concurring in part and dissenting in part.
I fully agree with Part II of the Court’s opinion. Teachers, like all other government officials, must conform their *354conduct to the Fourth Amendment’s protections of personal privacy and personal security. As Justice Stevens points out, post, at 373-374, 385-386, this principle is of particular importance when applied to schoolteachers, for children learn as much by example as by exposition. It would be incongruous and futile to charge teachers with the task of embuing their students with an understanding of our system of constitutional democracy, while at the same time immunizing those same teachers from the need to respect constitutional protections. See Board of Education v. Pico, 457 U. S. 853, 864-865 (1982) (plurality opinion); West Virginia State Board of Education v. Barnette, 319 U. S. 624, 637 (1943).
I do not, however, otherwise join the Court’s opinion. Today’s decision sanctions school officials to conduct full-scale searches on a “reasonableness” standard whose only definite content is that it is not the same test as the “probable cause” standard found in the text of the Fourth Amendment. In adopting this unclear, unprecedented, and unnecessary departure from generally applicable Fourth Amendment standards, the Court carves out a broad exception to standards that this Court has developed over years of considering Fourth Amendment problems. Its decision is supported neither by precedent nor even by a fair application of the “balancing test” it proclaims in this very opinion.
I
Three basic principles underly this Court’s Fourth Amendment jurisprudence. First, warrantless searches are per se unreasonable, subject only to a few specifically delineated and well-recognized exceptions. See, e. g., Katz v. United States, 389 U. S. 347, 357 (1967); accord, Welsh v. Wisconsin, 466 U. S. 740, 748-749 (1984); United States v. Place, 462 U. S. 696, 701 (1983); Steagald v. United States, 451 U. S. 204, 211-212 (1981); Mincey v. Arizona, 437 U. S. 385 (1978); Terry v. Ohio, 392 U. S. 1, 20 (1968); Johnson v. United States, 333 U. S. 10, 13-14 (1948). Second, full-scale searches — whether conducted in accordance with the war*355rant requirement or pursuant to one of its exceptions — are “reasonable” in Fourth Amendment terms only on a showing of probable cause to believe that a crime has been committed and that evidence of the crime will be found in the place to be searched. Beck v. Ohio, 379 U. S. 89, 91 (1964); Wong Sun v. United States, 371 U. S. 471, 479 (1963); Brinegar v. United States, 338 U. S. 160, 175-176 (1949). Third, categories of intrusions that are substantially less intrusive than full-scale searches or seizures may be justifiable in accordance with a balancing test even absent a warrant or probable cause, provided that the balancing test used gives sufficient weight to the privacy interests that will be infringed. Dunaway v. New York, 442 U. S. 200, 210 (1979); Terry v. Ohio, supra.
Assistant Vice Principal Choplick’s thorough excavation of T. L. O.’s purse was undoubtedly a serious intrusion on her privacy. Unlike the searches in Terry v. Ohio, supra, or Adams v. Williams, 407 U. S. 143 (1972), the search at issue here encompassed a detailed and minute examination of respondent’s pocketbook, in which the contents of private papers and letters were thoroughly scrutinized.1 Wisely, neither petitioner nor the Court today attempts to justify the search of T. L. O.’s pocketbook as a minimally intrusive search in the Terry line. To be faithful to the Court’s settled doctrine, the inquiry therefore must focus on the warrant and probable-cause requirements.
