basis for concluding that this 18-20-hour warrantless invasion of petitioners’ home complied with the Fourth Amendment. Because the agents unreasonably delayed in seeking judicial authorization for their seizure of petitioners’ apartment, that seizure was unreasonable.
*825Nevertheless, in what I can only characterize as an astonishing holding, The Chief Justice, joined by Justice O’Connor, concludes that the 18-20-hour seizure of the apartment was not unreasonable. He advances three reasons for that conclusion, none of which has any merit.
First, he seeks to justify the delay because “the officers focused first on the task of processing those whom they had arrested before turning to the task of securing the warrant.” Ante, at 812. But there is no evidence that this task presented any difficulties; indeed, since the arrest of the occupants itself was unconstitutional, it is truly ironic that The Chief Justice uses one wrong to justify another. Of greater significance, the District Court expressly found that the length of the delay was unreasonable and that the Government had made no attempt to justify it; that finding was upheld by the Court of Appeals, and in this Court the Government expressly concedes that the delay was unreasonable.17
Second, The Chief Justice suggests that it is relevant that the officers did not act in “bad faith.” Ante, at 798, 812. This is done despite the fact that there is no finding as to whether the agents acted in good or bad faith; the reason is that the litigants have never raised the issue. More impor*826tant, this Court has repeatedly held that a police officer’s good or bad faith in undertaking a search or seizure is irrelevant to its constitutional reasonableness,18 and does so again today.19
Finally, and “most important” to his conclusion, The Chief Justice suggests that there was no significant interference with petitioners’ possessory interests in their apartment because they were in custody anyway. Ante, at 813. The cases are legion holding that a citizen retains a protected possessory interest in his home and the effects within it which may not be infringed without a warrant even though that person is in custody. Mincey and Chimel are but two instances of that general rule — the defendants in both cases were in custody, yet both were held to have protected pos-sessory interests in their homes and the effects within them that could not be infringed without a warrant. Even when a person is in custody after an arrest based on probable cause, he still, of course, owns his house and his right to exclude others — including federal narcotics agents — remains inviolate. What is even more strange about The Chief Justice’s conclusion is that it permits the authorities to benefit from the fact that they had unlawfully arrested Colon. Colon was in her own home when she was arrested without a warrant. That was unconstitutional.20 If the agents had decided to obey the Constitution and not arrest Colon, then she would not have “relinquished control” over the property and presumably it would have been unreasonable for the agents to have remained on the premises under The Chief Justice’s analysis. However, because the agents conducted an unlawful arrest in addition to their pre*827vious unlawful entry, an otherwise unreasonable occupation becomes “reasonable.” The Chief Justice’s approach is as reasonable as was the agents’ conduct. Only in that sense does it achieve its purpose.
Thus, on the basis of the record evidence and the findings of the District Court, it is clear that the 18-20-hour occupation of petitioners’ apartment was a second independent violation of the Fourth Amendment. Not only was it the fruit of the initial illegal entry into that apartment, but it also constituted an unreasonable search and seizure of the apartment. The District Court concluded that both violations should be remedied by suppression of all of the evidence found in the apartment. The Court of Appeals agreed that suppression of the prewarrant evidence was the proper remedy for the first violation but prescribed no remedy for the second. The Chief Justice does not agree that there was a second violation, and the Court concludes that the unconstitutional conduct that did occur was neutralized by the ultimate issuance of a valid warrant. In reaching that conclusion the Court correctly recognizes that the law requires suppression of the evidence if it was ‘““come at by exploitation of [the initial] illegality” ’ ” instead of “ ‘ “by means sufficiently distinguishable to be purged of the primary taint.”’” Ante, at 804-805 (quoting Wong Sun v. United States, 371 U. S. 471, 488 (1963)). The Court fails, however, to discuss the reason for that rule or how it should apply to the facts of this case.
III
Every time a court holds that unconstitutionally obtained evidence may not be used in a criminal trial it is acutely aware of the social costs that such a holding entails.21 Only *828the most compelling reason could justify the repeated imposition of such costs on society. That reason, of course, is to prevent violations of the Constitution from occurring.22
As the Court has repeatedly stated, a principal purpose of the exclusionary rule is to deter violations of the Fourth Amendment. See, e. g., Stone v. Powell, 428 U. S. 465, 486 (1976); United States v. Janis, 428 U. S. 433, 446-447 (1976); United States v. Peltier, 422 U. S. 531, 536-539 (1975); United States v. Calandra, 414 U. S. 338, 347-348 (1974).
