delivered the opinion of the Court.

We granted certiorari to decide whether pretrial detainees have a right guaranteed by the United States Constitution to contact visits and to observe shakedown searches of their cells by prison officials.

I

Los Angeles County Central Jail is one of seven principal facilities operated by the Sheriff of Los Angeles County. The three-story jail complex, located in downtown Los Ange-les, is the largest jail in the country, with a capacity of over 5,000 inmates. It is the primary facility in Los Angeles County for male pretrial detainees, the vast majority of *578whom remain at the facility at most a few days or weeks while they await trial.

In 1975, respondents, pretrial detainees at Central Jail, brought a class action under 42 U. S. C. §§1983, 1985, against the County Sheriff, certain administrators of Central Jail, and the County Board of Supervisors, challenging various policies and practices of the jail and conditions of their confinement. Only respondents’ challenges to the policy of the jail denying pretrial detainees contact visits with their spouses, relatives, children, and friends, and to the jail’s practice of permitting irregularly scheduled shakedown searches of individual cells in the absence of the cell occupants are before this Court.1 The District Court sustained both of these challenges. Rutherford v. Pitchess, 457 F. Supp. 104 (CD Cal. 1978).

The District Court agreed with respondents that “the ability of a man to embrace his wife and his children from time to time during the weeks or months while he is awaiting trial is a matter of great importance to him,” id., at 110, yet it recognized that “unrestricted contact visitation would add greatly” to security problems at the jail. Ibid. The court ultimately concluded, however, that the danger of permitting low security risk inmates to have “physical contact with their loved ones” was not sufficiently great to warrant deprivation of such contact. Ibid. Striking what it believed was a “reasonable balance” between the twin considerations of prison *579security and the constitutional rights of the inmates, the court tentatively proposed to order contact visitation for those inmates who “have received other than high risk classification,” and who have been incarcerated for more than two weeks. Ibid.

With respect to the cell searches, the District Court concluded that allowing inmates to watch from a distance while their cells are searched would allay inmate concerns that their personal property will be unnecessarily confiscated or destroyed. The court concluded that “[fjuture shakedowns should be made while the respective inmates remain outside their cells but near enough to observe the process and raise or answer any relevant inquiry.” Id., at 116. The District Court viewed both of its proposed orders as “the least restrictive alternatives consistent with the purpose of [respondents’] incarceration.” Id., at 108.

The District Court withheld judgment on all of respondents’ complaints pending further evidentiary hearings. In its supplemental memorandum following the additional hearings, the court acknowledged that “many factors strongly militate against the allowing of contact visits,” App. to Pet. for Cert. 32, not the least of which being that “establishment of any program of contact visits [would] increase the importation of narcotics into [the] jail, despite all safeguards and precautions.” Id., at 31. The court again emphasized that if all or most of the inmates were allowed contact visits, a “great burden” would be imposed on the jail authorities and the public. Ibid. Modification of existing visiting areas, if not additional facilities, would be necessary. New procedures for processing visitors — possibly including interviews, personal searches, and searches of all packages carried by the visitors — would be required. Strip searches of inmates following contact visits would be needed.

The court found that the “hardship” on detainees of being unable to “embrace their loved ones” for only a few days or a few weeks could not justify imposition of these substantial *580burdens. Id., at 32. However, the court believed, the factors rendering contact visitation impracticable for detainees incarcerated for short periods are considerably less compelling when detention is prolonged.

The court reasoned that “the scope, burden and dangers of [a] program [of contact visitation] would be substantially diminished” were contact visitation limited to detainees “who have been in uninterrupted custody for a month or more and who are not determined to be drug oriented or escape risks,” and a ceiling imposed on the total number of contact visits that the jail must provide. Id., at 33 (emphasis added). With these limitations, the court suggested, a contact visitation program would require only “[m]odest alteration” of the existing facility. Ibid. Alternatively, the court said, the Sheriff could build or occupy a new facility for contact visits and transport inmates back and forth, as necessary.

The District Court also reaffirmed in the supplemental memorandum its earlier conclusion that inmates should be allowed to observe cell searches. The court believed that the interests of the inmates “in protecting their meager possessions outweigh[ed] the small increase in the burden upon the [petitioners].” Id., at 36.2

On appeal the Court of Appeals for the Ninth Circuit remanded the case to the District Court for consideration in light of our intervening decision in Bell v. Wolfish, 441 U. S. *581520 (1979), noting, among other things, that we rejected in Wolfish the suggestion that existence of less restrictive means for achievement of security objectives is proof of an exaggerated response to security concerns. App. to Pet. for Cert. 17.

The District Court on remand reaffirmed its prior orders, “[finding] nothing in Bell v. Wolfish that rendered] inappropriate any of the . . . challenged orders.” Id., at 24. Although the court acknowledged that the Central Jail authorities were not “consciously motivated by a desire to punish,” it reiterated its belief that the practices and policies in question were “excessive” in relation to the underlying security objectives. Id., at 25. It characterized petitioners’ rejection of all proposals for contact visitation as an “overreaction,” id., at 26, which “stem[med] from an unreasonable fixation upon security,” id., at 25.

