with whom Justice Brennan, Justice Marshall, and Justice Stevens join, dissenting.
The facts of this case are simple and straightforward. Dr. Ortega had an expectation of privacy in his office, desk, and file cabinets, which were the target of a search by petitioners that can be characterized only as investigatory in nature. Because there was no “special need,” see New Jersey v. T. L. O., 469 U. S. 325, 351 (1985) (opinion concurring in judgment), to dispense with the warrant and probable-cause requirements of the Fourth Amendment, I would evaluate the search by applying this traditional standard. Under that *733standard, this search clearly violated Dr. Ortega’s Fourth Amendment rights.
The problems in the plurality’s opinion all arise from its failure or unwillingness to realize that the facts here are clear. The plurality, however, discovers what it feels is a factual dispute: the plurality is not certain whether the search was routine or investigatory. Accordingly, it concludes that a remand is the appropriate course of action. Despite the remand, the plurality assumes it must announce a standard concerning the reasonableness of a public employer’s search of the workplace. Because the plurality treats the facts as in dispute, it formulates this standard at a distance from the situation presented by this case.
This does not seem to me to be the way to undertake Fourth Amendment analysis, especially in an area with which the Court is relatively unfamiliar.1 Because this analysis, when conducted properly, is always fact specific to an extent, it is inappropriate that the plurality’s formulation of a standard does not arise from a sustained consideration of a particular factual situation.2 Moreover, given that any standard *734ultimately rests on judgments about factual situations, it is apparent that the plurality has assumed the existence of hypothetical facts from which its standard follows. These “assumed” facts are weighted in favor of the public employer,3 and, as a result, the standard that emerges makes reasonable almost any workplace search by a public employer.
I
It is necessary to review briefly the factual record m this case because of the plurality’s assertion, ante, at 728, that *735“[t]here was a dispute of fact about the character of the search.” The plurality considers it to be either an inventory search to secure government property or an investigative search to gather evidence concerning Dr. Ortega’s alleged misdeeds. Ante, at 727-728. It is difficult to comprehend how, on the facts of this case, the search in any way could be seen as one for inventory purposes. As the plurality concedes, the search could not have been made pursuant to the Hospital’s policy of routinely inventorying state property in an office of a terminated employee, because at the time of the search Dr. Ortega was on administrative leave and had not been terminated. Ante, at 712-713.4 Napa had no policy of inventorying the office of an employee placed on administrative leave. Ante, at 713.
The plurality, however, observes that the absence of the policy does not dispositively eliminate inventorying or securing state property as a possible purpose for conducting the search. Ante, at 728. As evidence suggesting such a purpose, the plurality points to petitioners’ concern that Dr. Ortega may have removed from the Hospital’s grounds a computer owned by the Hospital and to their desire to secure such items as files located in Dr. Ortega’s office. See ante, at 727-728.
The record evidence demonstrates, however, that ensuring that the computer had not been removed from the Hospital was not a reason for the search. Mr. Friday, the leader of the “investigative team,” stated that the alleged removal of the computer had nothing to do with the decision to enter Dr. Ortega’s office. App. 59. Dr. O’Connor himself admitted that there was little connection between the entry and an at*736tempt by petitioners to ascertain the location of the computer. Id., at 39. The search had the computer as its focus only insofar as the team was investigating practices dealing with its acquisition. Id., at 32.
In deposition testimony, petitioners did suggest that the search was inventory in character insofar as they aimed to separate Dr. Ortega’s personal property from Hospital property in the office. Id., at 38, 40, 50. Such a suggestion, however, is overwhelmingly contradicted by other remarks of petitioners and particularly by the character of the search itself. Dr. O’Connor spoke of the individuals involved in the search as “investigators,” see id., at 37, and, even where he described the search as inventory in nature, he observed that it was aimed primarily at furthering investigative purposes. See, e. g., id., at 40 (“Basically what we were trying to do is to remove what was obviously State records or records that had to do with his program, his department, any of the materials that would be involved in running the residency program, around contracts, around the computer, around the areas that we were interested in investigating”). Moreover, as the plurality itself recognizes, ante, at 713-714, the “investigators” never made a formal inventory of what they found in Dr. Ortega’s office. Rather, they rummaged through his belongings and seized highly personal items later used at a termination proceeding to impeach a witness favorable to him. Ibid. Furthermore, the search was conducted in the evening, App. 53, and it was undertaken only after the investigators had received legal advice, id., at 51.