“The rule is calculated to prevent, not to repair. Its purpose is to deter — to compel respect for the constitu*829tional guaranty in the only effectively available way — by removing the incentive to disregard it.” Elkins v. United States, 364 U. S. 206, 217 (1960).
The deterrence rationale for the exclusionary rule sometimes, but not always, requires that it be applied to the indirect consequences of a constitutional violation. If the government could utilize evidence obtained through exploitation of illegal conduct, it would retain an incentive to engage in that conduct. “To forbid the direct use of methods thus characterized [as illegal] but to put no curb on their full indirect use would only invite the very methods deemed ‘inconsistent with ethical standards and destructive of personal liberty.’” Nardone v. United States, 308 U. S. 338, 340 (1939).
We have not, however, mechanically applied the rule to every item of evidence that has a causal connection with police misconduct. “The notion of the ‘dissipation of the taint’ attempts to mark the point at which the detrimental consequences of illegal police action become so attenuated that the deterrent effect of the exclusionary rule no longer justifies its cost.” Brown v. Illinois, 422 U. S. 590, 609 (1975) (Powell, J., concurring in part).23
This point is well illustrated by our cases concerning the use of confessions obtained as the result of unlawful arrests. In Wong Sun v. United States, 371 U. S. 471 (1963), we rejected a rule that any evidence that would not have been obtained but for the illegal actions of the police should be suppressed. See id., at 487-488, 491. Yet in Brown v. Illinois, 422 U. S. 590 (1975), while continuing to reject a “but-for” rule, see id., at 603, we held that the taint of an unlawful arrest could not be purged merely by warning the arrestee of his right to remain silent and to consult with *830counsel as required by Miranda v. Arizona, 384 U. S. 436 (1966). We explained:
“If Miranda warnings, by themselves, were held to attenuate the taint of an unconstitutional arrest, regardless of how wanton and purposeful the Fourth Amendment violation, the effect of the exclusionary rule would be substantially diluted. Arrests made without warrant or without probable cause, for questioning or ‘investigation,’ would be encouraged by the knowledge that evidence derived therefrom could well be made admissible at trial by the simple expedient of giving Miranda warnings. Any incentive to avoid Fourth Amendment violations would be eviscerated by making the warnings, in effect, a ‘cure-all,’ and the constitutional guarantee against unlawful searches and seizures could be said to be reduced to ‘a form of words.’ ” 422 U. S., at 602-603 (citation and footnote omitted).
These holdings make it clear that taint questions do not depend merely on questions of causation; causation is a necessary but not a sufficient condition for exclusion. In addition, it must be shown that exclusion is required to remove the incentive for the police to engage in the unlawful conduct. When it is, exclusion is mandated if the Fourth Amendment is to be more than “a form of words.”
C
The Court concludes that the evidence introduced against petitioners at trial was obtained from a source that was “independent” of the prior illegality — the search warrant. The Court explains that since the police had a legal basis for obtaining and executing the search warrant, the fruits of the authorized search were not produced by exploitation of the prior illegality. Ante, at 814-815. There are significant analytical difficulties lurking in the Court’s approach. First, the Court accepts the distinction between the evidence *831obtained pursuant to the warrant and the evidence obtained during the initial illegal entry. Ante, at 814; see also ante, at 812 (opinion of Burger, C. J.). I would not draw a distinction between the prewarrant evidence and the post-warrant evidence. The warrant embraced both categories equally and if there had been no unlawful entry, there is no more reason to believe that the evidence in plain view would have remained in the apartment and would have been obtained when the warrant was executed than the evidence that was concealed. The warrant provided an “independent” justification for seizing all the evidence in the apartment— that in plain view just as much as the items that were concealed. The “plain view” items were not actually removed from the apartment until the warrant was executed;24 thus there was no more interference with petitioners’ possessory interest in those items than with their interest in the concealed items. If the execution of a valid warrant takes the poison out of the hidden fruit, I should think that it would also remove the taint from the fruit in plain view.25
Second, the Court’s holding is inadequate to resolve the claims raised by petitioners. The Court states that the fruits of the judicially authorized search were untainted because “[n]o information obtained during the initial entry or occupation of the apartment was needed or used by the agents to secure the warrant.” Ante, at 814. That is sufficient to dispose only of a claim that petitioners do not make— that the information which led to the issuance of the search warrant was tainted. It does not dispose of the claim that *832petitioners do make — that the agents’ access to the fruits of the authorized search, rather than the information which led to that search, was a product of illegal conduct. On this question, the length of the delay in obtaining the warrant is surely relevant.
