with whom BROWNING and ELY, Circuit Judges, join, dissenting:

The majority of the court and I agree on many points in the analysis of the legality of this search: The Fourth Amendment governs parole searches; older authorities excluding parolees from the Fourth Amendment’s protection against unreasonable searches and seizures are constitutionally defunct. 1 A parolee has a legitimate expectation of privacy, although his expectation is not the same as that of other citizens who are not enmeshed in the parole system. In his dealings with law enforcement personnel other than his parole officer, a parolee is entitled to the same expectation of privacy as a wholly free citizen, but the parolee cannot expect the same privacy bulwark to be erected between him and his parole officer. To exercise his responsibilities of supervision, a parole officer must have considerable latitude in observing the parolee in and out of his home and, under some circumstances, he must be able to search the parolee’s person and home. However, I cannot agree with the majority that a parole officer may search his parolee’s home without probable cause and without a warrant, subject only to the restriction that the ensuing search, from a hindsight view, does not appear unreasonable to parole authorities or to the courts when criminal proceedings are instituted against the parolee.

The majority offers three intertwined justifications for jettisoning the warrant requirement: (1) the probable cause component of the warrant procedure cannot be applied without reducing warrants to useless formalities or seriously impairing effective parole supervision; (2) the parole officer’s relationship to the parolee is an adequate substitute for a warrant because the officer’s intimate knowledge of the parolee, his training in rehabilitative techniques, his concern for the well-being of his charge, and his unusual intuitive capacity will deter unreasonable searches; and (3) even if a parole officer conducts an unreasonable search, subsequent review by higher parole authorities or by the courts will protect the parolee from the consequences of such a search and will deter future unreasonable searches.

The majority’s conclusion that applying the warrant requirement necessitates a choice between unduly impairing the concept of probable cause and unduly impairing the parole officer’s functions reflects a fundamental misunderstanding of probable cause. The concept of probable cause is not rigid. It is flexible enough to be adapted to parole searches to give the parolee meaningful protection and to preserve the functions of parole. If the question were whether to adapt the warrant requirement as the general rule, rather than as an exception to it, and if the only alternatives in enforcing the warrant procedure were those assumed, the majority’s rationale would be forceful, although not necessarily convincing; but that is not the question, and those are not the only choices.

A warrantless search is per se unreasonable, “ ‘subject only to a few specifically established and well-delineated exceptions.’ ” (Schneckloth v. Bustamonte (1973) 412 U.S. 218, 219 , 93 S.Ct. 2041, 2043 , 36 L.Ed.2d 854 ; United States v. United States District Court (1972) 407 U.S. 297, 315 , 92 S.Ct. 2125 , 32 L.Ed.2d 752 ; Camara v. Municipal Court (1967) 387 U.S. 523, 528-29 , 87 S.Ct. 1727 , 18 L.Ed.2d 930 ; Jones v. United States (1958) 357 U.S. 493, 499 , 78 S.Ct. 1253 , 2 L.Ed.2d 1514 .) 2 The majority fails to confront this basic proposition. It as *255 sumes that the test is whether the search itself is reasonable — an interpretation flatly rejected by the Supreme Court when it overturned United States v. Rabinowitz (1950) 339 U.S. 56 , 70 S.Ct. 430 , 94 L.Ed. 653 (United States v. United States District Court, supra, 407 U.S. at 315-16 , 92 S.Ct. 2125 ). The test underlying the few exceptions to the per se rule is whether it was reasonable to procure a search warrant.

