BRENNAN, J., dissenting 466 U. S. the way for technology to override the limits of law in the area of criminal investigation. For example, under the Court's analysis in these cases, law enforcement officers could release a trained cocaine-sensitive dog-to paraphrase the California Court of Appeal, a "canine cocaine connoisseur"-to roam the streets at random, alerting the officers to people carrying cocaine. Cf. People v. Evans, 65 Cal. App. 3d 924, 932, 134 Cal. Rptr. 436, 440 (1977). Or, if a device were developed that, when aimed at a person, would detect instantaneously whether the person is carrying cocaine, there would be no Fourth Amendment bar, under the Court's approach, to the police setting up such a device on a street corner and scanning all passersby. In fact, the Court's analysis is so unbounded that if a device were developed that could detect, from the outside of a building, the presence of cocaine inside, there would be no constitutional obstacle to the police cruising through a residential neighborhood and using the device to identify all homes in which the drug is present. In short, under the interpretation of the Fourth Amendment first suggested in Place and first applied in this case, these surveillance techniques would not constitute searches and therefore could be freely pursued whenever and wherever law enforcement officers desire. Hence, at some point in the future, if the Court stands by the theory it has adopted today, search warrants, probable cause, and even "reasonable suspicion" may very well become notions of the past. Fortunately, we know from precedents such as Katz v. United States, supra, overruling the "trespass" doctrine of Goldman v. United States, 316 U. S. 129 (1942), and Olmstead v. United States, 277 U. S. 438 (1928), that this Court ultimately stands ready to prevent this Orwellian world from coming to pass. Although the Court accepts, as it must, the fundamental proposition that an investigative technique is a search within the meaning of the Fourth Amendment if it intrudes upon a privacy expectation that society considers to be reasonable, 109 BRENNAN, J., dissenting ante, at 113, the Court has entirely omitted from its discussion the considerations that have always guided our decisions in this area. In determining whether a reasonable expectation of privacy has been violated, we have always looked to the context in which an item is concealed, not to the identity of the concealed item. Thus in cases involving searches for physical items, the Court has framed its analysis first in terms of the expectation of privacy that normally attends the location of the item and ultimately in terms of the legitimacy of that expectation. In United States v. Chadwick, 433 U. S. 1 (1977), for example, we held that "[n]o less than one who locks the doors of his home against intruders, one who safeguards his possessions [by locking them in a footlocker] is due the protection of the Fourth Amendment . . . " Id., at 11. Our holding was based largely on the observation that, "[b]y placing personal effects inside a double-locked footlocker, respondents manifested an expectation that the contents would remain free from public examination." Ibid. The Court made the same point in United States v. Ross, 456 U. S. 798, 822-823 (1982), where it held that the "Fourth Amendment provides protection to the owner of every container that conceals its contents from plain view." The fact that a container contains contraband, which indeed it usually does in such cases, has never altered our analysis. Similarly, in Katz v. United States, we held that electronic eavesdropping constituted a search under the Fourth Amendment because it violated a reasonable expectation of privacy. In reaching that conclusion, we focused upon the private context in which the conversation in question took place, stating: "What a person knowingly exposes to the public. . . is not a subject of Fourth Amendment protection. . . . But what he seeks to preserve as private, even in an area accessible to the public, may be constitutionally protected." 389 U. S., at 351-352. Again, the fact that the conversations involved in Katz were incriminating did not alter our consideration of the BRENNAN, J., dissenting 466 U. S. privacy issue. Nor did such a consideration affect our analysis in Payton v. New York, 445 U. S. 573 (1980), in which we reaffirmed the principle that the home is private even though it may be used to harbor a fugitive. In sum, until today this Court has always looked to the manner in which an individual has attempted to preserve the private nature of a particular fact before determining whether there is a reasonable expectation of privacy upon which the government may not intrude without substantial justification. And it has always upheld the general conclusion that searches constitute at least "those more extensive intrusions that significantly jeopardize the sense of security which is the paramount concern of Fourth Amendment liberties." United States v. White, 401 U. S. 745, 786 (1971) (Harlan, J., dissenting). Nonetheless, adopting the suggestion in Place, the Court has veered away from