If Segura had not returned home at all that night, or during the next day, it is probable that the occupants of the apartment would have become concerned and might at least have destroyed the records of their illegal transactions, or removed some of the evidence. If one of the occupants had left the apartment and taken evidence with him or her during the 18-20-hour period prior to the execution of the search warrant, then obviously that evidence would not have been accessible to the agents when the warrant finally was executed.26 The District Court concluded that there was a possibility that the evidence’s availability when the warrant was executed hinged solely on the illegal impoundment. It found: “The evidence would not inevitably have been discovered. In fact, Colon might well have destroyed the evidence had she not been illegally excluded [from the apartment].” App. 15. This finding indicates that there is substantial doubt as to whether all of the evidence that was actually seized would have been discovered if there had been no illegal entry and occupation.
The majority insists that the idea that access to evidence is a relevant consideration is “unsound” because it would “extend” the exclusionary rule and “further ‘protect’ criminal activity,” ante, at 816. However, this very point is far from *833novel; it actually has been the long-settled rule. It is implicit in virtually every case in which we have applied the exclusionary rule. In the seminal case, Weeks v. United States, 232 U. S. 383 (1914), federal agents illegally entered Weeks’ house and seized evidence. The Court ordered the evidence suppressed precisely because absent the illegality, the agents would never have obtained access to the evidence. See id., at 393-394. More recently, in Payton v. New York, 445 U. S. 573 (1980), we held that suppression was required because the agents were not authorized to enter the house; it was the Fourth Amendment violation that enabled them to obtain access to the evidence. Indeed, we have regularly invoked the exclusionary rule because the evidence would have eluded the police absent the illegality.27 Here, too, if the evidence would not have been available to the agents at the time they finally executed the warrant had they not illegally entered and impounded petitioners’ apartment, then it cannot be said that the agents’ access to the evidence was “independent” of the prior illegality.
The unlawful delay provides the same justification for suppression as does the unlawful entry: both violations precluded the possibility that evidence would have been moved out of the reach of the agents. We approved of exactly that principle only last Term, in United States v. Place, 462 U. S. 696 (1983). There, luggage was detained for some 90 minutes until a trained narcotics detection dog arrived. The dog then sniffed the luggage, signaled the presence of narcotics, a *834warrant was obtained on the strength of the dog's reaction, and when the warrant was executed, narcotics were discovered. The Court held that while the initial seizure was lawful, it became unreasonable because of its duration. Thus, absent the illegality, the authorities would have had to give the luggage back to Place, who would have then taken it away.28 The evidence was obtained in violation of the Fourth Amendment because it was the unlawful delay that prevented the evidence from disappearing before it could be obtained by the authorities. That is precisely the claim made by petitioners here.