A
I agree that schoolteachers or principals, when not acting as agents of law enforcement authorities, generally may conduct a search of their students’ belongings without first *356obtaining a warrant. To agree with the Court on this point is to say that school searches may justifiably be held to that extent to constitute an exception to the Fourth Amendment’s warrant requirement. Such an exception, however, is not to be justified, as the Court apparently holds, by assessing net social value through application of an unguided “balancing test” in which “the individual’s legitimate expectations of privacy and personal security” are weighed against “the government’s need for effective methods to deal with breaches of public order.” Ante, at 337. The Warrant Clause is something more than an exhortation to this Court to maximize social welfare as we see fit. It requires that the authorities must obtain a warrant before conducting a full-scale search. The undifferentiated governmental interest in law enforcement is insufficient to justify an exception to the warrant requirement. Rather, some special governmental interest beyond the need merely to apprehend lawbreakers is necessary to justify a categorical exception to the warrant requirement. For the most part, special governmental needs sufficient to override the warrant requirement flow from “exigency” — that is, from the press of time that makes obtaining a warrant either impossible or hopelessly infeasible. See United States v. Place, supra, at 701-702; Mincey v. Arizona, supra, at 393-394; Johnson v. United States, supra, at 15. Only after finding an extraordinary governmental interest of this kind do we — or ought we — engage in a balancing test to determine if a warrant should nonetheless be required.2
*357To require a showing of some extraordinary governmental interest before dispensing with the warrant requirement is not to undervalue society’s need to apprehend violators of the criminal law. To be sure, forcing law enforcement personnel to obtain a warrant before engaging in a search will predictably deter the police from conducting some searches that they would otherwise like to conduct. But this is not an unintended result of the Fourth Amendment’s protection of privacy; rather, it is the very purpose for which the Amendment was thought necessary. Only where the governmental interests at stake exceed those implicated in any ordinary law enforcement context — that is, only where there is some extraordinary governmental interest involved — is it legitimate to engage in a balancing test to determine whether a warrant is indeed necessary.
In this case, such extraordinary governmental interests do exist and are sufficient to justify an exception to the warrant requirement. Students are necessarily confined for most of the schoolday in close proximity to each other and to the school staff. I agree with the Court that we can take judicial notice of the serious problems of drugs and violence that plague our schools. As Justice Blackmun notes, teachers must not merely “maintain an environment conducive to learning” among children who “are inclined to test the outer boundaries of acceptable conduct,” but must also “protect the very safety of students and school personnel.” Ante, at 352-353. A teacher or principal could neither carry out essential teaching functions nor adequately protect students’ safety if required to wait for a warrant before conducting a necessary search.
B
I emphatically disagree with the Court’s decision to east aside the constitutional probable-cause standard when assessing the constitutional validity of a schoolhouse search. The Court’s decision jettisons the probable-cause standard— the only standard that finds support in the text of the Fourth *358Amendment — on the basis of its Rohrschach-like “balancing test.” Use of such a “balancing test” to determine the standard for evaluating the validity of a full-scale search represents a sizable innovation in Fourth Amendment analysis. This innovation finds support neither in precedent nor policy and portends a dangerous weakening of the purpose of the Fourth Amendment to protect the privacy and security of our citizens. Moreover, even if this Court’s historic understanding of the Fourth Amendment were mistaken and a balancing test of some kind were appropriate, any such test that gave adequate weight to the privacy and security interests protected by the Fourth Amendment would not reach the preordained result the Court's conclusory analysis reaches today. Therefore, because I believe that the balancing test used by the Court today is flawed both in its inception and in its execution, I respectfully dissent.
1
An unbroken line of cases in this Court have held that probable cause is a prerequisite for a full-scale search. In Carroll v. United States, 267 U. S. 132, 149 (1925), the Court held that “[o]n reason and authority the true rule is that if the search and seizure . . . are made upon probable cause .. . the search and seizure are valid.” Under our past decisions probable cause — which exists where “the facts and circumstances within [the officials’] knowledge and of which they had reasonably trustworthy information [are] sufficient in themselves to warrant a man of reasonable caution in the belief” that a criminal offense had occurred and the evidence would be found in the suspected place, id., at 162 — is the constitutional minimum for justifying a full-scale search, regardless of whether it is conducted pursuant to a warrant or, as in Carroll, within one of the exceptions to the warrant requirement. Henry v. United States, 361 U. S. 98, 104 (1959) (Carroll “merely relaxed the requirements for a warrant on grounds of practicality,” but “did not dispense *359with the need for probable cause”); accord, Chambers v. Maroney, 399 U. S. 42, 51 (1970) (“In enforcing the Fourth Amendment’s prohibition against unreasonable searches and seizures, the Court has insisted upon probable cause as a minimum requirement for a reasonable search permitted by the Constitution”).3
Our holdings that probable cause is a prerequisite to a full-scale search are based on the relationship between the two Clauses of the Fourth Amendment. The first Clause (“The right of the people to be secure in their persons, houses, papers and effects, against unreasonable searches and seizures, shall not be violated . . .”) states the purpose of the Amendment and its coverage. The second Clause (“. . . and no Warrants shall issue but upon probable cause . . .”) gives content to the word “unreasonable” in the first Clause. “For all but . . . narrowly defined intrusions, the requisite ‘balancing’ has been performed in centuries of precedent and is embodied in the principle that seizures are ‘reasonable’ only if supported by probable cause.” Dunaway v. New York, 442 U. S., at 214.