The District Court conceded that Wolfish invalidated a similar order requiring that detainees be allowed to observe searches of their cells, but it went on to identify several factors that it thought distinguished its order from that in Wolfish.3

On petitioner’s second appeal, the Court of Appeals affirmed the District Court’s orders requiring that certain of the detainees be allowed contact visits and that inmates be allowed to watch searches of their cells.4 Rutherford v. *582Pitchess, 710 F. 2d 572 (1983). The Court of Appeals held that the District Court’s order on contact visitation “fits harmoniously within [the] pattern” of federal cases following Wolfish “recognizing] the important security interests of the [penal] institution but at the same time recognizing] the psychological and punitive effects which the prolonged loss of contact visitation has upon detainees . . . 710 F. 2d, at 577. It suggested that a blanket prohibition of contact visits for all detainees would be an “unreasonable, exaggerated response to security concerns.” Ibid.

The Court of Appeals also rejected petitioners’ contention that Wolfish precluded an order that pretrial detainees be permitted to observe cell searches. The Court of Appeals, as had the District Court, identified “significant differences” between the order invalidated in Wolfish and that entered by the District Court.5

We granted certiorari because of both the importance of the issue to the administration of detention facilities and the conflict among the Federal Courts of Appeals.6 464 U. S. 959 (1983). We reverse.

*583II

The administration of seven separate jail facilities for a metropolitan area of more than seven million people is a task of monumental proportions. Housed in these facilities annually are 200,000 persons awaiting trial and confined because they are unable to meet the requirements for release on bail. Generalizations are of little value, but no one familiar with even the barest outline of the problems of the administration of a prison or jail, or with the administration of criminal justice, could fail to be aware of the ease with which one can obtain release on bail or personal recognizance. The very fact of nonrelease pending trial thus is a significant factor bearing on the security measures that are imperative to proper administration of a detention facilitv.

Four Terms ago, in Bell v. Wolfish, 441 U. S. 520 (1979), we considered for the first time, in light of these security concerns, the scope of constitutional protection that must be accorded pretrial detainees. The respondents in Wolfish challenged numerous conditions of their confinement at the pretrial detention facility in New York City and various policies and practices of that institution. We held that, where it is alleged that a pretrial detainee has been deprived of liberty without due process, the dispositive inquiry is whether the challenged condition, practice, or policy constitutes punishment, “[fjor under the Due Process Clause, a detainee must not be punished prior to an adjudication of guilt in accordance with due process of law.” Id., at 535 (footnote omitted).

*584In addressing the particular challenges in Wolfish, we carefully outlined the principles to be applied in evaluating the constitutionality of conditions of pretrial detention. Specifically, we observed that “[a] court must decide whether the disability is imposed for the purpose of punishment or whether it is but an incident of some other legitimate governmental purpose.” Id., at 538 (citation omitted). Absent proof of intent to punish, we noted, this determination “generally will turn on ‘whether an alternative purpose to which [the restriction] may rationally be connected is assignable for it, and whether it appears excessive in relation to the alternative purpose assigned [to it].’” Ibid, (quoting Kennedy v. Mendoza-Martinez, 372 U. S. 144, 168-169 (1963)). We concluded:

“[I]f a particular condition or restriction of pretrial detention is reasonably related to a legitimate governmental objective, it does not, without more, amount to ‘punishment.’ Conversely, if a restriction or condition is not reasonably related to a legitimate goal — if it is arbitrary or purposeless — a court permissibly may infer that the purpose of the governmental action is punishment that may not constitutionally be inflicted upon detainees qua detainees.” 441 U. S., at 539 (footnote and citation omitted).

In setting forth these guidelines, we reaffirmed the very limited role that courts should play in the administration of detention facilities. In assessing whether a specific restriction is “reasonably related” to security interests, we said, courts should

“heed our warning that ‘[s]uch considerations are peculiarly within the province and professional expertise of corrections officials, and, in the absence of substantial evidence in the record to indicate that the officials have exaggerated their response to these considerations courts should ordinarily defer to their expert judgment *585in such matters.’” Id., at 540-541, n. 23 (quoting Pell v. Procunier, 417 U. S. 817, 827 (1974)).

We also cautioned:

“[P]rison administrators [are to be] accorded wide-ranging deference in the adoption and execution of policies and practices that in their judgment are needed to preserve internal order and discipline and to maintain institutional security.” 441 U. S. at 547 (citing cases).

The principles articulated in Wolfish govern resolution of this case.