When it finally does confront petitioners’ claim concerning the relationship between the unlawful occupation of their apartment and the evidence obtained at the conclusion of that occupation, ante, at 815-816, the Court rejects it for two reasons. First, it finds the possibility that the evidence would not have been in the apartment had it not been impounded to be speculative. However, the District Court found a distinct, nonspeculative possibility that the evidence would not have been available to the police had they not entered the apartment illegally. The Court is obligated to respect that finding unless found to be clearly erroneous, which it is not. Indeed, it is equally speculative to assume that the occupants of the apartment would not have become concerned enough to take some action had Segura been missing for 18-20 hours.29 Second, the Court thinks it “prudentially unsound” *835to suppress the evidence, noting a certain irony in extending the protection of the Constitution simply because criminals may destroy evidence if given the chance. This analysis confuses two separate issues however: (1) whether the initial entry was justified by exigent circumstances; and (2) whether the discovery of the evidence can be characterized as “inevitable” notwithstanding the 18-hour delay. There is no dispute that the risk of immediate destruction did not justify the entry. The argument petitioners make is not that there was some immediate threat of destruction of evidence, but that there was a substantial possibility that over the course of 18-20 hours at least some of the evidence would have been removed or destroyed.30
*836For me, however, the controlling question should not be answered merely on the basis of such speculation, but rather by asking whether the deterrent purposes of the exclusionary rule would be served or undermined by suppression of this evidence. That is the appropriate “prudential” consideration identified in our exclusionary rule cases. The District Court found that there was a distinct possibility that the evidence was preserved only through an illegal occupation of petitioners’ apartment. That possibility provides a sufficient reason for asking whether the deterrent rationale of the exclusionary rule is applicable to the second constitutional violation committed by the police in this case.
V
The importance of applying the exclusionary rule to the police conduct in this case is underscored by its facts. The 18-20-hour occupation of petitioners’ home was blatantly unconstitutional. At the same time, the law enforcement justification for engaging in such conduct is exceedingly weak. There can be no justification for inordinate delay in securing a warrant. Thus, applying the exclusionary rule to such conduct would impair no legitimate interest in law enforcement. Moreover, the deterrence rationale of the rule is plainly applicable. The agents impounded this apartment precisely because they wished to avoid risking a loss of access to the evidence within it. Thus, the unlawful benefit they acquired through the impoundment was not so “attenuated” as to make it unlikely that the deprivation of that benefit through the exclusionary rule would have a deterrent effect. To the contrary, it was exactly the benefit identified by the District *837Court — avoiding a risk of loss of evidence — that motivated the agents in this case to violate the Constitution. Thus, the policies underlying the exclusionary rule demand that some deterrent be created to this kind of unconstitutional conduct. Yet the majority’s disposition of this case creates none. Under the majority’s approach, the agents could have remained indefinitely — impounding the apartment for a week or a month — without being deprived of the advantage derived from the unlawful impoundment. We cannot expect such an approach to prevent similar violations of the Fourth Amendment in the future.
In my opinion the exclusionary rule should be applied to both of the constitutional violations to deprive the authorities of the advantage they gained as a result of their unconstitutional entry and impoundment of petitioners’ apartment. The deterrence rationale of the exclusionary rule requires suppression unless the Government can prove that the evidence in fact would have remained in the apartment had it not been unlawfully impounded. The risk of uncertainty as to what would have happened absent the illegal conduct posed by the facts of this case should be borne by the party that created that uncertainty, the Government. That is the teaching of our exclusionary rule cases. See Taylor v. Alabama, 457 U. S. 687, 690 (1982); Dunaway v. New York, 442 U. S. 200, 218 (1979); Brown v. Illinois, 422 U. S., at 604.
Further proceedings are necessary in this case if petitioners’ claim is to be properly evaluated. The District Court found only that there was a demonstrable possibility that the evidence obtained during the execution of the search warrant would have been destroyed absent the illegal entry and impoundment. While this finding is sufficient to establish prima facie that the Government exploited the illegality by avoiding a risk of losing the evidence in the apartment, the existence of a mere possibility cannot be equated with an ultimate finding that such exploitation did in fact occur. The *838District Court made no specific finding as to whether the Government had demonstrated that the evidence obtained pursuant to the search warrant would have remained in the apartment had the agents not illegally entered and impounded it. It may be that an evidentiary hearing would be necessary to supplement the record on this point. Accordingly, I would remand this case to the Court of Appeals with instructions that it be remanded to the District Court for further proceedings.
VI
The Government did not contest the blatant unconstitutionality of the agents’ conduct in this case. Nevertheless, today’s holding permits federal agents to benefit from that conduct by avoiding the risk that evidence would be unavailable when the search warrant was finally executed. The majority’s invocation of the “enormous price” of the exclusionary rule and its stated unwillingess to “protect criminal activity,” ante, at 816, is the most persuasive support that the Court provides for its holding. Of course, the Court is quite right to be ever mindful of the cost of excessive attention to procedural safeguards. But an evenhanded approach to difficult cases like this requires attention to countervailing considerations as well. There are two that I would stress.