The search in question stemmed neither from a Hospital policy nor from a practice of routine entrances into Dr. Ortega’s office. It was plainly exceptional and investigatory in nature. Accordingly, there is no significant factual dispute in this case.
II
Before examining the plurality’s standard of reasonableness for workplace searches, I should like to state both my *737agreement and disagreement with the plurality’s discussion of a public employee’s expectation of privacy. What is most important, of course, is that in this case the plurality acknowledges that Dr. Ortega had an expectation of privacy in his desk and file cabinets, ante, at 719, and that, as the plurality concedes, ante, at 718, the majority of this Court holds that he had a similar expectation in his office. With respect to the plurality’s general comments, I am in complete agreement with its observation that “[individuals do not lose Fourth Amendment rights merely because they work for the government instead of a private employer.” Ante, at 717. Moreover, I would go along with the plurality’s observation that, in certain situations, the “operational realities” of the workplace may remove some expectation of privacy on the part of the employee. Ibid. However, I am disturbed by the plurality’s suggestion, see ante, at 717-718, that routine entries by visitors might completely remove this expectation.
First, this suggestion is contrary to the traditional protection that this Court has recognized the Fourth Amendment accords to offices. See Oliver v. United States, 466 U. S. 170, 178, n. 8 (1984) (“The Fourth Amendment’s protection of offices and commercial buildings, in which there may be legitimate expectations of privacy, is also based upon societal expectations that have deep roots in the history of the Amendment”); Hoffa v. United States, 385 U. S. 293, 301 (1966) (‘What the Fourth Amendment protects is the security a man relies upon when he places himself or his property within a constitutionally protected area, be it his home or his office, his hotel room or his automobile”). The common understanding of an office is that it is a place where a worker receives an occasional business-related visitor. Thus, when the office has received traditional Fourth Amendment protection in our cases, it has been with the understanding that such routine visits occur there.
*738Moreover, as the plurality appears to recognize, see ante, at 717-718, the precise extent of an employee’s expectation of privacy often turns on the nature of the search. This observation is in accordance with the principle that the Fourth Amendment may protect an individual’s expectation of privacy in one context, even though this expectation may be unreasonable in another. See New Jersey v. T. L. O., 469 U. S., at 339. See also Lo-Ji Sales, Inc. v. New York, 442 U. S. 319, 329 (1979) (the opening of a retail store to the public does not mean that “it consents to wholesale searches and seizures that do not conform to Fourth Amendment guarantees”). As Justice Scalia observes, “[constitutional protection against unreasonable searches by the government does not disappear merely because the government has the right to make reasonable intrusions in its capacity as employer.” Ante, at 731. Thus, although an employee might well have no reasonable expectation of privacy with respect to an occasional visit by a fellow employee, he would have such an expectation as to an afterhours search of his locked office by an investigative team seeking materials to be used against him at a termination proceeding.5
*739Finally and most importantly, the reality of work in modern time, whether done by public or private employees, reveals why a public employee’s expectation of privacy in the workplace should be carefully safeguarded and not lightly set aside. It is, unfortunately, all too true that the workplace has become another home for most working Americans. Many employees spend the better part of their days and much of their evenings at work. See R. Kanter, Work and Family in the United States: A Critical Review and Agenda for Research and Policy 31-32 (1977); see also R. Bellah, R. Madsen, W. Sullivan, A. Swidler, & S. Tipton, Habits of the Heart: Individualism and Commitment in American Life 288-289 (1985) (a “less frantic concern for advancement and a reduction of working hours” would make it easier for both men and women to participate fully in'working and family life). Consequently, an employee’s private life must intersect with the workplace, for example, when the employee takes advantage of work or lunch breaks to make personal telephone calls, to attend to personal business, or to receive personal visitors in the office. As a result, the tidy distinctions (to which the plurality alludes, see ante, at 715-716) between the workplace and professional affairs, on the one hand, and personal possessions and private activities, on the other, do not exist in reality.6 Not all of an employee’s pri*740vate possessions will stay in his or her briefcase or handbag. Thus, the plurality’s remark that the “employee may avoid exposing personal belongings at work by simply leaving them at home,” ante, at 725, reveals on the part of the Members of the plurality a certain insensitivity to the “operational realities of the workplace,” ante, at 717, they so value.7
*741Dr. Ortega clearly had an expectation of privacy in his office, desk, and file cabinets, particularly with respect to the type of investigatory search involved here. In my view, when examining the facts of other cases involving searches of the workplace, courts should be careful to determine this expectation also in relation to the search in question.