Ill

A

Petitioners’ first contention is that it was error to conclude that even low risk detainees incarcerated for more than a month are constitutionally entitled to contact visits from friends and relatives. Petitioners maintain, as they have throughout these proceedings that, in the interest of institutional and public security, it is within their discretion as officials of a detention facility to impose an absolute prohibition on contact visits.7 The District Court did not find, nor did the Court of Appeals suggest, that the purpose of petitioners’ policy of denying contact visitation is to punish the inmates. To the contrary, the District Court found that petitioners are *586fully cognizant of the possible value of contact visitation, and it commended petitioners for their conscientious efforts to accommodate the large numbers of inmates at Central Jail.

The question before us, therefore, is narrow: whether the prohibition of contact visits is reasonably related to legitimate governmental objectives. More particularly, because there is no dispute that internal security of detention facilities is a legitimate governmental interest,8 our inquiry is simply whether petitioners’ blanket prohibition on contact visits at Central Jail is reasonably related to the security of that facility.

That there is a valid, rational connection between a ban on contact visits and internal security of a detention facility is too obvious to warrant extended discussion. The District Court acknowledged as much. Contact visits invite a host of security problems. They open the institution to the introduction of drugs, weapons, and other contraband. Visitors can easily conceal guns, knives, drugs, or other contraband in countless ways and pass them to an inmate unnoticed by even the most vigilant observers. And these items can readily be slipped from the clothing of an innocent child, or transferred by other visitors permitted close contact with inmates.

Contact visitation poses other dangers for a detention facility, as well. Detainees — by definition persons unable to meet bail — often are awaiting trial for serious, violent offenses, and many have prior criminal convictions. Exposure of this type person to others, whether family, friends, or jail administrators, necessarily carries with it risks that the safety of innocent individuals will be jeopardized in various ways. They *587may, for example, be taken as hostages or become innocent pawns in escape attempts. It is no answer, of course, that we deal here with restrictions on pretrial detainees rather than convicted criminals. For, as we observed in Wolfish, in this context, “[t]here is no basis for concluding that pretrial detainees pose any lesser security risk than convicted inmates.” 441 U. S., at 546, n. 28. Indeed, we said, “it may be that in certain circumstances [detainees] present a greater risk to jail security and order.” Ibid.

The District Court and Court of Appeals held that totally disallowing contact visits is excessive in relation to the security and other interests at stake. We reject this characterization. There are many justifications for denying contact visits entirely, rather than attempting the difficult task of establishing a program of limited visitation such as that imposed here. It is not unreasonable to assume, for instance, that low security risk detainees would be enlisted to help obtain contraband or weapons by their fellow inmates who are denied contact visits. Additionally, identification of those inmates who have propensities for violence, escape, or drug smuggling is a difficult if not impossible task, and the chances of mistaken identification are substantial. The burdens of identifying candidates for contact visitation — glossed over by the District Court — are made even more difficult by the brevity of detention and the constantly changing nature of the inmate population. Or a complete prohibition could reasonably be thought necessary because selectively allowing contact visits to some — even if feasible — could well create tension between those allowed contact visits and those not.

In Wolfish, we sustained against a Fourth Amendment challenge the practice of conducting routine body cavity searches following contact visits, even though there had been only one reported attempt to smuggle contraband into the facility in a body cavity. 441 U. S., at 558-560. The purpose of the cavity searches in Wolfish was to discover and deter smuggling of weapons and contraband, which was found to be *588a byproduct of contact visits. Given the security demands and the need to protect not only other inmates but also the facility’s personnel, we did not regard full body cavity searches as excessive. Petitioners’ flat prohibition on contact visits cannot be considered a more excessive response to the same security objectives. See id., at 559-560, n. 40. In any event, we have emphasized that we are unwilling to substitute our judgment on these difficult and sensitive matters of institutional administration and security for that of “the persons who are actually charged with and trained in the running,” id., at 562, of such facilities.9 In sum, we conclude that petitioners’ blanket prohibition is an entirely reasonable, nonpunitive response to the legitimate security concerns identified, consistent with the Fourteenth Amendment.

The District Court acknowledged that “many factors strongly militate against the allowing of contact visits.” App. to Pet. for Cert. 32. The court appears to have accepted petitioners’ testimony that contact visits significantly increase the possibility that there will be breaches of security and that the safety of others will be placed in jeopardy. It noted that, “despite all safeguards and precautions,” id., at 31, any program of contact visitation would inevitably increase importation of narcotics into the jail. We can take judicial notice that the unauthorized use of narcotics is a problem that plagues virtually every penal and detention *589center in the country. While explicitly acknowledging the security risks that inhere in even a limited program of contact visitation, the District Court nonetheless invalidated petitioners’ practice of denying contact visitation.

On this record, we must conclude that the District Court simply misperceived the limited scope of judicial inquiry under Wolfish. When the District Court found that many factors counseled against contact visits, its inquiry should have ended. The court’s further “balancing” resulted in an impermissible substitution of its view on the proper administration of Central Jail for that of the experienced administrators of that facility. Here, as in Wolfish, “[i]t is plain from [the] opinions that the lower courts simply disagreed with the judgment of [the jail] officials about the extent of the security interests affected and the means required to further those interests.” 441 U. S., at 554.