I therefore fully agree with the Court that “the underlying command of the Fourth Amendment is always that searches and seizures be reasonable.” Ante, at 337. But this “underlying command” is not directly interpreted in each category of cases by some amorphous “balancing test.” Rather, the provisions of the Warrant Clause — a warrant and probable cause — provide the yardstick against which official searches *360and seizures are to be measured. The Fourth Amendment neither requires nor authorizes the conceptual free-for-all that ensues when an unguided balancing test is used to assess specific categories of searches. If the search in question is more than a minimally intrusive Terry stop, the constitutional probable-cause standard determines its validity.
To be sure, the Court recognizes that probable cause “ordinarily” is required to justify a full-scale search and that the existence of probable cause “bears on” the validity of the search. Ante, at 340-341. Yet the Court fails to cite any case in which a full-scale intrusion upon privacy interests has been justified on less than probable cause. The line of cases begun by Terry v. Ohio, 392 U. S. 1 (1968), provides no support, for they applied a balancing test only in the context of minimally intrusive searches that served crucial law enforcement interests. The search in Terry itself, for instance, was a “limited search of the outer clothing.” Id., at 30. The type of border stop at issue in United States v. Brignoni-Ponce, 422 U. S. 873, 880 (1975), usually “consume[d] no more than a minute”; the Court explicitly noted that “any further detention . . . must be based on consent or probable cause.” Id., at 882. See also United States v. Hensley, ante, at 224 (momentary stop); United States v. Place, 462 U. S., at 706-707 (brief detention of luggage for canine “sniff”); Pennsylvania v. Mimms, 434 U. S. 106 (1977) (per curiam) (brief frisk after stop for traffic violation); United States v. Martinez-Fuerte, 428 U. S. 543, 560 (1976) (characterizing intrusion as “minimal”); Adams v. Williams, 407 U. S. 143 (1972) (stop and frisk). In short, all of these cases involved “‘seizures’ so substantially less intrusive than arrests that the general rule requiring probable cause to make Fourth Amendment ‘seizures’ reasonable could be replaced by a balancing test.” Dunaway, supra, at 210.
Nor do the “administrative search” cases provide any comfort for the Court. In Camara v. Municipal Court, 387 U. S. 523 (1967), the Court held that the probable-cause standard governed even administrative searches. Although *361the Camara Court recognized that probable-cause standards themselves may have to be somewhat modified to take into account the special nature of administrative searches, the Court did so only after noting that “because [housing code] inspections are neither personal in nature nor aimed at the discovery of evidence of crime, they involve a relatively limited invasion of the urban citizen’s privacy.” Id., at 537. Subsequent administrative search cases have similarly recognized that such searches intrude upon areas whose owners harbor a significantly decreased expectation of privacy, see, e. g., Donovan v. Dewey, 452 U. S. 594, 598-599 (1981), thus circumscribing the injury to Fourth Amendment interests caused by the search.