1 — ( h — ( b — 4
A
At the outset of its analysis, the plurality observes that an appropriate standard of reasonableness to be applied to a public employer’s search of the employee’s workplace is arrived at from “balancing” the privacy interests of the employee against the public employer’s interests justifying the intrusion. Ante, at 719-720. Under traditional Fourth Amendment jurisprudence, however, courts abandon the warrant and probable-cause requirements, which constitute the standard of reasonableness for a government search that the Framers established, “[o]nly in those exceptional circumstances in which special needs, beyond the normal need for law enforcement, make the warrant and probable-cause requirement impracticable . . . .” New Jersey v. T. L. O., 469 U. S., at 351 (opinion'Concurring in judgment); see United States v. Place, 462 U. S. 696, 721-722, and n. 1 (1983) (opinion concurring in judgment). In sum, only when the practical realities of a particular situation suggest that a government official cannot obtain a warrant based upon probable cause without sacrificing the ultimate goals to which a search would contribute, does the Court turn to a “balancing” test to formulate a standard of reasonableness for this context.
In New Jersey v. T. L. O., supra, I faulted the Court for neglecting this “crucial step” in Fourth Amendment analysis. See 469 U. S., at 351. I agreed, however, with the T. L. O. Court’s standard because of my conclusion that this step, had *742it been taken, would have revealed that the case presented a situation of “special need.” Id., at 353. I recognized that discipline in this country’s secondary' schools was essential for the promotion of the overall goal of education, and that a teacher could not maintain this discipline if, every time a search was called for, the teacher would have to procure a warrant based on probable cause. Id., at 352-353. Accordingly, I observed: “The special need for an immediate response to behavior that threatens either the safety of schoolchildren and teachers or the educational process itself justifies the Court in excepting school searches from the warrant and probable-cause requirements, and in applying a standard determined by balancing the relevant interests.” Id., at 353.
The plurality repeats here the T. L. 0. Court’s error in analysis. Although the plurality mentions the “special need” step, ante, at 720, it turns immediately to a balancing test to formulate its standard of reasonableness. This error is significant because, given the facts of this case, no “special need” exists here to justify dispensing with the warrant and probable-cause requirements. As observed above, the facts suggest that this was an investigatory search undertaken to obtain evidence of charges of mismanagement at a time when Dr. Ortega was on administrative leave and not permitted to enter the Hospital’s grounds. There was no special practical need that might have justified dispensing with the warrant and probable-cause requirements. Without sacrificing their ultimate goal of maintaining an effective institution devoted to training and healing, to which the disciplining of Hospital employees contributed, petitioners could have taken any evidence of Dr. Ortega’s alleged improprieties to a magistrate in order to obtain a warrant.
Furthermore, this seems to be exactly the kind of situation where a neutral magistrate’s involvement would have been helpful in curtailing the infringement upon Dr. Ortega’s privacy. See United States v. United States District Court, *743407 U. S. 297, 317 (1972) (“The historical judgment, which the Fourth Amendment accepts, is that unreviewed executive discretion may yield too readily to pressures to obtain incriminating evidence and overlook potential invasions of privacy and protected speech”). Petitioners would have been forced to articulate their exact reasons for the search and to specify the items in Dr. Ortega’s office they sought, which would have prevented the general rummaging through the doctor’s office, desk, and file cabinets. Thus, because no “special need” in this case demanded that the traditional warrant and probable-cause requirements be dispensed with, petitioners’ failure to conduct the search in accordance with the traditional standard of reasonableness should end the analysis, and the judgment of the Court of Appeals should be affirmed.