In rejecting the District Court’s order, we do not in any sense denigrate the importance of visits from family or friends to the detainee. Nor do we intend to suggest that contact visits might not be a factor contributing to the ultimate reintegration of the detainee into society. We hold only that the Constitution does not require that detainees be allowed contact visits when responsible, experienced administrators have determined, in their sound discretion, that such visits will jeopardize the security of the facility.

B

It has been the petitioners’ practice, as it is of all such facilities, to conduct irregular or random “shakedown” searches of the cells of detainees while the detainees are away at meals, recreation, or other activities. Respondents do not dispute the need for these searches; they challenge the searches only to the extent that detainees are not permitted to observe them.

Petitioners respond that their method of conducting cell searches is a security measure virtually identical to that chai-*590lenged in Wolfish. See 441 U. S., at 555-557.10 We agree. The Court described the practice in Wolfish as follows:

“The MCC staff conducts unannounced searches of inmate living areas at irregular intervals. These searches generally are formal unit ‘shakedowns’ during which all inmates are cleared of the residential units, and a team of guards searches each room. . . . [IJnmates [are] not permitted to watch the searches.” Id., at 555.

The search practices described are essentially identical to those employed at Central Jail, see n. 1, supra.

Respondents attempt to distinguish Wolfish principally on the ground that the District Court’s order invalidated in Wolfish rested on the Fourth Amendment, while the District Court’s order here was predicated on its holding that searches in the absence of the detainees violate their rights under the Due Process Clause of the Fourteenth Amendment. We did hold in Wolfish that the room search rule challenged did not violate the Fourth Amendment. But we also explicitly rejected the contention that the room search rule, including the feature of the rule prohibiting observation of the searches by the detainees, violated the detainees’ due process rights:

“Nor do we think that the four MCC security restrictions and practices described in Part III, supra [one of which was the rule permitting room searches in the absence of the detainees] constitute ‘punishment’ in viola*591tion of the rights of pretrial detainees under the Due Process Clause of the Fifth Amendment.” 441U. S., at 560-561 (footnote omitted).

We held that all of the restrictions “were reasonable responses by [the] officials to legitimate security concerns.” Id., at 561.

Thus, contrary to respondents’ suggestion, we have previously considered not only a Fourth Amendment challenge but also a due process challenge to a room search procedure almost identical to that used at Central Jail, and we sustained the practice on both scores. We have no reason to reconsider that issue; the identical arguments made by respondents here were advanced by the respondents in Wolfish. The security concerns that we held justified the same restriction in Wolfish, see id., at 555, n. 36, are no less compelling here.11 Moreover, we could not have been clearer in our holding in Wolfish that this is a matter lodged in the sound discretion of the institutional officials. We reaffirm that “proper deference to the informed discretion of prison authorities demands that they, and not the courts, make the difficult judgments which reconcile conflicting claims affecting the security of the institution, the welfare of the prison staff, and the property rights of the detainees.” Id., at 557, n. 38.12

*592Accordingly, the judgment of the Court of Appeals is reversed.

It is so ordered.

concurring in the judgment.

I agree with the Court that neither the contact-visitation policy nor the cell-search policy at issue in this case violates respondents’ due process rights under the Fourteenth Amendment. I write separately, however, because I do not believe that the Court adequately has addressed the gravamen of respondents’ constitutional claims.

1. I disagree with the Court’s treatment of the contact-visitation issue chiefly because, in my view, the Court has invoked principles of judicial deference to administrative judgment that have no place in the present litigation. As the Court made clear in Bell v. Wolfish, 441 U. S. 520 (1979), and as it reaffirms here, a pretrial detainee who challenges conditions of confinement on the ground that they amount to punishment in violation of the Due Process Clause must show that the conditions are the product of punitive intent. See id., at 538-539, and nn. 19 and 20, and ante, at 584. When a detainee attempts to demonstrate the existence of punitive intent, either through direct proof of motive or through a demonstration that the challenged conditions are not “reasonably related to a legitimate governmental objective,” 441 U. S., at 539, he necessarily is calling into question the good faith of prison administrators. Under those circumstances, it seems to me to be somewhat perverse to insist that a court assessing the rationality of a particular administrative prac*593tice must accord prison administrators “ ‘wide-ranging deference in the adoption and execution of policies and practices that in their judgment are needed to preserve internal order and discipline and to maintain institutional security.’ ” Ante, at 585, quoting Bell v. Wolfish, 441 U. S., at 547. Such a requirement boils down to a command that when a court is confronted with a charge of administrative bad faith, it must evaluate the charge by assuming administrative good faith.