this sound and well-settled approach and has focused instead solely on the product of the would-be search. In so doing, the Court has ignored the fundamental principle that "[a] search prosecuted in violation of the Constitution is not made lawful by what it brings to light." Byars v. United States, 273 U. S. 28, 29 (1927). The unfortunate product of this departure from precedent is an undifferentiated rule allowing law enforcement officers free rein in utilizing a potentially broad range of surveillance techniques that reveal only whether or not contraband is present in a particular location. The Court's new rule has rendered irrelevant the circumstances surrounding the use of the technique, the accuracy of the technique, and the privacy interest upon which it intrudes. Furthermore, the Court's rule leaves no room to consider whether the surveillance technique is employed randomly or selectively, a consideration that surely implicates Fourth Amendment concerns. See 2 W. LaFave, Search and Seizure §2.2(f) (1978). Although a technique that reveals only the presence or absence of illegal 109 BRENNAN, J., dissenting activity intrudes less into the private life of an individual under investigation than more conventional techniques, the fact remains that such a technique does intrude. In my view, when the investigation intrudes upon a domain over which the individual has a reasonable expectation of privacy, such as his home or a private container, it is plainly a search within the meaning of the Fourth Amendment. Surely it cannot be that the individual's reasonable expectation of privacy dissipates simply because a sophisticated surveillance technique is employed. This is not to say that the limited nature of the intrusion has no bearing on the general Fourth Amendment inquiry. Although there are very few exceptions to the general rule that warrantless searches are presumptively unreasonable, the isolated exceptions that do exist are based on a "balancing [of] the need to search against the invasion which the search entails." Camara v. Municipal Court, 387 U. S. 523, 537 (1967). Hence it may be, for example, that the limited intrusion effected by a given surveillance technique renders the employment of the technique, under particular circumstances, a "reasonable" search under the Fourth Amendment. See United States v. Place, 462 U. S., at 723 (BLACKMUN, J., concurring in judgment) (“a dog sniff may be a search, but a minimally intrusive one that could be justified in this situation under Terry"). At least under this wellsettled approach, the Fourth Amendment inquiry would be broad enough to allow consideration of the method by which a surveillance technique is employed as well as the circumstances attending its use. More important, however, it is only under this approach that law enforcement procedures, like those involved in this case and in Place, may continue to be governed by the safeguards of the Fourth Amendment. B In sum, the question whether the employment of a particular surveillance technique constitutes a search depends on BRENNAN, J., dissenting 466 U. S. whether the technique intrudes upon a reasonable expectation of privacy. This inquiry, in turn, depends primarily on the private nature of the area or item subjected to the intrusion. In cases involving techniques used to locate or identify a physical item, the manner in which a person has attempted to shield the item's existence or identity from public scrutiny will usually be the key to determining whether a reasonable expectation of privacy has been violated. Accordingly, the use of techniques like the dog sniff at issue in Place constitutes a search whenever the police employ such techniques to secure any information about an item that is concealed in a container that we are prepared to view as supporting a reasonable expectation of privacy. The same would be true if a more technologically sophisticated method were developed to take the place of the dog. In this case, the chemical field test was used to determine whether certain white powder was cocaine. Upon visual inspection of the powder in isolation, one could not identify it as cocaine. In the abstract, therefore, it is possible that an individual could keep the powder in such a way as to preserve a reasonable expectation of privacy in its identity. For instance, it might be kept in a transparent pharmaceutical vial and disguised as legitimate medicine. Under those circumstances, the use of a chemical field test would constitute a search. However, in this case, as hypothesized above, see supra, at 134, the context in which the powder was found could not support a reasonable expectation of privacy. In particular, the substance was found in four plastic bags, which had been inside a tube wrapped with tape and sent to respondents via Federal Express. It was essentially inconceivable that a legal substance would be packaged in this manner for transport by a common carrier. Thus, viewing the powder as they did at the offices of Federal Express, the DEA agent could identify it with "virtual certainty"; it was essentially as though the chemical identity of the powder was |