B
Even were I to accept the proposition that this case presents a situation of “special need” calling for an exception to the warrant and probable-cause standard, I believe that the plurality’s balancing of the public employer’s and the employee’s respective interests to arrive at a different standard is seriously flawed. Once again, the plurality fails to focus on the facts. Instead, it arrives at its conclusion on the basis of “assumed” facts. First, sweeping with a broad brush, the plurality announces a rule that dispenses with the warrant requirement in every public employer’s search of an employee’s office, desk, or file cabinets because it “would seriously disrupt the routine conduct of business and would be unduly burdensome.” Ante, at 722. The plurality reasons that a government agency could not conduct its work in an efficient manner if an employer needed a warrant for every routine entry into an employee’s office in search of a file or correspondence, or for every investigation of suspected employee misconduct. In addition, it argues that the warrant requirement, if imposed on an employer who would be unfamiliar with this procedure, would prove “unwieldy.” Ibid.
*744The danger in formulating a standard on the basis of “assumed” facts becomes very clear at this stage of the plurality’s opinion. Whenever the Court has arrived at a standard of reasonableness other than the warrant and probable-cause requirements, it has first found, through analysis of a factual situation, that there is a nexus between this other standard, the employee’s privacy interests, and the government purposes to be served by the search. Put another way, the Court adopts a new standard only when it is satisfied that there is no alternative in the particular circumstances.8 In Terry v. Ohio, 392 U. S. 1, 20 (1968), the Court concluded that, as a practical matter, brief, on-the-spot stops of individuals by police officers need not be subject to a warrant. Still concerned, however, with the import of the warrant requirement, which provides the “neutral scrutiny of a judge,” id., at 21, the Court weighed in detail the law enforcement and the suspect’s interests in the circumstances of the protective search. The resulting standard constituted the equivalent of the warrant: judging the officer’s behavior from a reasonable or objective standard, id., at 21, 27. In Camara v. Municipal Court, 387 U. S. 523 (1967), on the other hand, the Court declined to abandon the warrant as a standard in the case of a municipal health inspection in fight of the interests of the target of the health investigation and those of the government in enforcing health standards. Id., at 532-533.
*745A careful balancing with respect to the warrant requirement is absent from the plurality’s opinion, an absence that is inevitable in light of the gulf between the plurality’s analysis and any concrete factual setting. It is certainly correct that a public employer cannot be expected to obtain a warrant for every routine entry into an employee’s workplace.9 This situation, however, should not justify dispensing with a warrant in all searches by the employer. The warrant requirement is perfectly suited for many work-related searches, including the instant one.10 Moreover, although the plurality abandons the warrant requirement, it does not explain what it will substitute or how the standard it adopts retains anything of the normal “neutral scrutiny of the judge.”11 In sum, the plurality’s general result is preordained because, cut off from a particular factual setting, it cannot make the necessary distinctions among types of searches, or formulate an alternative to the warrant requirement that derives from a precise weighing of competing interests.
*746When the plurality turns to the balancing that will produce an alternative to probable cause, it states that it is limiting its analysis to the two situations arguably presented by the facts of this case — the “noninvestigatory work-related intrusion” (i. e., inventory search) and the “investigatory search for evidence of suspected work-related employee misfeasance” (i. e., investigatory search). Ante, at 723. This limitation, however, is illusory. The plurality describes these searches in such a broad fashion that it is difficult to imagine a search that would not fit into one or the other of the categories. Moreover, it proposes the same standard, one taken from New Jersey v. T. L. 0., for both inventory and investigatory searches. See ante, at 725-726. Therefore, in the context of remanding a case because the facts are unclear, the plurality is announcing a standard to apply to all public employer searches.