When a constitutional challenge to prison conditions necessarily places the good faith of prison administrators at issue, I regard it as improper to make the plaintiff prove his case twice by requiring a court to defer to administrators’ putative professional judgment. Instead, I think it sufficient to rest on the substantive due process standard announced in Bell v. Wolfish itself: absent direct proof of punitive intent, “a court permissibly may infer that the purpose of [a challenged] governmental action is punishment” if, but only if, the action “is not reasonably related to a legitimate goal.” Id., at 539. The requirement that prison policies be reasonably related to a legitimate goal is hardly a stringent one, and, for many of the reasons given by the Court, ante, at 586-587, I have no doubt that the requirement has been met on the record presented here. I therefore am mystified by the Court’s insistence on invoking principles of judicial deference, since those principles are not only inappropriate but entirely unnecessary to the result in this case.

More generally, I am concerned about the Court’s apparent willingness to substitute the rhetoric of judicial deference for meaningful scrutiny of constitutional claims in the prison setting. See Rhodes v. Chapman, 452 U. S. 337, 369 (1981) (opinion concurring in judgment). Courts unquestionably should be reluctant to second-guess prison administrators’ opinions about the need for security measures; when constitutional standards look in whole or in part to the effectiveness of administrative practices, good-faith administrative *594judgments are entitled to substantial weight. The fact that particular measures advance prison security, however, does not make them ipso facto constitutional. Cf. Bell v. Wolfish, 441 U. S., at 539, n. 20. I recognize that constitutional challenges to prison conditions, like similarly expansive challenges to the workings of other institutions, pose a danger of excessive judicial intervention. At the same time, however, careless invocations of “deference” run the risk of returning us to the passivity of several decades ago, when the then-prevailing barbarism and squalor of many prisons were met with a judicial blind eye and a “hands off” approach. As we recognized in Bell v. Wolfish, the fact that initial responsibility for the Nation’s prisons is vested in prison administrators “does not mean that constitutional rights are not to be scrupulously observed.” Id., at 562. It is only because I am satisfied that the contact-visitation policy satisfies this standard under the Due Process Clause that I join the Court’s judgment.

2. The Court’s treatment of the cell-search policy misconstrues respondents’ claim. The Court assumes that respondents are challenging their exclusion from cell searches on substantive due process grounds and hence that the decision in Bell v. Wolfish is dispositive. Ante, at 590-591. It is quite clear, however, that respondents are challenging the cell-search policy on procedural due process grounds. See Tr. of Oral Arg. 38 (“[Tjhis is a procedural due process issue . . . rather than [an issue of] freedom from punishment as a matter of substantive due process”); Brief for Respondents 33-36. In essence, respondents are arguing that cell searches result in the deprivation of their personal property and that the process due them under the Fourteenth Amendment includes an opportunity to observe cell searches in order to minimize erroneous deprivations. Because the Court did not address a procedural due process claim in Bell v. Wolfish, something more must be said to support the judgment in this case.

*595Under Mathews v. Eldridge, 424 U. S. 319 (1976), the adequacy of governmental procedures that accompany deprivations of property normally depends on a balance of three factors: the private interest that will be affected by the official action, the risk that the existing procedures will result in an erroneous deprivation and the probable value of additional procedural safeguards, and the governmental interest in relying on the challenged procedures. Id., at 335. Here, I do not dispute that the private interests at stake in cell searches are potentially significant. See Hudson v. Palmer, ante, at 521, 524-525 (opinion of the Court), 553-554 (Stevens, J., concurring in part and dissenting in part). Nor is it possible to maintain that a pretrial detainee’s presence never would contribute to the avoidance of erroneous deprivations. We noted in Bell v. Wolfish that the “prevention of] theft or misuse by those conducting the search” was a “conceivable beneficial effect” of allowing detainees to observe cell searches, 441 U. S., at 557, and the District Judge in this case witnessed a search in which a prisoner was able to prevent the mistaken seizure of two magazines from his cell by explaining why they complied with prison regulations. App. to Pet. for Cert. 36.

The countervailing governmental interests in conducting cell searches outside the presence of pretrial detainees, however, are substantial enough to persuade me that the Court of Appeals erred in its due process determination. First, there is no reason to think that “friction between the inmates and security guards,” Bell v. Wolfish, 441 U. S., at 555, is any less likely to result from the presence of detainees here than it was in Bell v. Wolfish itself. Second, and more significant, detainees may well learn where to hide contraband if they are allowed to watch searches of their cells. As a result, although the requirement of a detainee’s presence during the course of a search may not prevent the seizure of contraband during the search itself, cf. Calero-Toledo v. Pearson Yacht Leasing Co., 416 U. S. 663, 679 (1974), it may *596frustrate future searches of the same detainee’s cell. Quite apart from security concerns, moreover, there undoubtedly are “administrative burdens [entailed by] the additional or substitute procedural requirement.” Mathews v. Eldridge, 424 U. S., at 335. For example, as petitioners point out, the jail now would be required to dedicate an increased number of guards to the task of accompanying each detainee from a holding area to his cell while the search is being conducted. Just as different exigencies have excused the requirement of predeprivation hearings in other contexts, see, e. g., Commissioner v. Shapiro, 424 U. S. 614, 629-630, and n. 12 (1976); Calero-Toledo, 416 U. S., at 676-680; Phillips v. Commissioner, 283 U. S. 589, 596-597 (1931); North American Cold Storage Co. v. Chicago, 211 U. S. 306, 315-316 (1908), so do these considerations tip the balance against a defacto predeprivation “hearing” for pretrial detainees here. It is for this reason that I join the judgment of the Court.