Considerations of the deepest significance for the freedom of our citizens counsel strict adherence to the principle that no search may be conducted where the official is not in possession of probable cause — that is, where the official does not know of “facts and circumstances [that] warrant a prudent man in believing that the offense has been committed.” Henry v. United States, 361 U. S., at 102; see also id., at 100-101 (discussing history of probable-cause standard). The Fourth Amendment was designed not merely to protect against official intrusions whose social utility was less as measured by some “balancing test” than its intrusion on individual privacy; it was designed in addition to grant the individual a zone of privacy whose protections could be breached only where the “reasonable” requirements of the probable-cause standard were met. Moved by whatever momentary evil has aroused their fears, officials — perhaps even supported by a majority of citizens — may be tempted to conduct searches that sacrifice the liberty of each citizen to assuage the perceived evil.4 But the Fourth Amendment *362rests on the principle that a true balance between the individual and society depends on the recognition of “the right to be let alone — the most comprehensive of rights and the right most valued by civilized men.” Olmstead v. United States, 277 U. S. 438, 478 (1928) (Brandéis, J., dissenting). That right protects the privacy and security of the individual unless the authorities can cross a specific threshold of need, designated by the term “probable cause.” I cannot agree with the Court’s assertions today that a “balancing test” can replace the constitutional threshold with one that is more convenient for those enforcing the laws but less protective of the citizens’ liberty; the Fourth Amendment’s protections should not be defaced by “a balancing process that overwhelms the individual’s protection against unwarranted official intrusion by a governmental interest said to justify the search and seizure.” United States v. Martinez-Fuerte, supra, at 570 (Brennan, J., dissenting).
2
I thus do not accept the majority’s premise that “[t]o hold that the Fourth Amendment applies to searches conducted by school authorities is only to begin the inquiry into the standards governing such searches.” Ante, at 337. For me, the finding that the Fourth Amendment applies, coupled with the observation that what is at issue is a full-scale search, is the end of the inquiry. But even if I believed that a “balancing test” appropriately replaces the judgment of the Framers of the Fourth Amendment, I would nonetheless object to the cursory and shortsighted “test” that the Court employs to justify its predictable weakening of Fourth Amendment protections. In particular, the test employed by the Court vastly overstates the social costs that a probable-cause standard entails and, though it plausibly articulates the serious privacy interests at stake, inexplicably fails to accord them adequate weight in striking the balance.
*363The Court begins to articulate its “balancing test” by-observing that “the government’s need for effective methods to deal with breaches of public order” is to be weighed on one side of the balance. Ibid. Of course, this is not correct. It is not the government’s need for effective enforcement methods that should weigh in the balance, for ordinary Fourth Amendment standards — including probable cause— may well permit methods for maintaining the public order that are perfectly effective. If that were the case, the governmental interest in having effective standards would carry no weight at all as a justification for departing from the probable-cause standard. Rather, it is the costs of applying probable cause as opposed to applying some lesser standard that should be weighed on the government’s side.5
In order to tote up the costs of applying the probable-cause standard, it is thus necessary first to take into account the nature and content of that standard, and the likelihood that it would hamper achievement of the goal — vital not just to “teachers and administrators,” see ante, at 339 — of maintaining an effective educational setting in the public schools. The seminal statement concerning the nature of the probable-cause standard is found in Carroll v. United States, 267 U. S. 132 (1925). Carroll held that law enforcement authorities have probable cause to search where “the facts and circumstances within their knowledge and of which they had reasonably trustworthy information [are] sufficient in themselves to *364warrant a man of reasonable caution in the belief” that a criminal offense had occurred. Id., at 162. In Brinegar v. United States, 338 U. S. 160 (1949), the Court amplified this requirement, holding that probable cause depends upon “the factual and practical considerations of everyday life on which reasonable and prudent men, not legal technicians, act.” Id., at 175.
Two Terms ago, in Illinois v. Gates, 462 U. S. 213 (1983), this Court expounded at some length its view of the probable-cause standard. Among the adjectives used to describe the standard were “practical,” “fluid,” “flexible,” “easily applied,” and “nontechnical.” See id., at 232, 236, 239. The probable-cause standard was to be seen as a “common-sense” test whose application depended on an evaluation of the “totality of the circumstances.” Id., at 238.
