with whom Justice Marshall joins, dissenting.
The underlying question in this case is very simple: whether a second search after a prior legal search and a “controlled delivery” will ordinarily require a warrant. The Court answers that question by announcing that the second search is not a search at all, but merely a “reopening,” ante, *774at 772, not subject to the protection of the Fourth Amendment. I suppose one should be grateful that the Court has not explicitly opened one more breach in the general rule that “ ‘ “searches conducted outside the judicial process, without prior approval by judge or magistrate, are per se unreasonable under the Fourth Amendment — subject only to a few specifically established and well-delineated exceptions.”’” United States v. Ross, 456 U. S. 798, 825 (1982), quoting Mincey v. Arizona, 437 U. S. 385, 390 (1978), in turn quoting Katz v. United States, 389 U. S. 347, 357 (1967).1 On the other hand, the Court’s rationale, even though limited to a very specific fact pattern, is nevertheless astounding in its implications. We have, to my knowledge, never held that the physical opening and examination of a container in the possession of an individual was anything other than a “search.” It might be a permissible search or an impermissible search, require a warrant or not require a warrant, but it is in any event a “search.”2
I
A
The Court’s primary argument in favor of its “no-search” holding can be stated briefly:
“The threshold question ... is whether an individual has a legitimate expectation of privacy in the contents of a *775previously lawfully searched container. It is obvious that the privacy interest in the contents of a container diminishes with respect to a container that law enforcement authorities have already lawfully opened and found to contain illicit drugs. No protected privacy interest remains in contraband in a container once government officers lawfully have opened that container and identified its contents as illegal. The simple act of sealing the container to enable the police to make a controlled delivery does not operate to revive or restore the lawfully invaded privacy rights.” Ante, at 771.
The validity of this reasoning depends, however, on what the Court means by “protected privacy interest.” Clearly, one aspect of the privacy interest protected by the Fourth Amendment is the right to keep certain information beyond official scrutiny. See United States v. Knotts, 460 U. S. 276, 281-282 (1983) (no reasonable expectation of privacy in location of automobile on public roads). If this were all that were meant by the notion of privacy embodied in the Fourth Amendment, the Court’s analysis would be essentially correct. Respondent knowingly and voluntarily rendered his container vulnerable to a perfectly legal and perfectly proper border search. And as soon as that search revealed the presence of contraband, any reasonable expectation respondent may have had that the existence of the contraband would remain secret was lost, and could not be regained.
The Fourth Amendment, however, does not protect only information. It also protects, in its own sometimes-forgotten words, “[t]he right of the people to be secure in their persons, houses, papers, and effects . . .” (emphasis added). As Justice Brandéis put the matter in his dissent in Olmstead v. United States, 277 U. S. 438, 478 (1928), the Fourth Amendment “conferred, as against the Government, the right to be let alone — the most comprehensive of rights and the right most valued by civilized men” (emphasis added). The right to be “let alone” is, at the very least, the right not to have one’s repose and possessions disturbed. See, e. g., Rakas v. *776Illinois, 439 U. S. 128 (1978); United States v. United States District Court, 407 U. S. 297, 326-327 (1972) (Douglas, J., concurring); Alderman v. United States, 394 U. S. 165, 179-180 (1969); Silverman v. United States, 365 U. S. 505, 511-512 (1961); Taylor v. United States, 286 U. S. 1 (1932); Boyd v. United States, 116 U. S. 616, 626-630 (1886).3 In this case, respondent had the right to maintain the integrity of his container. Admittedly, he waived that right temporarily when the container passed through customs inspection; as Carroll v. United States, 267 U. S. 132 (1925), teaches us, the right of the Government to search, with or without probable cause, persons and property entering the country is necessary to “national self protection.” Id., at 154. But however justified the search at customs may have been, that justification no longer existed once the container was sent on its way, and certainly did not exist once the container was delivered to respondent.
That the Court’s reduction of the right to privacy to the right to secrecy is incorrect, and that its implicit analogy between a border search and a loss of amateur status is inapt, is made quite clear by a number of our recent cases.4 In Lo-Ji *777Sales, Inc. v. New York, 442 U. S. 319 (1979), for example, we reviewed the warrantless search of an “adult bookstore” by local law enforcement officials. The Chief Justice, speaking for a unanimous Court, stated:
“The suggestion is [made] that by virtue of its display of the items at issue to the general public in areas of its store open to them, petitioner had no legitimate expectation of privacy against governmental intrusion, see Rakas v. Illinois, 439 U. S. 128 (1978), and that accordingly no warrant was needed. But there is no basis for the notion that because a retail store invites the public to enter, it consents to wholesale searches and seizures that do not conform to Fourth Amendment guarantees. See Lewis v. United States, 385 U. S. 206, 211 (1966).” Id., at 329.