with whom Justice Brennan and Justice Stevens join, dissenting.

This case marks the fourth time in recent years that the Court has turned a deaf ear to inmates’ claims that the conditions of their confinement violate the Federal Constitution. See Rhodes v. Chapman, 452 U. S. 337 (1981); Bell v. Wolfish, 441 U. S. 520 (1979); Hudson v. Palmer, ante, p. 517. Guided by an unwarranted confidence in the good faith and “expertise” of prison administrators and by a pinched conception of the meaning of the Due Process Clauses and the Eighth Amendment, a majority of the Court increasingly appears willing to sanction any prison condition for which the majority can imagine a colorable rationale, no matter how oppressive or ill-justified that condition is in fact. So, here, the Court upholds two policies in force at the Los Angeles County Central Jail. Under one, a pretrial detainee is not permitted any physical contact with members of his family, *597regardless of how long he is incarcerated pending his trial or how slight is the risk that he will abuse a visitation privilege. Under the other, detainees are not allowed to observe searches of their cells, despite the fact that such searches frequently result in arbitrary destruction or confiscation of the detainees’ property. In my view, neither of these policies comports with the Constitution.

t — (

In Bell v. Wolfish, supra, the Court established a set of principles defining constitutionally permissible treatment of incarcerated persons who have not been convicted of crimes. In the years since Wolfish, I have not abandoned my view that the Court’s decision in that case was fundamentally misconceived. See 441 U. S., at 563-579 (Marshall, J., dissenting). However, even if I thought the doctrine enunciated in Wolfish was defensible, I could not abide the manner in which the majority construes and applies that doctrine to dispose of respondents’ challenge to the jail’s rule against contact visitation.

One of the premises of the principal holding in Wolfish was that the plaintiffs’ claims did not implicate any “fundamental liberty interests” such as those “delineated in . . . Roe v. Wade, 410 U. S. 113 (1973); Eisenstadt v. Baird, 405 U. S. 438 (1972); Stanley v. Illinois, 405 U. S. 645 (1972); Griswold v. Connecticut, 381 U. S. 479 (1965); [and] Meyer v. Nebraska, 262 U. S. 390 (1923).” Id., at 534-535. Aside from the right not to be punished prior to adjudication of guilt, the only general interest that could be asserted by the plaintiffs in Wolfish, the Court contended, was a “desire to be free from discomfort.” Id., at 534.1 The comparatively un*598important nature of that interest made it possible for the Court to adopt a deferential legal standard: “[A] particular condition or restriction of pretrial detention” passes muster under the Due Process Clause as long as it “is reasonably related to a legitimate governmental objective,” id., at 539.

The Court today reiterates and relies on the foregoing test. Ante, at 586. In so doing, however, the Court ignores a crucial difference between the interests at stake in Wolfish and in this case. Unlike the Wolfish plaintiffs, respondents can and do point to a fundamental right abridged by the jail’s policy — namely, their freedom to engage in and prevent the deterioration of their relationships with their families.

The importance of the right asserted by respondents was acknowledged by the District Court. “[T]he ability of a man to embrace his wife and his children from time to time during the weeks or months while he is awaiting trial,” the court found, “is a matter of great importance to him.” Rutherford v. Pitehess, 457 F. Supp. 104, 110 (1978).2 Denial of contact visitation, the court concluded, is “very traumatic treatment.” App. to Pet. for Cert. 25. Substantial evidence in the record supports the District Court’s findings. William Nagel, an expert in the field of corrections, testified that contact visitation was crucial in allowing prisoners to maintain their familial bonds. Tr. 4174-4175. Similarly, Dr. Terry Kupers, a psychiatrist, testified that denial of contact visitation contributes to the breakup of prisoners’ marriages and generally threatens their mental health. Id., at 4647-4651. *599The secondary literature buttresses these assertions,3 as do the conclusions reached by other courts.4