Ignoring what Gates took such great pains to emphasize, the Court today holds that a new “reasonableness” standard is appropriate because it “will spare teachers and school administrators the necessity of schooling themselves in the niceties of probable cause and permit them to regulate their conduct according to the dictates of reason and common sense.” Ante, at 343. I had never thought that our pre-Gates understanding of probable cause defied either reason or common sense. But after Gates, I would have thought that there could be no doubt that this “nontechnical,” “practical,” and “easily applied” concept was eminently serviceable in a context like a school, where teachers require the flexibility to respond quickly and decisively to emergencies.
A consideration of the likely operation of the probable-cause standard reinforces this conclusion. Discussing the issue of school searches, Professor LaFave has noted that the cases that have reached the appellate courts “strongly suggest that in most instances the evidence of wrongdoing prompting teachers or principals to conduct searches is sufficiently detailed and specific to meet the traditional probable cause test.” 3 W. LaFave, Search and Seizure § 10.11, *365pp. 459-460 (1978).6 The problems that have caused this Court difficulty in interpreting the probable-cause standard have largely involved informants, see, e. g., Illinois v. Gates, supra; Spinelli v. United States, 393 U. S. 410 (1969); Aguilar v. Texas, 378 U. S. 108 (1964); Draper v. United States, 358 U. S. 307 (1959). However, three factors make it likely that problems involving informants will not make it difficult for teachers and school administrators to make probable-cause decisions. This Court’s decision in Gates applying a “totality of the circumstances” test to determine whether an informant’s tip can constitute probable cause renders the test easy for teachers to apply. The fact that students and teachers interact daily in the school building makes it more likely that teachers will get to know students who supply information; the problem of informants who remain anonymous even to the teachers — and who are therefore unavailable for verification or further questioning — is unlikely to arise. Finally, teachers can observe the behavior of students under suspicion to corroborate any doubtful tips they do receive.
As compared with the relative ease with which teachers can apply the probable-cause standard, the amorphous “reasonableness under all the circumstances” standard freshly coined by the Court today will likely spawn increased litigation and greater uncertainty among teachers and administrators. Of course, as this Court should know, an essential purpose of developing and articulating legal norms is to enable individuals to conform their conduct to those norms. A school system conscientiously attempting to obey the Fourth Amendment’s dictates under a probable-cause standard could, for example, consult decisions and other legal materials and prepare a booklet expounding the rough outlines of the concept. Such a booklet could be distributed to *366teachers to provide them with guidance as to when a search may be lawfully conducted. I cannot but believe that the same school system faced with interpreting what is permitted under the Court’s new “reasonableness” standard would be hopelessly adrift as to when a search may be permissible. The sad result of this uncertainty may well be that some teachers will be reluctant to conduct searches that are fully permissible and even necessary under the constitutional probable-cause standard, while others may intrude arbitrarily and unjustifiably on the privacy of students.7
One further point should be taken into account when considering the desirability of replacing the constitutional probable-cause standard. The question facing the Court is not whether the probable-cause standard should be replaced by a test of “reasonableness under all the circumstances.” Rather, it is whether traditional Fourth Amendment standards should recede before the Court’s new standard. Thus, although the Court today paints with a broad brush and holds its undefined “reasonableness” standard applicable to all school searches, I would approach the question with considerably more reserve. I would not think it necessary to develop a single standard to govern all school searches, any more *367than traditional Fourth Amendment law applies even the probable-cause standard to all searches and seizures. For instance, just as police officers may conduct a brief stop and frisk on something less than probable cause, so too should teachers be permitted the same flexibility. A teacher or administrator who had reasonable suspicion that a student was carrying a gun would no doubt have authority under ordinary Fourth Amendment doctrine to conduct a limited search of the student to determine whether the threat was genuine. The “costs” of applying the traditional probable-cause standard must therefore be discounted by the fact that, where additional flexibility is necessary and where the intrusion is minor, traditional Fourth Amendment jurisprudence itself displaces probable cause when it determines the validity of a search.
A legitimate balancing test whose function was something more substantial than reaching a predetermined conclusion acceptable to this Court’s impressions of what authority teachers need would therefore reach rather a different result than that reached by the Court today. On one side of the balance would be the costs of applying traditional Fourth Amendment standards — the “practical” and “flexible” probable-cause standard where a full-scale intrusion is sought, a lesser standard in situations where the intrusion is much less severe and the need for greater authority compelling. Whatever costs were toted up on this side would have to be discounted by the costs of applying an unprecedented and ill-defined “reasonableness under all the circumstances” test that will leave teachers and administrators uncertain as to their authority and will encourage excessive fact-based litigation.