Cf. Walter v. United States, 447 U. S. 649, 660-662 (1980) (White, J., concurring in judgment). Similarly, in Michigan v. Tyler, 436 U. S. 499 (1978), we held that, although a building fire and its immediate aftermath are “exigent circumstances” justifying the warrantless entry of the building both by firefighters and by investigators, any further intrusions that take place after the exigent circumstances have passed require a warrant. The fire may suspend the right to be let alone, but it does not extinguish it, and an initial search does not validate the legality of subsequent independent warrantless searches, let alone render them nonsearches. Cf. G. M. Leasing Corp. v. United States, 429 U. S. 338, 358-359 (1977).
*778Thus, in its analysis today, the Court breaks new ground and erodes the principles of the Fourth Amendment. Moreover, by claiming that the right to “title and possession” confers no right to “privacy,” ante, at 771, the Court adopts a view curiously out of touch with the genius of the American system of liberties.
B
The Court supports its “no-search” analysis by an analogy to the “reasoning underlying the ‘plain-view’ doctrine.” Ibid. In fact, however, the “plain-view” doctrine hurts rather than helps the Court’s case, for it recognizes and indeed emphasizes that the Fourth Amendment protects security as well as secrecy.
“We recognized in Payton v. New York, 445 U. S. 573, 587 (1980), the well-settled rule that ‘objects such as weapons or contraband found in a public place may be seized by the police without a warrant. The seizure of property in plain view involves no invasion of privacy and is presumptively reasonable, assuming that there is probable cause to associate the property with criminal activity.’ A different situation is presented, however, when the property in open view is ‘situated on private premises to which access is not otherwise available for the seizing officer.’ Ibid., quoting G. M. Leasing Corp. v. United States, 429 U. S. 338, 354 (1977). As these cases indicate, ‘plain view’ provides grounds for seizure of an item when an officer’s access to an object has some prior justification under the Fourth Amendment. ‘Plain view’ is perhaps better understood, therefore, not as an independent ‘exception’ to the Warrant Clause, but simply as an extension of whatever the prior justification for an officer’s ‘access to an object’ may be.” Texas v. Brown, 460 U. S. 730, 738-739 (1983) (opinion of Rehnquist, J.) (footnote omitted).
See also id., at 747-749 (Stevens, J., concurring in judgment); Coolidge v. New Hampshire, 403 U. S. 443, 464-471 (1971) *779(plurality opinion). Thus, under the “plain-view” doctrine, the fact that a person displays incriminating evidence in his living room window5 (or allows it to pass through customs inspection) is not enough by itself to authorize a search and seizure of that evidence. More is necessary, and that “more” must be some independent reason for breaching the individual’s right to repose and to security in his possessions. Moreover, as the Court itself admits, “plain view” can only justify a search or seizure of an item if the authorities have “probable cause to suspect that the item is connected with criminal activity.” Ante, at 771. Obviously, there would be no need to require probable cause if the protections of the Fourth Amendment did not apply at all to the search or seizure in question. Cf. n. 2, supra.
C
The plain-view doctrine does, of course, highlight the fact that there are certain “specifically established and well-delineated exceptions” to the Fourth Amendment’s warrant requirement. See supra, at 774. Such exceptions, however, require at the very least that there be some compelling government interest at stake, not merely in the search at issue, but in the right to conduct the search without a warrant.6 Moreover, we have repeatedly made clear that “the police must, whenever practicable, obtain advance judicial approval of searches and seizures through the warrant procedure.” Terry v. Ohio, 392 U. S. 1, 20 (1968). See United States v. Chadwick, 433 U. S. 1, 13 (1977); United States v. United States District Court, 407 U. S., at 315-318; Chimel v. California, 395 U. S. 752, 762 (1969); Johnson v. United *780States, 333 U. S. 10, 15 (1948); Carroll v. United States, 267 U. S., at 153. Indeed, each of the limited exceptions we have established to the warrant requirement arose in a context in which, at the very least, a warrantless search was necessary to preserve the safety of law enforcement officers, see, e. g., Chimel v. California, supra (search incident to arrest), or to prevent the loss or destruction of evidence, see, e. g., Chambers v. Maroney, 399 U. S. 42, 48-51 (1970) (automobile exception), or in which the very special nature of the government interest made it appropriate to allow a search based on something less than probable cause, see, e. g., Carroll v. United States, supra, at 154 (border search). In the plain-view context, the compelling government interest is evident: the legal search has already put potential suspects on notice that they are the objects of official interest; the delay inherent in obtaining a warrant at that point might risk the destruction of the evidence and even the security of the officers. See Coolidge, supra, at 467-468 (plurality opinion). This case, however, presents none of the conditions that we have previously held indispensable to the recognition of an exception to the warrant requirement. The police officers who conducted the search of respondent’s container could have obtained, were indeed in the process of obtaining, a search warrant, but decided instead — for no apparent reason other than the hope of vindication in this Court — to conduct the search without a warrant. Thus, even if one were to recharacterize the Court’s novel “no-search” analysis as simply another exception to the warrant requirement, it would be difficult to square that result with the clear mandate of our previous decisions.