The significant injury to familial relations wrought by the jail’s policy of denying contact visitation means that that policy must be tested against a legal standard more constraining than the rule announced in Wolfish. Our cases leave no doubt that persons’ freedom to enter into, maintain, and cultivate familial relations is entitled to constitutional protection. E. g., Santosky v. Kramer, 455 U. S. 745, 753 (1982). Among the relationships that we have expressly shielded from state interference are bonds between spouses, see Zablocki v. Redhail, 434 U. S. 374 (1978), and between parents and their children, see Wisconsin v. Yoder, 406 U. S. 205 (1972); Stanley v. Illinois, supra. The special status of these relationships in our constitutional scheme derives from several considerations: the fact that traditionally they have been regarded as sacrosanct,5 the important role they have played in fostering diversity and pluralism in our culture,6 and their centrality to the emotional life of many persons.7

Determination of exactly how the doctrine established in the aforementioned cases bears upon a ban on contact visita*600tion by pretrial detainees would be difficult. On the one hand, it could be argued that the “withdrawal or limitation of many privileges and rights” that necessarily accompanies incarceration, Price v. Johnston, 334 U. S. 266, 285 (1948), combined with the fact that the inmates’ familial bonds are not altogether severed by such a ban, means that something less than a “compelling” government interest would suffice to legitimate the impairment of the inmates’ rights.8 On the other hand, two factors suggest that only a very important public purpose could sustain the policy. First, even persons lawfully incarcerated after being convicted of crimes retain important constitutional rights;9 presumptively innocent persons surely are entitled to no less.10 Second, we have previously insisted upon very persuasive justifications for government regulations that significantly, but not prohibitively, interfered with the exercise of familial rights;11 arguably, a similarly stringent test should control here. However, a sensitive balancing of these competing considerations is unnecessary to resolve the case before us. At a minimum, *601petitioners, to prevail, should be required to show that the jail’s policy materially advances a substantial government interest. Petitioners have not made and, on this record, could not make such a demonstration.12

It should be emphasized that what petitioners must defend is not their reluctance to allow unlimited contact visitation, but rather their refusal to adopt the specific reforms ordered by the lower courts. The District Court’s order, it should be recalled, was carefully circumscribed:

“Commencing not more than ninety days following the date of this order, the defendants will make available a contact visit once each week to each pretrial detainee that has been held in the jail for one month or more, and concerning whom there is no indication of drug or escape propensities; provided, however, that no more than fifteen hundred such visits need be allowed in any one week. In the event that the number of requested visits in any week exceeds fifteen hundred, or such higher number as the Sheriff voluntarily undertakes to accommodate, a reasonable system of rotation or other priorities may be maintained. The lengths of such visits shall remain in the discretion of the Sheriff.” App. to Pet. for Cert. 38.

Petitioners object to this order, and defend their current rule prohibiting all contact between inmates and their families, on two main grounds. Neither of the proffered justifications survives scrutiny.

First, petitioners contend that a ban on contact visitation is necessary to prevent the introduction into the jail of drugs *602and weapons. It must be admitted that this is a legitimate and important goal. However, petitioners fail to show that its realization would be materially impaired by adoption of the reforms ordered by the District Court. Indeed, evidence adduced at trial establishes the contrary. Several witnesses testified that security procedures could be implemented that would make importation of contraband very difficult. Among the precautions effectively used at other institutions are: searches of prisoners before and after visits; dressing of prisoners in special clothes for visitation; examination of prisoners and visitors with metal detectors and fluoroscopes; exclusion of parcels from the visiting area; rejection of visitors who do not comply with visiting rules; and continuous observation of the visiting area by guards. E. g., Tr. 4164-4166, 4232, 4676-4577.13 Mr. Nagel testified that these procedures would “prevent everything except the most extreme methods of introducing drugs into the institution.” Id., at 4170. Further protection against the transmission of contraband from visitors to inmates is provided by the District Court’s restriction of its order to inmates who have been classified as low risk. In short, there is no reason to think that compliance with the lower courts’ directive would result in more than a negligible increase in the flow of drugs or weapons into the jail.14

Second, petitioners contend that allowance of contact visitation would endanger innocent visitors who are placed in near proximity to dangerous detainees. Again, though the *603importance of the objective is apparent, the nexus between it and the jail’s current policy is not. As indicated above, the District Court’s order applies only to detainees who are unlikely to try to escape. And security measures could be employed by petitioners that would make it very difficult for inmates to hurt or take advantage of visitors. See supra, at 602. Finally, the administrators of other institutions that have long permitted contact visits between inmates and their families testified at trial that violent incidents resulting from such visitation are rare, apparently because inmates value their visitation privileges so highly.15

The majority seeks to shore up petitioners’ two arguments with miscellaneous subsidiary claims. In an effort to discredit the limitations on the District Court’s order, the majority argues that determination of which inmates have a sufficiently low propensity to misbehave would be difficult and time-consuming, especially in light of “the brevity of detention and the constantly changing nature of the inmate population.” Ante, at 587. This contention is rebutted by the District Court’s finding that, after an inmate has been incarcerated for a month, jail officials have considerable information regarding his background and behavior patterns, and by evidence in the record that the jail already has a classification system that, with some modification and improvement, could be used to evaluate detainees’ propensities for escape and drug abuse. App. to Pet. for Cert. 33.16 Next, the majority contends that compliance with the Dis*604trict Court’s order would be expensive. Ante, at 588, n. 9. Again, the District Court’s findings are decisive; the court found that only “modest” changes in the jail facilities would be required. App. to Pet. for Cert. 83. More fundamentally, a desire to run a jail as cheaply as possible is not a legitimate reason for abridging the constitutional rights of its occupants. Finally, the majority suggests that the District Court’s order might cause some dissension in the jail, because inmates denied visitation privileges would resent those granted such privileges. Ante, at 587. There is no evidence whatsoever in the record to support this speculative observation.