On the other side of the balance would be the serious privacy interests of the student, interests that the Court admirably articulates in its opinion, ante, at 337-339, but which the Court’s new ambiguous standard places in serious jeopardy. I have no doubt that a fair assessment of the two *368sides of the balance would necessarily reach the same conclusion that, as I have argued above, the Fourth Amendment’s language compels — that school searches like that conducted in this case are valid only if supported by probable cause.
I — { 1 — I
Applying the constitutional probable-cause standard to the facts of this case, I would find that Mr. Choplick’s search violated T. L. O.’s Fourth Amendment rights. After escorting T. L. O. into his private office, Mr. Choplick demanded to see her purse. He then opened the purse to find evidence of whether she had been smoking in the bathroom. When he opened the purse, he discovered the pack of cigarettes. At this point, his search for evidence of the smoking violation was complete.
Mr. Choplick then noticed, below the cigarettes, a pack of cigarette rolling papers. Believing that such papers were “associated,” see ante, at 328, with the use of marihuana, he proceeded to conduct a detailed examination of the contents of her purse, in which he found some marihuana, a pipe, some money, an index card, and some private letters indicating that T. L. O. had sold marihuana to other students. The State sought to introduce this latter material in evidence at a criminal proceeding, and the issue before the Court is whether it should have been suppressed.
On my view of the case, we need not decide whether the initial search conducted by Mr. Choplick — the search for evidence of the smoking violation that was completed when Mr. Choplick found the pack of cigarettes — was valid. For Mr. Choplick at that point did not have probable cause to continue to rummage through T. L. O.’s purse. Mr. Choplick’s suspicion of marihuana possession at this time was based solely on the presence of the package of cigarette papers. The mere presence without more of such a staple item of commerce is insufficient to warrant a person of reasonable caution in inferring both that T. L. O. had violated the law *369by possessing marihuana and that evidence of that violation would be found in her purse. Just as a police officer could not obtain a warrant to search a home based solely on his claim that he had seen a package of cigarette papers in that home, Mr. Choplick was not entitled to search possibly the most private possessions of T. L. O. based on the mere presence of a package of cigarette papers. Therefore, the fruits of this illegal search must be excluded and the judgment of the New Jersey Supreme Court affirmed.
h-H H — H
In the past several Terms, this Court has produced a succession of Fourth Amendment opinions in which “balancing tests” have been applied to resolve various questions concerning the proper scope of official searches. The Court has begun to apply a “balancing test” to determine whether a particular category of searches intrudes upon expectations of privacy that merit Fourth Amendment protection. See Hudson v. Palmer, 468 U. S. 517, 527 (1984) (“Determining whether an expectation of privacy is ‘legitimate’ or ‘reasonable’ necessarily entails a balancing of interests”). It applies a “balancing test” to determine whether a warrant is necessary to conduct a search. See ante, at 340; United States v. Martinez-Fuerte, 428 U. S., at 564-566. In today’s opinion, it employs a “balancing test” to determine what standard should govern the constitutionality of a given category of searches. See ante, at 340-341. Should a search turn out to be unreasonable after application of all of these “balancing tests,” the Court then applies an additional “balancing test” to decide whether the evidence resulting from the search must be excluded. See United States v. Leon, 468 U. S. 897 (1984).
All of these “balancing tests” amount to brief nods by the Court in the direction of a neutral utilitarian calculus while the Court in fact engages in an unanalyzed exercise of judicial will. Perhaps this doctrinally destructive nihilism is merely *370a convenient umbrella under which a majority that cannot agree on a genuine rationale can conceal its differences. Compare ante, p. 327 (White, J., delivering the opinion of the Court), with ante, p. 348 (Powell, J., joined by O’Connor, J., concurring), and ante, p. 351 (Blackmun, J., concurring in judgment). And it may be that the real force underlying today’s decision is the belief that the Court purports to reject — the belief that the unique role served by the schools justifies an exception to the Fourth Amendment on their behalf. If so, the methodology of today’s decision may turn out to have as little influence in future cases as will its result, and the Court’s departure from traditional Fourth Amendment doctrine will be confined to the schools.