I agree entirely with the Court that “controlled delivery” is a proper and effective tool of responsible law enforcement. See ante, at 769-770. If contraband is discovered in a package passing through customs inspection, the authorities are not required to seize it then and there, but may make use of their discovery to obtain more evidence and to capture the culprits behind the contraband. The “controlled delivery” tech-*781ñique, however, would be just as effective, and decidedly more proper, if the second search that came at its culmination were authorized by a valid search warrant. Under these circumstances, I am not at all sure what interest the Court thinks it is vindicating by its determined if awkward exertions.
II
Even if the Court were correct that the “reopening” of a package after a properly controlled “controlled delivery” is not a “search,” I could still not agree with the standard it fashions to put that principle into effect, or with the result it reaches in this case. The Court holds that a “reopening” is not a “search” as long as there is not a “substantial likelihood that the contents of the container have been changed during [a] gap in surveillance.” Ante, at 773. Of course, “the rigors and contingencies inescapable in an investigation into illicit drug traffic often make 'perfect’ controlled deliveries and . . . ‘absolute certainty’. . . impossible,” ante, at 772. Nevertheless, the very justifications proffered by the Court for its “no-search” analysis should have at least led it to require something very close to “absolute certainty.” Cf. post, p. 782 (Stevens, J., dissenting). After all, if a person has no reasonable expectation of privacy in a package whose contents are already legally known to the authorities, a reasonable expectation of privacy should reattach if the person has unobserved access to the package and any opportunity to change its contents. By adopting a vague intermediate standard, the Court makes more likely serious intrusions into what even it would consider to be “reasonable expectations of privacy.” Moreover, I cannot see how as indistinct a phrase as “substantial likelihood” could in any way serve the Court’s interest in fashioning a standard “workable for application by rank-and-file, trained police officers.” Ante, at 772.
In this case, the package subject to a “controlled delivery” was in respondent’s possession for between 30 and 45 minutes. For a good deal of that time, it was unobserved. I am by no means convinced that there was, as an ex ante matter, *782even a “substantial likelihood” that the container still contained contraband when it was searched, or “reopened.” In any event, I fail to see how, in light of the very justifications put forward by the Court for a “controlled delivery” gloss on the Fourth Amendment, a warrantless search can be justified here as in any way consistent with the principles embodied in that Amendment.
I dissent.
See also, e. g., Arkansas v. Sanders, 442 U. S. 753, 759 (1979); G. M. Leasing Corp. v. United States, 429 U. S. 338, 358 (1977); United States v. United States District Court, 407 U. S. 297, 318 (1972); Camara v. Municipal Court, 387 U. S. 523, 528-529 (1967); Jones v. United States, 357 U. S. 493, 499 (1958).
Indeed, if the “reopening” of a package in a controlled delivery context is not a “search,” it is not even clear why it should require probable cause. Fortunately, though, the Court seems to reject this implication of its reasoning. See ante, at 771 (“No protected privacy interest remains in contraband in a container once government officials lawfully have opened that container and identified its contents as illegal”); ante, at 771-772 (“once a container has been found to a certainty to contain illicit drugs, the contraband becomes like objects physically within the plain view of the police, and the claim to privacy is lost”) (footnote omitted).
See generally Posner, The Uncertain Protection of Privacy by the Supreme Court, 1979 S. Ct. Rev. 173 (discussing “seclusion” and “secrecy” aspects of privacy right protected by the Fourth Amendment); cf. Whalen v. Roe, 429 U. S. 589, 599, nn. 24-25 (1977).
The Court’s confusion may be in part an unintended consequence of our decision in Katz v. United States, 389 U. S. 347 (1967), where we held that electronic eavesdropping was subject to the warrant requirement even if it involved no physical intrusion into a suspect’s “protected area.” Before Katz, this Court may have focused too much on the “security” aspect of the right of privacy, while giving short shrift to its “secrecy” aspect. In recognizing the importance of secrecy, however, Katz did not extinguish the relevance of security. As I wrote only recently, Katz “made quite clear that the Fourth Amendment protects against governmental invasions of a person’s reasonable ‘expectation^] of privacy,’ even when those invasions are not accompanied by physical intrusions. Cases such as Silverman v. United States, 365 U. S. 505, 509-512 (1961), how*777ever, hold that, when the Government does engage in physical intrusion of a constitutionally protected area in order to obtain information, that intrusion may constitute a violation of the Fourth Amendment even if the same information could have been obtained by other means. I do not believe that Katz, or its progeny, have eroded that principle.” United States v. Knotts, 460 U. S. 276, 286 (1983) (Brennan, J., concurring in judgment).
Cf. Coolidge v. New Hampshire, 403 U. S., at 468, and n. 25 (plurality opinion); Taylor v. United States, 286 U. S. 1 (1932).
“In assessing whether the public interest demands creation of a general exception to the Fourth Amendment’s warrant requirement the question is not whether the public interest justifies the type of search in question, but whether the authority to search should be evidenced by a warrant. . . .” Camara v. Municipal Court, 387 U. S., at 533.