Moreover, the plurality also abandons any effort at careful balancing in arriving at its substitute for probable cause. Just as the elimination of the warrant requirement requires some nexus between its absence, the employee’s privacy interests, and the government interests to be served by the search, so also does the formulation of a standard less than probable cause for a particular search demand a similar connection between these factors. See, e. g., United States v. Brignoni-Ponce, 422 U. S. 873, 881 (1975). The plurality’s discussion of investigatory searches reveals no attempt to set forth the appropriate nexus.12 It is certainly true, as the plurality observes, that a public employer has an interest in eliminating incompetence and work-related misconduct in order to enable the government agency to accomplish its tasks in an efficient manner. It is also conceivable that a public employee’s privacy interests are somewhat limited in the workplace, although, as noted above, not to the extent suggested by the plurality. The plurality, however, fails to *747explain why the balancing of these interests necessarily leads to the standard borrowed from New Jersey v. T. L. O., as opposed to other imaginable standards. Indeed, because the balancing is simply asserted rather than explicated,13 the plurality never really justifies why probable cause, characterized by this Court as a “practical, nontechnical conception,” Brinegar v. United States, 338 U. S. 160, 176 (1949), would not protect adequately the public employer’s interests in the situation presented by this case. See New Jersey v. T. L. O., 469 U. S., at 363-364 (Brennan, J., concurring in part and dissenting in part).14
*748I — I <1
I have reviewed at too great length the plurality’s opinion because the question of public employers’ searches of their employees’ workplaces, like any relatively unexplored area of Fourth Amendment law, demands careful analysis. These searches appear in various factual settings, some of which courts are only now beginning to face, and present different problems.15 Accordingly, I believe that the Court should examine closely the practical realities of a particular situation and the interests implicated there before replacing the traditional warrant and probable-cause requirements with some other standard of reasonableness derived from a balancing test. The Fourth Amendment demands no less. By ignoring the specific facts of this case, and by announcing in the abstract a standard as to the reasonableness of an employer’s workplace searches, the plurality undermines not only the Fourth Amendment rights of public employees but also any further analysis of the constitutionality of public employer searches.
I respectfully dissent.
Although there has been some development on these issues in federal courts, see ante, at 720-721, this Court has not yet squarely faced them.
It is true that this Court has expressed concern about the workability of “ ‘an ad hoe, ease-by-case definition of Fourth Amendment standards to be applied in differing factual circumstances.’” Ante, at 730 (Scalia, J., concurring in judgment), quoting Oliver v. United States, 466 U. S. 170, 181 (1984). Given, however, the number and types of workplace searches by public employers that can be imagined — ranging all the way from the employer’s routine entry for retrieval of a file to a planned investigatory search into an employee’s suspected criminal misdeeds — development of a jurisprudence in this area might well require a ease-by-case approach. See California v. Carney, 471 U. S. 386, 400 (1985) (Stevens, J., dissenting) (“The only true rules governing search and seizure have been formulated and refined in the painstaking scrutiny of ease-by-ease adjudication”); New Jersey v. T. L. O., 469 U. S. 325, 366-367 (1985) (BRENNAN, J., concurring in part and dissenting in part) (“I would not think it necessary to develop a single standard to govern all school searches, any more than traditional Fourth Amendment law applies even the probable-cause standard *734to all searches and seizures” (emphasis in original)). Under a case-by-case approach, a rule governing a particular type of workplace search, unlike the standard of the plurality here, should emerge from a concrete set of facts and possess the precision that only the exploration of “every aspect of a multifaced situation embracing conflicting and demanding interests” can produce. See United States v. Fruehauf, 365 U. S. 146, 157 (1961). The manner in which the plurality arrives at its standard, it seems to me, thus not only harms Dr. Ortega and other public employees, but also does a disservice to Fourth Amendment analysis.
It could be argued that the plurality removes its analysis from the facts of this case in order to arrive at a result unfavorable to public employees, whose position members of the plurality do not look upon with much sympathy. As Justice Cardozo long ago explained, judges are never free from the feelings of the times or those emerging from their own personal lives:
“I have spoken of the forces of which judges avowedly avail to shape the form and content of their judgments. Even these forces are seldom fully in consciousness. They lie so near the surface, however, that their existence and influence are not likely to be disclaimed. But the subject is not exhausted with the recognition of their power. Deep below consciousness are other forces, the likes and the dislikes, the predilections and the prejudices, the complex of instincts and emotions and habits and convictions, which make the man, whether he be litigant or judge.” B. Cardozo, The Nature of the Judicial Process 167 (1921).
It seems to me that whenever, as here, courts fail to concentrate on the facts of a case, these predilections inevitably surface, no longer held in check by the “discipline” of the facts, and shape, more than they ever should and even to an extent unknown to the judges themselves, any legal standard that is then articulated. This, I believe, is the central problem of the opinion of the plurality and, indeed, of the concurrence.