On my view, the presence of the word “unreasonable” in the text of the Fourth Amendment does not grant a shifting majority of this Court the authority to answer all Fourth Amendment questions by consulting its momentary vision of the social good. Full-scale searches unaccompanied by probable cause violate the Fourth Amendment. I do not pretend that our traditional Fourth Amendment doctrine automatically answers all of the difficult legal questions that occasionally arise. I do contend, however, that this Court has an obligation to provide some coherent framework to resolve such questions on the basis of more than a conclusory recitation of the results of a “balancing test.” The Fourth Amendment itself supplies that framework and, because the Court today fails to heed its message, I must respectfully dissent.
A purse typically contains items of highly personal nature. Especially for shy or sensitive adolescents, it could prove extremely embarrassing for a teacher or principal to rummage through its contents, which could include notes from friends, fragments of love poems, caricatures of school authorities, and items of personal hygiene.
Administrative search cases involving inspection schemes have recognized that “if inspection is to be effective and serve as a credible deterrent, unannounced, even frequent, inspections are essential. In this context, the prerequisite of a warrant could easily frustrate inspection . . . .” United States v. Biswell, 406 U. S. 311, 316 (1972); accord, Donovan v. Dewey, 452 U. S. 594, 603 (1981). Cf. Marshall v. Barlow’s, Inc., 436 U. S. 307 (1978) (holding that a warrant is nonetheless necessary in some administrative search contexts).
In fact, despite the somewhat diminished expectation of privacy that this Court has recognized in the automobile context, see South Dakota v. Opperman, 428 U. S. 364, 367-368 (1976), we have required probable cause even to justify a warrantless automobile search, see United States v. Ortiz, 422 U. S. 891, 896 (1975) (“A search, even of an automobile, is a substantial invasion of privacy. To protect that privacy from official arbitrariness, the Court always has regarded probable cause as the minimum requirement for a lawful search”) (footnote omitted); Chambers v. Maroney, 399 U. S., at 51.
As Justice Stewart said in Coolidge v. New Hampshire, 403 U. S. 443, 455 (1971): “In times of unrest, whether caused by crime or racial conflict or fear of internal subversion, this basic law and the values that it represents may appear unrealistic or ‘extravagant’ to some. But the values were those of the authors of our fundamental constitutional concepts.”
I speak of the “government’s side” only because it is the terminology used by the Court. In my view, this terminology itself is seriously misleading. The government is charged with protecting the privacy and security of the citizen, just as it is charged with apprehending those who violate the criminal law. Consequently, the government has no legitimate interest in conducting a search that unduly intrudes on the privacy and security of the citizen. The balance is not between the rights of the government and the rights of the citizen, but between opposing conceptions of the constitutionally legitimate means of carrying out the government’s varied responsibilities.
It should be noted that Professor LaFave reached this conclusion in 1978, before this Court’s decision in Gates made clear the “flexibility” of the probable-cause concept.
A comparison of the language of the standard (“reasonableness under all the circumstances”) with the traditional language of probable cause (“facts sufficient to warrant a person of reasonable caution in believing that a crime had been committed and the evidence would be found in the designated place”) suggests that the Court’s new standard may turn out to be probable cause under a new guise. If so, the additional uncertainty caused by this Court’s innovation is surely unjustifiable; it would be naive to expect that the addition of this extra dose of uncertainty would do anything other than “burden the efforts of school authorities to maintain order in their schools,” ante, at 342. If, on the other hand, the new standard permits searches of students in instances when probable cause is absent— instances, according to this Court’s consistent formulations, when a person of reasonable caution would not think it likely that a violation existed or that evidence of that violation would be found — the new standard is genuinely objectionable and impossible to square with the premise that our citizens have the right to be free from arbitrary intrusions on their privacy.