Although the Fourth Amendment requires a warrant to search, it does not dictate that the probable cause necessary to justify issuance of a warrant to a parole officer to search his parolee’s home must be the same as that governing issuance of a warrant to a policeman to search a home. Probable cause, together with its evidentiary underpinnings, is not and cannot be inflexible. (Cf. Camara v. Municipal Court, supra; Almeida-Sanchez v. United States (1973) 413 U.S. 266, 283-84 , 93 S.Ct. 2535 , 37 L.Ed.2d 596 (Powell, J., concurring).) Conceptual rigidity would destroy the living force of the Fourth Amendment by disabling it from meeting new challenges that arise from changing conditions. 3

The organic quality of the probable cause requirement is reflected by the developing law controlling administrative searches. In Camara v. Municipal Court, supra, the Supreme Court adhered to the warrant requirement but declined to limit the showing of antecedent justification to probable cause to believe “that a particular dwelling contains violations of the minimum standards prescribed by the code being enforced” ( 387 U.S. at 534 , 87 S.Ct. at 1734 ), holding instead that probable cause could be established ’ on an area-wide basis when “reasonable legislative or administrative standards for conducting an area inspection are satisfied with respect to a particular dwelling. Such standards, which will vary with the municipal program being enforced, may be based upon the passage of time, the nature of the building . or the condition of the entire area, but they will not necessarily depend upon specific knowledge of the condition of the particular dwelling” ( 387 U.S. at 538 , 87 S.Ct. at 1736 ). The Court expressly rejected the contention that modification of the probable cause standard to meet the problem arising from this type of administrative search would authorize “ ‘a synthetic search warrant’ . If a valid public interest justifies the intrusion contemplated, then there is probable cause to issue a suitably restricted search warrant. . . . Such an approach neither endangers time-honored doctrines applicable to criminal investigations nor makes a nullity of the probable cause requirement in this area.” ( 387 U.S. at 538-39 , 87 S.Ct. at 1736 .)

The Supreme Court has permitted warrantless administrative searches in a single context under severely limiting circumstances. An administrative search may be made of business premises of licensed .dealers in liquor or firearms pursuant to statute or ordinance supported by sufficiently detailed regulations so far describing the dealer’s obligations and defining the inspector’s authority that neither inspector nor dealer is “left to wonder about the purposes of the inspector or the limits of his task.” (United States v. Biswell (1972) 406 U.S. 311, 316 , 92 S.Ct. 1593, 1596 , 32 L.Ed.2d 87 ; Colonnade Catering Corp. v. United States (1970) 397 U.S. 72 , 90 S.Ct. 774 , 25 L.Ed.2d 60 .)

The majority’s effort to squeeze parole searches into the Biswell-Colonnade mold and out of Camara-See’s, 4 distorts both sets of opinions. 5 No statutory authority for parole searches exists here, and there are no regulations of any kind defining *256 the limits of the parole officer’s authority or the scope of the searches he may make. The suggestion is untenable that appellate court decisions in the realm of search and seizure are an adequate substitute for the statutes and regulations that Biswell and Colonnade require. Appellate court decisions on points of search and seizure law are not susceptible to neat categorization, involving as they do endless twists of facts and circumstances viewed always retrospectively and almost always after a search has turned up something incriminating. The assumption is wholly unrealistic that appellate opinions, incorporated by reference into a parole officer’s hypothetical manual of standard operating procedures would provide any black-letter directions governing future interactions between parole officers and their parolees.

Wyman v. James (1971) 400 U.S. 309 , 91 S.Ct. 381 , 27 L.Ed.2d 408 sharply distinguishes home visits from searches; 6 it cannot be used to turn a search into a home visit. As Wyman takes pains to point out, Camara and See are “significantly different” because those cases “arose in a criminal context where a genuine search was denied and prosecution followed” ( 400 U.S. at 325 , 91 S.Ct. at 390 ). The instant case arose in a criminal context where a genuine search was made and prosecution followed.

Our decision in United States v. Davis (9th Cir. 1973) 482 F.2d 893 , upholding warrantless airport search programs, suggests a helpful analytical framework for determining whether an administrative search can be conducted without a warrant. The court took the following factors into account: (1) airport searches are conducted as part of “a general regulatory scheme in furtherance of an administrative purpose, rather than as part of a regular criminal investigation to secure evidence of crime” ( 482 F.2d at 908 ); (2) the decision to search a particular passenger’s luggage “is not ‘subject to the discretion of the official in the field’ ” ( 482 F.2d at 910 ); (3) a warrant requirement would effectively frustrate the governmental purpose behind the search; (4) the public has a very substantial interest in preventing skyjacking; and (5) a warrantless search is nevertheless unreasonable if it is unnecessarily intrusive (undue intrusiveness is avoided if a passenger can prevent a search by electing not to board the plane).