First, we should consider the impact of the Court’s holding on the leaders of the law enforcement community who have achieved great success in creating the kind of trained, professional officers who deservedly command the respect of the communities they serve. The image of the “keystone cop” whose skills seldom transcended the ham-handed employment of the “third degree” is largely a matter of memory for those of us who lived through the 1920’s, 1930’s and 1940’s. For a congeries of reasons, among which unquestionably is the added respect for the constitutional rights of the individual engendered by cases like Miranda v. Arizona, 384 U. S. 436 (1966), and Mapp v. Ohio, 367 U. S. 643 (1961), the professionalism that has always characterized the Federal *839Bureau of Investigation is now typical of police forces throughout the land. A rule of law that is predicated on the absurd notion that a police officer does not have the skill required to obtain a valid search warrant in less than 18 or 20 hours, or that fails to deter the authorities from delaying unreasonably their attempt to obtain a warrant after they have entered a home, is demeaning to law enforcement and can only encourage sloppy, undisciplined procedures.
Second, the Court’s rhetoric cannot disguise the fact that when it not only tolerates but also provides an affirmative incentive for warrantless and plainly unreasonable and unnecessary intrusions into the home, the resulting erosion of the sanctity of the home is a “price” paid by the innocent and guilty alike.31 More than half a century ago, Justice Holmes explained why the Government cannot be permitted to benefit from its violations of the Constitution.
“The Government now, while in form repudiating and condemning the illegal seizure, seeks to maintain its right to avail itself of the knowledge obtained by that means which otherwise it would not have had.
*840“The proposition could not be presented more nakedly. It is that although of course its seizure was an outrage which the Government now regrets, . . . the protection of the Constitution covers the physical possession but not any advantages that the Government can gain over the object of its pursuit by doing the forbidden act. . . . In our opinion such is not the law. It reduces the Fourth Amendment to a form of words. The essence of a provision forbidding the acquisition of evidence in a certain way is that not merely evidence so acquired shall not be used before the Court but that it shall not be used at all.” Silverthorne Lumber Co. v. United States, 251 U. S. 385, 391-392 (1920) (citation omitted).
If we are to give more than lipservice to protection of the core constitutional interests that were twice violated in this case, some effort must be made to isolate and then remove the advantages the Government derived from its illegal conduct.
I respectfully dissent.
The Chief Justice seems to think that this problem was caused by the unavailability of a magistrate to issue a warrant at this hour, ante, at 812-813. However, as the Government candidly admits, the fault here lies not with the judiciary, but with the United States Attorney’s office for failing to exercise due diligence in attempting to procure a warrant. One of the agents testified that the Assistant United States Attorney told him only that “perhaps a Magistrate could not be found at that particular time in the evening.” Tr. 154 (emphasis supplied). The Assistant United States Attorney testified that he did not even attempt to locate a magistrate or obtain a search warrant. Id., at 441-442. As the Government concedes in its brief:
“It is not clear why a greater effort was not made to obtain a search warrant when the officers first sought one, and we do not condone the failure to do so ... . We note that, subsequent to the events in this case, the United States Attorney circulated an internal memorandum reemphasizing that search warrants should be sought when at all possible, regardless of the hour, in order to avoid the need for warrantless entries to secure premises.” Brief for United States 40, n. 23.
Based on the information they had been given prior to their arrival at the apartment, the agents believed, correctly as it turned out, that Segura was not in the apartment. Tr. 394.
Id., at 366, 392.
However, none of this evidence could be seen until after the agents had entered the apartment. Id., at 405.
In Vale v. Louisiana, 399 U. S. 30 (1970), we held that absent a demonstrable threat of imminent destruction of evidence, the authorities may not enter a residence in order to preserve that evidence without a warrant. See also United States v. Jeffers, 342 U. S. 48, 51-52 (1951); McDonald v. United States, 335 U. S. 451, 454-455 (1948); Johnson v. United States, 333 U. S. 10, 13-15 (1948). The illegality is even more plain in this case because the entry was effected by force late at night.