In sum, neither petitioners nor the majority have shown that permitting low-risk pretrial detainees who have been incarcerated for more than a month occasionally to have contact visits with their spouses and children would frustrate the achievement of any substantial state interest.17 Because such visitation would significantly alleviate the adverse impact of the jail’s current policies upon respondents’ familial rights, its deprivation violates the Due Process Clause.

II

The majority brusquely rejects respondents’ challenge to the jail’s policy of refusing to permit detainees to observe *605searches of their cells on the ground that respondents’ claim is foreclosed by the decision in Wolfish. If respondents’ claim were indeed identical to that presented by the Wolfish plaintiffs, I would vote to affirm on this issue for the reasons stated in my dissenting opinion in Wolfish. See 441 U. S., at 576. In fact, however, the two cases differ in a crucial respect, and that difference provides an independent ground for sustaining the judgment below.

The Court in Wolfish held that the policy adopted by the Metropolitan Correctional Center of not allowing pretrial detainees to observe searches of their cells did not violate the Fourth Amendment and did not constitute punishment vio-lative of the Due Process Clause. Id,., at 556-557, 560-561. Respondents in this case make a quite different claim. They assert that the Central Jail’s policy of searching cells and confiscating or destroying personal possessions found therein, without allowing inmates to observe those searches, deprives inmates of property without due process of law. On the record before us, I think respondents’ claim is meritorious.

One of the purposes of the Due Process Clause is to reduce the incidence of error in deprivations of life, liberty, or property. See Fuentes v. Shevin, 407 U. S. 67, 80-81 (1972). One of the ways such error can be reduced, in turn, is by allowing persons whose interests may be affected adversely by government decisions to participate in those decisions. In Mathews v. Eldridge, 424 U. S. 319 (1976), the Court identified a complex of considerations that are helpful in determining whether the Constitution mandates such participation in particular contexts:

“[Identification of the specific dictates of due process generally requires consideration of three distinct factors: First, the private interest that will be affected by the official action; second, the risk of an erroneous deprivation of such interest through the procedures used, and the probable value, if any, of additional or substitute proce*606dural safeguards; and finally, the Government’s interest, including the function involved and the fiscal and administrative burdens that the additional or substitute procedural requirement would entail.” Id., at 335.

Application of these factors to the facts of the instant case provides strong support for the judgment of the courts below. As the District Court aptly observed, the private interests affected by the jail’s cell-search procedure are important. “The possessions that a man is allowed to keep in his cell are meager indeed, being limited to things like a few pictures, magazines, cigarettes, candy bars, and perhaps an extra pair of socks. Nonetheless, these items are cherished by the inmates.” App. to Pet. for Cert. 27-28.18 Next, the District Court found that the risk, under the jail’s current policy, that inmates’ possessions will be destroyed unnecessarily is substantial. Unannounced shakedown searches inevitably are somewhat hasty, and the officers conducting them have significant discretion in deciding what to leave and what to confiscate. Id., at 28. If allowed to observe the process, inmates can persuade the officers to preserve possessions that would otherwise be destroyed. Ibid.19 Finally, to allow detainees to witness searches of their cells *607would impose only slight burdens on the jail officials. In response to the District Court’s original order, petitioners developed alternative methods of conducting shakedown searches, each of which made it possible for inmates to be present. One of those procedures, known as “Method C,” proved to be no less effectual, no more time-consuming, and only slightly more expensive than the practice challenged by respondents.20 The demonstrated feasibility21 and minor cost of this option renders indefensible, in my view, petitioners’ insistence that detainees not be permitted to observe cell searches.

In sum, this seems a classic instance in which an “established state procedure,” as distinguished from “a random and *608unauthorized act by a state employee,” has the effect of causing unnecessary deprivations of private property. Compare Logan v. Zimmerman Brush Co., 455 U. S. 422, 435-436 (1982), with Hudson v. Palmer, ante, p. 517, and Parratt v. Taylor, 451 U. S. 527, 541 (1981). In view of the ease with which petitioners could implement an alternative procedure that would reduce the incidence of wanton destruction of inmates’ possessions, I would affirm the judgment of the courts below that the jail’s current practice violates the Due Process Clause.22

I respectfully dissent.