The parole system is a regulatory scheme embodying both rehabilitative and crime prevention components, in each of which the public has a very substantial interest. However, searches of the parolee’s home are not a necessary ingredient of every parolee’s supervision, nor are such searches the subject of promulgated regulations. The decision to search, absent a warrant, rests on the parole officer’s discretion. Unlike the airline passenger, the parolee is not given a choice whether to submit to a search. Undoubtedly there are occasions when a parole officer must search his charge’s home to fulfill the purposes of the parole system. But the majority opinion fails to explain (1) why the power to search should not be confined to those cases in which less intrusive means will not adequately serve the public interest in parole supervision, or (2) why the appropriate method of determining whether a search shall be made and how extensive it shall be should not be a warrant procedure. When probable cause is adapted to the special circumstances of parole, the warrant requirement fulfills constitutional mandates without unreasonably restricting parole supervision.

A warrant should issue to a parole officer to search his parolee’s residence upon the officer’s showing that the described home to be searched is the residence of his parolee, in which he lives alone or in the company of persons identified or otherwise described; that the *257 parole officer has reasonable cause to believe that the parolee is violating, or is in imminent danger of violating, one or more specified conditions of his parole; and that he has reasonable cause to believe that evidence of such actual or impending violations will be found in the home to be searched. Evidentiary support for the probable cause showing need not meet the high standards of AguilarSpinelli, 7 but it could not be based on the officer’s hunches unsupported by articulated facts. Rather, the standard should be sufficiently flexible to accommodate the parole officer’s supervisory obligations, but not so loose as to offer the parolee and his family no protection from arbitrary intrusions by the parole officer or from searches that are unjustifiably broad. To this end, the officer’s showing need not be confined to evidence admissible in a courtroom. It could even include information from others whose reliability had not been tested.

In deciding whether to issue the warrant and in defining its terms, the magistrate would take into account the strength of the showing of reasonable cause and such additional factors as the nature of the parole violations suspected, the extent to which persons other than the parolee would have their privacy invaded by the search, and the existence of means less intrusive than the search to meet the parole officer’s supervisory responsibilities.

The issuance of a warrant on this kind of showing is no “paper tiger”; 8 neither is it an undue burden on the functioning of parole. Even if we were to assume, as does the majority, that “in most cases the magistrate would have to take the parole’s officer’s word for it,” 9 which I interpret as referring to the factual recitals of the parole officer, the protection afforded by the warrant is by no means negligible. The requirement that an officer articulate his reasons for making a search before he searches is a substantial deterrent to impulsive and arbitrary official conduct and a real safeguard against after-the-fact justifications. (See Green-berg, “The Balance of Interests Theory and the Fourth Amendment,” 61 Calif. L.Rev. 1011, 1025 n.60 (1973).) 10

Moreover, the assumption is dubious at best that magistrates “in most cases” will exercise no independent judgment in issuing search warrants to parole officers. A neutral magistrate is not and should not be assumed to be an automaton. “ ‘[T]he procedure of antecedent justification ... is central to the Fourth Amendment’ ” (Katz v. United States (1967) 389 U.S. 347, 359 , 88 S.Ct. 507, 515 , 19 L.Ed.2d 576 ). “The Fourth Amendment contemplates a prior judicial judgment, not the risk that executive discretion may be reasonably exercised” (United States v. United States District Court, supra, 407 U.S. at 317 , 92 S.Ct. at 2137 (footnote omitted).)