See Oliver v. United States, 466 U. S. 170, 177 (1984); Illinois v. Andreas, 463 U. S. 765, 771 (1983); United States v. Knotts, 460 U. S. 276, 280-281 (1983); Smith v. Maryland, 442 U. S. 735, 739-741 (1979); Terry v. Ohio, 392 U. S. 1, 9 (1968).
See also, e. g., Welsh v. Wisconsin, 466 U. S. 740, 748 (1984); Michigan v. Clifford, 464 U. S. 287, 296-297 (1984) (plurality opinion); Steagald *821v. United States, 451 U. S. 204, 212 (1981); Payton v. New York, 445 U. S. 573, 583-590 (1980); Coolidge v. New Hampshire, 403 U. S. 443, 481 (1971); McDonald v. United States, 335 U. S., at 455-456; Johnson v. United States, 333 U. S., at 13-14.
See also 395 U. S., at 764-765:
“It is argued in the present case that it is ‘reasonable’ to search a man’s house when he is arrested in it. But that argument is founded on little more than a subjective view regarding the acceptability of certain sorts of police conduct, and not on considerations relevant to Fourth Amendment interests. Under such an unconfined analysis, Fourth Amendment protection in this area would approach the evaporation point. It is not easy to explain why, for instance, it is less subjectively ‘reasonable’ to search a man’s house when he is arrested on his front lawn — or just down the street — than it is when he happens to be in the house at the time of arrest.” 9
At oral argument, the Government conceded that the agents’ occupation of the apartment constituted a “continuing search” for exactly this reason. Tr. of Oral Arg. 27, 31.
See United States v. Karo, ante, at 712-713; United States v. Jacobsen, 466 U. S. 109, 120-121, 124-125 (1984); United States v. Place, 462 U. S. 696, 707-708 (1983); id., at 716 (Brennan, J., concurring in result); Texas v. Brown, 460 U. S. 730, 747-748 (1983) (Stevens, J., concurring in judgment).
While Segura was lawfully in custody during this period, Colon and her three companions were not. They were unknown to the agents prior to the illegal entry and, as the District Court noted, would have been able to remain in the apartment free from governmental interference had the unlawful entry not occurred.
The Chief Justice’s parsimonious approach to Fourth Amendment rights is vividly illustrated by the fact that, as though he were preparing an adversary’s brief, he is unwilling even to acknowledge explicitly that the apartment and its contents were seized, but only “assum[es]” that was the ease. Ante, at 806.
See Mincey v. Arizona, 437 U. S. 385, 393 (1978); G. M. Leasing Corp. v. United States, 429 U. S. 338, 358-359 (1977); Vale v. Louisiana, 399 U. S., at 34-35; Chimel v. California, 395 U. S. 752, 762-763 (1969).
The Chief Justice’s misuse of Place, ante, at 813, n. 8, is quite remarkable. He suggests that Place approved the almost 3-day detention of Place’s luggage before a warrant was obtained, when in fact the Court had *824no occasion to reach that issue because it held that the initial 90-minute detention of the luggage pending a “sniff test” using a trained narcoties-detecting dog was unreasonable. See 462 U. S., at 710. Other than this reference to Place, The Chief Justice’s diligent search for support for his holding has produced nothing but dissenting opinions and a law review article. See ante, at 809-810, n. 7. Dean Griswold’s article, however, did not even purport to answer the question presented by this case. See Griswold, Criminal Procedure, 1969 — Is It a Means or an End?, 29 Md. L. Rev. 307, 317 (1969).
Since these premises were impounded “from the inside,” I assume im-poundment would be permissible even absent exigent circumstances when it occurs “from the outside” — when the authorities merely seal off premises pending the issuance of a warrant but do not enter.
See also Steagald v. United States, 451 U. S. 204 (1981); Payton v. New York, 445 U. S. 573 (1980); Mincey v. Arizona, 437 U. S. 385 (1978); Vale v. Louisiana, 399 U. S. 30 (1970). In fact, except for an aberrational warrantless “search incident to an arrest” exception recognized in United States v. Rabinowitz, 339 U. S. 56 (1950), and repudiated by Chimel, this rule has been settled since Agnello v. United States, 269 U. S. 20, 32-33 (1925). See also Trupiano v. United States, 334 U. S. 699 (1948).