The plurality is correct in pointing out that the District Court erred in its conclusion that there was a Hospital policy that would have justified this search. Ante, at 728. This was not the only error on the District Court’s part. That court also concluded that Dr. Ortega was notified of the search and could have participated in it, see App. 23, a conclusion at odds with the record, see id., at 24, 40.
This common-sense notion that public employees have some expectation of privacy in the workplace, particularly with respect to private documents or papers kept there, was exemplified by recent remarks of the Attorney General. In responding to questions concerning the possibility of a search and seizure of papers and offices of Government employees in connection with an investigation into allegedly illegal diversion of funds to Central American recipients, he is reported to have stated: “I’m not sure we would have any opportunity or any legal right to get into those personal papers. . . . There was certainly no evidence of any criminality that would have supported a search warrant at that time. ... I don’t think public employees’ private documents belong to the Government.” N. Y. Times, Dec. 3, 1986, p. All, col. 3.
Moreover, courts have recognized that a public employee has a legitimate expectation of privacy as to an employer’s search and seizure at the workplace. See, e. g., Gillard v. Schmidt, 579 F. 2d 825, 829 (CA3 1978) (search of desk); United States v. McIntyre, 582 F. 2d 1221, 1224 (CA9 1978) (monitoring conversations at office desk). But see Williams v. Col*739lins, 728 F. 2d 721, 728 (CA5 1984) (search of desk). In some cases, courts have decided that an employee had no such expectation with respect to a workplace search because an established regulation permitted the search. See United States v. Speights, 557 F. 2d 362, 364-365 (CA3 1977) (describing cases); United States v. Donato, 269 F. Supp. 921 (ED Pa.), aff’d, 379 F. 2d 288 (CA3 1967) (Government regulation notified employees that lockers in the United States Mint were not to be viewed by employees as private lockers). The question of such a search pursuant to regulations is not now before this Court.
Perhaps the greatest sign of the disappearance of the distinction between work and private life is the fact that women — the traditional representatives of the private sphere and family life — have entered the *740work force in increasing numbers. See BNA Special Report, Work & Family: A Changing Dynamic, 1, 3, 13-15 (1986). It is therein noted:
“The myth of ‘separate worlds’ — one of work and the other of family lifelong harbored by employers, unions, and even workers themselves has been effectively laid to rest. Their inseparability is undeniable, particularly as two-earner families have become the norm where they once were the exception and as a distressing number of single parents are required to raise children on their own. The import of work-family conflicts — for the family, for the workplace, and, indeed, for the whole of society — will grow as these demographic and social transformations in the roles of men and women come to be more fully clarified and appreciated.” Id., at 217 (remarks of Professor Phyllis Moen).
As a result of this disappearance, moreover, the employee must attempt to maintain the difficult balance between work and personal life. Id., at 227 (remarks of Barney Olmsted and Suzanne Smith).
I am also troubled by the plurality’s implication that a public employee is entitled to a lesser degree of privacy in the workplace because the public agency, not the employee, owns much of what constitutes the workplace. This implication emerges in the distinction the plurality draws between the workplace “context,” which includes “the hallways, cafeteria, offices, desks, and file cabinets,” and an employee’s “closed personal luggage, a handbag, or a briefcase.” Ante, at 715-716. This Court, however, has made it clear that privacy interests protected by the Fourth Amendment do not turn on ownership of particular premises. See, e. g., Rakas v. Illinois, 439 U. S. 128, 143 (1978) (“[T]he protection of the Fourth Amendment depends not upon a property right in the invaded place but upon whether the person who claims the protection of the Amendment has a legitimate expectation of privacy in the invaded place”); Katz v. United States, 389 U. S. 347, 353 (1967) (Fourth Amendment protects people and not simply “areas”). To be sure, the public employer’s ownership of the premises is relevant in determining an employee’s expectation of privacy, for often it is the main reason for the routine visits into an employee’s office. The employee is assigned an office for work purposes; it is expected that the employee will receive work-related visitors and that the employer will main*741tain the office. This fact of ownership, however, like the routine visits, does not abrogate the employee’s expectation of privacy.