Post search reviews of reasonableness neither deter unreasonable searches nor remedy those that have occurred. In all but the most egregious cases, the searching officers will be able retrospectively to point to specific facts that justified the search. (Beck v. Ohio (1964) 379 U.S. 89, 96 , 85 S.Ct. 223 , 13 L.Ed.2d 142 ; see United States v. United States District Court, supra, 407 U.S. at 317—18, 92 S.Ct. 2125 .) Unreasonable searches that yield nothing incriminating do not sur *258 face in parole review proceedings or in the criminal courts, and it will be a very brave or very foolhardy parolee who attempts to vindicate his Fourth Amendment rights by suing his parole officer pursuant to 42 U.S.C. § 1983 and Bivens v. Six Unknown Named Agents of Federal Bureau of Investigation (1971) 403 U.S. 388 , 91 S.Ct. 1999 , 29 L.Ed.2d 619 -even if he has the financial wherewithal to do so. 11 Furthermore, review by parole authorities or criminal courts provide no protection and no solace to the parolee’s family and friends whose privacy is invaded by unreasonable searches of the parolee’s home or temporary abode, which they may share with him.

The majority’s reliance upon the expertise of parole officers and their supposed knowledge of the standards of reasonableness developed in numerous decisions of appellate courts is misplaced. Of course, there are parole officers who maintain ideal relationships with parolees and who are well trained in the rehabilitative arts. There may even be parole officers who keep up with advance sheets. But surely they are not representative of the mine run of parole officers. “In a number of parole systems, too many parole officers still see their major role as that of policeman-enforcer.” (National Advisory Commission on Criminal Justice Standards and Goals, Corrections 434 (1973).) Parole systems are more often than not characterized by inadequate training programs and by individual workloads so burdensome that parole officers, no matter how well trained and dedicated, cannot function effectively. (Id. at 435.) To the extent that there is a check on unreasonable searches, it is more attributable to case overload than to a regard for Fourth Amendment values; parole officers sím-ply do not have enough time to search many of their parolees’ homes. 12

The rehabilitative goals of parole would be advanced, not impeded, by a warrant requirement. As the majority concedes, indiscriminate searches undermine the rehabilitative process. The Constitution places primary responsibility for preventing indiscriminate searches on the warrant requirement and not, as the majority concludes, on the parole authorities or the courts in later criminal proceedings.

The warrant requirement does not deprive the parole officer of any legitimate tool that he needs to practice his craft. He needs no warrant to visit his charge’s home. (Cf. Wyman v. James, supra.) If during the course of a home visit and without a search a parole officer observes evidence leading him reasonably to believe that his parolee is in violation of the conditions of parole, he may seize that evidence. His observations during a home visit could provide probable cause to search the residence forthwith, if the circumstances appear exigent. In short, the parole officer has all the leeway to conduct a warrantless search that is accorded a police officer, with the substantial advantage that the parole officer, unlike a policeman, can gain warrantless entry for visiting purposes.

The need for a warrant is vividly demonstrated by the facts of the case before us. At the time the parole officer, accompanied by a police officer, searched the parolee’s residence, the parolee was in custody. Violation of his parole had theretofore been established. No exigent circumstances existed justifying any immediate search. The only purpose served was to gather evidence of crime. The search invaded the privacy of the *259 parolee’s family as well as that of the parolee. No function of parole would have been impeded or embarrassed by requiring the parole officer to procure a warrant. The search falls squarely within the Fourth Amendment prohibition of unreasonable searches.

The majority opinion says that post-search review of the reasonableness of the search is an adequate substitute for a warrant. The unsoundness of that conclusion is illustrated by the majority’s application of its rule to uphold the search in this case. The search would have been patently unreasonable, as the majority tacitly concedes, if it had been made by a lone policeman. It becomes reasonable solely by virtue of the fact that it was made by a parole officer, accompanied by a police officer. The unarticulated majority rule is that all searches of a parolee’s home by his parole officer are reasonable unless the particular search later is deemed to have been harassing, or intimidating, or too overblown. In short, the majority creates a presumption of reasonableness in respect of searches by parole officers, to be dispelled, if ever, upon subsequent review of a particular search. Under a warrant requirement, the burden would be on the government official to establish the reasonableness of the prospective search. Under the majority’s rule, the parolee has the burden of establishing the unreasonableness of the search. Since the parolee would have to establish unreasonableness after a search has presumably turned up something incriminating, his burden is practically insurmountable. How can the majority’s reasoning be squared with its conclusion that its reasonableness standard is an acceptable substitute for a warrant? The impact of the majority’s rule is to obliterate Fourth Amendment protections of a parolee from unreasonable searches and seizures by his parole officer.