The only explanation the Government has offered for the delay is that most of February 13 was taken up with “processing” the arrests. Brief for United States 5, n. 4. At oral argument, the Government conceded that the delay was unreasonable. Tr. of Oral Arg. 27. At the suppression hearing in the District Court, one of the agents testified that the warrant application was not even presented to a Magistrate until 5 p. m. on February 13. He explained: “Well, it’s very hard to get secretarial services today.” Tr. 162-163. The Assistant United States Attorney responsible for procuring the warrant testified similarly. Id., at 445. The attorney did not explain why he did not simply write out the two-page application by hand, or seek a telephonic warrant under Federal Rule of Criminal Procedure 41(c)(2). The District Court found that the delay was unreasonable, App. 15-16, a finding that the Court of Appeals did not disturb. The Government does not challenge that finding in this Court.
See Terry v. Ohio, 392 U. S., at 22; Beck v. Ohio, 379 U. S. 89, 97 (1964); Henry v. United States, 361 U. S. 98, 102 (1959).
United States v. Leon, post, at 915, n. 13.
Welsh v. Wisconsin, 466 U. S. 740 (1984); Payton v. New York, 445 U. S. 573 (1980).
Justice Holmes commented on this dilemma:
“[W]e must consider the two objects of desire, both of which we cannot have, and make up our minds which to choose. It is desirable that criminals should be detected, and to that end that all available evidence should *828be used. It also is desirable that the Government should not itself foster and pay for other crimes, when they are the means by which the evidence is to be obtained. If it pays its officers for having got evidence by crime I do not see why it may not as well pay them for getting it in the same way, and I can attach no importance to protestations of disapproval if it knowingly accepts and pays and announces that in future it will pay for the fruits. We have to choose, and for my part I think it a less evil that some criminals should escape than that the Government should play an ignoble part.” Olmstead v. United States, 277 U. S. 438, 470 (1928) (dissenting opinion).
Justice Stewart has written:
“[T]he Framers did not intend the Bill of Rights to be no more than unenforceable guiding principles — no more than a code of ethics under an honor system. The proscriptions and guarantees in the amendments were intended to create legal rights and duties.
“The Bill of Rights is but one component of our legal system — the one that limits the government’s reach. The primary responsibility for enforcing the Constitution’s limits on government, at least since the time of Marbury v. Madison, has been vested in the judicial branch. In general, when law enforcement officials violate a person’s Fourth Amendment rights, they do so in attempting to obtain evidence for use in criminal proceedings. To give effect to the Constitution’s prohibition against illegal searches and seizures, it may be necessary for the judiciary to remove the incentive for violating it. Thus, it may be argued that although the Constitution does not explicitly provide for exclusion, the need to enforce the Constitution’s limits on government — to preserve the rule of law — requires an exclusionary rule.” Stewart, The Road to qMapp v. Ohio and Beyond: The Origins, Development and Future of the Exclusionary Rule, 83 Colum. L. Rev. 1365, 1383-1384 (1983) (footnotes omitted).
See 3 W. LaFave, Search and Seizure § 11.4(a) (1978); Amsterdam, Search, Seizure, and Section 2255: A Comment, 112 U. Pa. L. Rev. 378, 388-390 (1964); Pitler, “The Fruit of the Poisonous Tree” Revisited and Shepardized, 56 Calif. L. Rev. 579, 586-589 (1968).
Tr. 259.
I recognize that the legality of the seizure of the evidence that was in plain view when the officers entered is not before us, but I find it necessary to discuss it since it affects the analysis of the issue that is in dispute. The Chief Justice does so as well; he relies on the deterrent effect of the suppression of the evidence found in plain view in responding to petitioners’ argument that the Court of Appeals’ decision will encourage illegal entries in the course of securing premises from the inside. Ante, at 812.
It is by no means impossible that at least one of the occupants might have been able to leave the apartment. None of them was known to the agents, and if the agents were located outside the apartment building, they would not have known that a person leaving the building would have come from petitioners’ apartment. There were quite a few apartments on each floor of the apartment building. Tr. 253. Moreover, as the District Court noted, the agents could not see petitioners’ apartment from their position in the front of the building.