This part of the analysis is related to the “special need” step. Courts turn to the balancing test only when they conclude that the traditional warrant and probable-cause requirements are not a practical alternative. Through the balancing test, they then try to identify a standard of reasonableness, other than the traditional one, suitable for the circumstances. The warrant and probable-cause requirements, however, continue to serve as a model in the formulation of the new standard. It is conceivable, moreover, that a court, having initially decided that it is faced with a situation of “special need” that calls for balancing, may conclude after application of the balancing test that the traditional standard is a suitable one for the context after all.
In some workplace investigations, the particular goals of the government agency coupled with a need for special employee discipline may justify dispensing with the warrant requirement. See, e. g., Security and Law Enforcement Employees, Dist. Council 82, American Federation of State, County and Municipal Employees, AFL-CIO v. Carey, 737 F. 2d 187, 203-204 (CA2 1984) (government interest in maintaining security of a correctional facility justifies strip searches of correctional officers, in certain circumstances, in absence of a warrant).
While the warrant requirement might be “unwieldy” for public employers if it was required for every workplace search, the plurality has failed to explain why, on the facts of this case, obtaining a warrant would have been burdensome for petitioners, even if one assumes that they were unfamiliar with this requirement. In fact, the opposite seems true. Moreover, contrary to the plurality’s suggestion, see ante, at 722, the warrant requirement is not limited to the criminal context. See Camara v. Municipal Court, 387 U. S. 523, 530-531 (1967).
The plurality adopts a “standard of reasonableness under all the circumstances.” Ante, at 725-726. It fails completely to suggest how this standard captures any of the protection of the traditional warrant requirement; indeed, the standard appears to be simply an alternative to probable cause.
The same holds true for the plurality’s discussion of inventory searches.
The plurality’s attempt at explication consists of little more than a series of assertions: that the probable-cause requirement “would impose intolerable burdens on public employers”; that the delay caused by such a requirement would result in “tangible and often irreparable damage” to a government agency; and that public employers cannot be expected “to learn the subtleties of the probable cause standard.” See ante, at 724-725. Such assertions cannot pass for careful balancing on the facts of this case, given that the search was conducted during Dr. Ortega’s administrative leave from the Hospital, with the advice of counsel, and by an investigating party that included a security officer. My observation that a particular Fourth Amendment standard of reasonableness should be developed from a context bears here.
Even if I believed that this case were an appropriate vehicle for development of a standard on public-employer searches, I would fault the plurality for its failure to give much substance to the standard it has borrowed almost verbatim from New Jersey v. T. L. O. See ante, at 714-715. The T. L. O. Court described in some detail the substance of its test, which was tailored to the circumstances of the case before it and thus is not directly transferable from the halls of a high school to the offices of government. In any event, were I to apply the rather stark standard of reasonableness announced by the plurality, I would conclude that petitioners here did not satisfy it. Assuming, without deciding, that petitioners had an individualized suspicion that Dr. Ortega was mismanaging the psychiatric residency program, I believe the scope of the search was not reasonably related to this concern. If petitioners were truly in search of evidence of respondent’s mismanagement, it is difficult to understand why they looked through the personal belongings of Dr. Ortega, a search that resulted in the seizure of a Valentine’s Day card, a photograph, and a book of poetry, which could have no conceivable relation to the claimed purpose of the search. Although, in *748the plurality’s view, the seizure of these items is not an issue in this case, see ante, at 729, n., I would think that this seizure is relevant to determining the reasonableness of the scope of the search. Accordingly, under the plurality’s own standard, this search was unreasonable.
One example is the Fourth Amendment problem associated with drug and alcohol testing of employees. See, e. g., Shoemaker v. Handel, 795 F. 2d 1136, 1141-1143 (CA3) (administrative-search exception extended to warrantless breath and urine testing of jockeys, given the heavily regulated nature of the horse-racing industry), cert. denied, 479 U. S. 986 (1986); National Treasury Employees Union v. Von Raab, 649 F. Supp. 380 (ED La. 1986) (wide-scale urinalysis of United States Customs Service employees without probable cause or reasonable suspicion struck down as violative of the Fourth Amendment).