The warrant requirement must not be unjustifiably and easily cast aside. We should be ever mindful of the truth and wisdom of Mr. Justice Frankfurter’s observation: “The history of liberty has largely been the history of observance of procedural safeguards.” (McNabb v. United States (1943) 318 U.S. 332, 347 , 63 S.Ct. 608, 616 , 87 L.Ed. 819 .)

I would reverse.

. Op’n, Duniway, J., at 248-249.

. For a synoptical discussion of exceptions to the warrant requirement, see Amsterdam, “Perspectives on the Fourth Amendment,” 58 Minn.L.Rev. 349, 358-60 (1974) and accompanying notes.

. Parole is one such changing condition. Parole began over a century after the Fourth Amendment was drafted, when in 1876 the Elmira Reformatory in New York commenced the practice. (National Advisory Commission on Criminal Justice Standards and Goals, Corrections 391 (1973).)

. See v. Seattle (1967) 387 U.S. 541 , 87 S.Ct. 1737 , 18 L.Ed.2d 943 .

. The limitations of the Biswell-Colonnade exception to the warrant requirement are sharply drawn by Almeida-Sanchez v. United States (1973) 413 U.S. 266, 270-72 , 93 S.Ct. 2535 , 37 L.Ed.2d 596 .

. “This natural and quite proper protective attitude, however, is not a factor in this case, for the seemingly obvious and simple reason that we are not concerned here with any search by the New York social service agency in the Fourth Amendment meaning of that term.” (Wyman v. James, supra, 400 U.S. at 317 , 91 S.Ct. at 386 .)

. Aguilar v. Texas (1964) 378 U.S. 108 , 84 S.Ct. 1509 , 12 L.Ed.2d 723 ; Spinelli v. United States (1969) 393 U.S. 410 , 89 S.Ct. 584 , 21 L.Ed.2d 637 .

. To the majority’s charge that I would create a “paper tiger” (ante at 252), I might reply that the majority’s sentinel of privacy is a Cheshire kitten that dematerializes with the first glimmer of a parole officer’s incipient hunch. But the differences between us are much deeper than a metaphorical debate.

. Op’n, Duniway, J., at 252.

. “Harassment is not confined to inspections motivated solely by personal spite and bad faith, but includes overzealous enforcement. The latter problem is considerably more acute . [where] one of the main aims of the inspection scheme is to ‘serve as a credible deterrent.’ Government use of the search power to deter, unchecked by neutral inquiry into the necessity for a given level of deterrence, creates the possibility of substantial abuse, since it is undeniable that the greater the governmental abuse, the greater the deterrent effect.” (Greenberg, supra, 61 Calif.L. Rev. at 1023 (footnote omitted).)

. Cf. Amsterdam, supra note 2, 58 Minn.L. Rev. at 429 n. 30.

. The warrant procedure would mitigate the possibility that a particular search would be “based on caprice or on personal spite” (Ohio ex rel. Eaton v. Price (1960) 364 U.S. 263, 271 , 80 S.Ct. 1463, 1468 , 4 L.Ed.2d 1708 (Brennan, J.)) or that “the administrative officer who invades the privacy of the home may be only a front for the police who are thus saved the nuisance of getting a warrant” (Abel v. United States (1960) 362 U.S. 217, 242 , 80 S.Ct. 683, 699 , 4 L.Ed.2d 668 (Douglas, J., dissenting)).