The element of access, rather than information, is central to virtually the whole of our jurisprudence under the Warrant Clause of the Fourth Amendment. In all of our cases suppressing evidence because it was obtained pursuant to a warrantless search, we have focused not on the authorities’ lack of appropriate information to authorize the search, but rather on the fact that that information was not presented to a magistrate. Thus, suppression is the consequence not of a lack of information, but of the fact that the authorities’ access to the evidence in question was not properly authorized and hence was unconstitutional.
Even more recently, in Welsh v. Wisconsin, 466 U. S. 740 (1984), we again employed this concept. The Court held that police could not justify under the Fourth Amendment the warrantless arrest of Welsh, who was suspected of drunken driving, in his own home, “simply because evidence of the petitioner’s blood-alcohol level might have dissipated while the police obtained a warrant.” Id., at 754 (footnote omitted).
The Court of Appeals, with which this Court agrees, noted that the District Court’s ruling depended on “speculative assumptions,” such as that the agents would not have kept the apartment under surveillance after Segura’s arrest had they not illegally entered it, that Colon would have *835destroyed the evidence rather than merely removed it from the apartment, or that the evidence could have been destroyed unobtrusively. However, each of these “assumptions” is supported by the evidence. First, the agents would have had no reason to keep the apartment under surveillance subsequent to the arrests of all the persons that they had surveilled, Parra, Rivudalla-Vidal, and Segura. Second, even if Colon had merely removed the evidence from the apartment, there is reason to believe the agents would not have intercepted her. See n. 26, swpra. Third, since the agents were outside the apartment and would have had no reason to remain on the scene after Segura’s arrest, they would not have been around to notice had evidence been removed or destroyed unobtrusively. Moreover, even if it would have been difficult to remove or destroy some of the evidence, such as the triple-beam scale petitioners owned, that does not mean that all of the evidence would have remained in the apartment over the course of an 18-20-hour period. The Court of Appeals’ assumptions to the contrary are just as “speculative” as the finding of the District Court.
The cases in the lower courts the majority cites in support of its holding, ante, at 814, n. 9, are plainly distinguishable. In United States v. Perez, 700 F. 2d 1232, 1237-1238 (CA8 1983), the court remanded for a hearing as to whether the search and seizure authorized by a warrant was tainted by prior illegality. In United States v. Kinney, 638 F. 2d 941, 945 (CA6), cert. denied, 452 U. S. 918 (1981), the court found no taint, but in that case there was no occupation of the searched premises prior to obtaining the warrant and hence no claim of the type made here. The same is true of the other cases the Court cites, United States v. Bosby, 675 F. 2d *8361174, 1180-1181 (CA11 1982); United States v. Fitzharris, 633 F. 2d 416 (CA5 1980), cert. denied, 451 U. S. 988 (1981); United States v. Agapito, 620 F. 2d 324, 338 (CA2), cert. denied, 449 U. S. 834 (1980). As the Court concedes, United States v. Lomas, 706 F. 2d 886 (CA9 1983), and United States v. Allard, 634 F. 2d 1182 (CA9 1980), are contrary to its holding.
The words that this case calls to my mind are not those of Nardone, ante, at 816, but rather those in two of Justice Jackson’s dissents. With respect to the claim that the Fourth Amendment “protect[s] criminal activity,” he wrote: “Only occasional and more flagrant abuses come to the attention of the courts, and then only those where the search and seizure yields incriminating evidence and the defendant is at least sufficiently compromised to be indicted. . . . Courts can protect the innocent against such invasions only indirectly and through the medium of excluding evidence obtained against those who frequently are guilty. ... So a search against Brinegar’s car must be regarded as a search of the car of Everyman.” Brinegar v. United States, 338 U. S. 160, 181 (1949). And with respect to the “price” exacted by the exclusionary rule, he wrote: “[T]he forefathers thought this was not too great a price to pay for that decent privacy of home, papers and effects which is indispensable to individual dignity and self-respect. They may have overvalued privacy, but I am not disposed to set their command at naught.” Harris v. United States, 331 U. S. 145, 198 (1947).