This case presents the question whether evidence seized in violation of the Fourth Amendment by an officer who acted in reliance on a police record indicating the existence of an outstanding arrest warrant—a record that is later determined to be erroneous—must be suppressed by virtue of the exclusionary rule regardless of the source of the error. The Supreme Court of Arizona held that the exclusionary rule required suppression of evidence even if the erroneous information resulted from an error committed by an employee of the office of the Clerk of Court. We disagree.

In January 1991, Phoenix police officer Bryan Sargent observed respondent Evans driving the wrong way on a one-way street in front of the police station. The officer stopped respondent and asked to see his driver's license. After respondent told him that his license had been suspended, the officer entered respondent's name into a computer data terminal located in his patrol car. The computer inquiry confirmed that respondent's license had been suspended and also indicated that there was an outstanding misdemeanor warrant for his arrest. Based upon the outstanding warrant, Officer Sargent placed respondent under arrest. While being handcuffed, respondent dropped a hand-rolled cigarette that the officers determined smelled of marijuana. Officers proceeded to search his car and discovered a bag of marijuana under the passenger's seat.

The State charged respondent with possession of marijuana. When the police notified the Justice Court that they had arrested him, the Justice Court discovered that the arrest warrant previously had been quashed and so advised the police. Respondent argued that because his arrest was based on a warrant that had been quashed 17 days prior to his arrest, the marijuana seized incident to the arrest should be suppressed as the fruit of an unlawful arrest. Respondent also argued that "[t]he 'good faith' exception to the exclusionary rule [was] inapplicable . . . because it was police error, not judicial error, which caused the invalid arrest." App. 5.

At the suppression hearing, the Chief Clerk of the Justice Court testified that a Justice of the Peace had issued the arrest warrant on December 13, 1990, because respondent had failed to appear to answer for several traffic violations. On December 19, 1990, respondent appeared before a pro tem Justice of the Peace who entered a notation in respondent's file to "quash warrant." Id., at ----.

The Chief Clerk also testified regarding the standard court procedure for quashing a warrant. Under that procedure a justice court clerk calls and informs the warrant section of the Sheriff's Office when a warrant has been quashed. The Sheriff's Office then removes the warrant from its computer records. After calling the Sheriff's Office, the clerk makes a note in the individual's file indicating the clerk who made the phone call and the person at the Sheriff's Office to whom the clerk spoke. The Chief Clerk testified that there was no indication in respondent's file that a clerk had called and notified the Sheriff's Office that his arrest warrant had been quashed. A records clerk from the Sheriff's Office also testified that the Sheriff's Office had no record of a telephone call informing it that respondent's arrest warrant had been quashed. Id., at 42-43.

At the close of testimony, respondent argued that the evidence obtained as a result of the arrest should be suppressed because "the purposes of the exclusionary rule would be served here by making the clerks for the court, or the clerk for the Sheriff's office, whoever is responsible for this mistake, to be more careful about making sure that warrants are removed from the records." Id., at 47. The trial court granted the motion to suppress because it concluded that the State had been at fault for failing to quash the warrant. Presumably because it could find no "distinction between State action, whether it happens to be the police department or not," id., at 52, the trial court made no factual finding as to whether the Justice Court or Sheriff's Office was responsible for the continued presence of the quashed warrant in the police records.

A divided panel of the Arizona Court of Appeals reversed because it "believe[d] that the exclusionary rule [was] not intended to deter justice court employees or Sheriff's Office employees who are not directly associated with the arresting officers or the arresting officers' police department." 172 Ariz. 314, 317, 836 P.2d 1024, 1027 (1992). Therefore, it concluded, "the purpose of the exclusionary rule would not be served by excluding the evidence obtained in this case." Ibid.

The Arizona Supreme Court reversed. 177 Ariz. 201, 866 P.2d 869 (1994). The court rejected the "distinction drawn by the court of appeals . . . between clerical errors committed by law enforcement personnel and similar mistakes by court employees." Id., at 203, 866 P.2d, at 871. The court predicted that application of the exclusionary rule would "hopefully serve to improve the efficiency of those who keep records in our criminal justice system." Id., at 204, 866 P.2d, at 872. Finally, the Court concluded that "[e]ven assuming that deterrence is the principal reason for application of the exclusionary rule, we disagree with the court of appeals that such a purpose would not be served where carelessness by a court clerk results in an unlawful arrest." Ibid.

We granted certiorari to determine whether the exclusionary rule requires suppression of evidence seized incident to an arrest resulting from an inaccurate computer record, regardless of whether police personnel or court personnel were responsible for the record's continued presence in the police computer. 511 U.S. ----, 114 S.Ct. 2131, 128 L.Ed.2d 862 (1994).1 We now reverse.

We first must consider whether we have jurisdiction to review the Arizona Supreme Court's decision. Respondent argues that we lack jurisdiction under 28 U.S.C. § 1257 because the Arizona Supreme Court never passed upon the Fourth Amendment issue and instead based its decision on the Arizona good-faith statute, Ariz.Rev.Stat.Ann. § 13-3925 (1993), an adequate and independent state ground. In the alternative, respondent asks that we remand to the Arizona Supreme Court for clarification.

In Michigan v. Long, 463 U.S. 1032, 103 S.Ct. 3469, 77 L.Ed.2d 1201 (1983), we adopted a standard for determining whether a state-court decision rested upon an adequate and independent state ground. When "a state court decision fairly appears to rest primarily on federal law, or to be interwoven with the federal law, and when the adequacy and independence of any possible state law ground is not clear from the face of the opinion, we will accept as the most reasonable explanation that the state court decided the case the way it did because it believed that federal law required it to do so." Id., at 1040-1041, 103 S.Ct. at 3476. We adopted this practice, in part, to obviate the "unsatisfactory and intrusive practice of requiring state courts to clarify their decisions to the satisfaction of this Court." Id., at 1041, 103 S.Ct. at 3476. We also concluded that this approach would "provide state judges with a clearer opportunity to develop state jurisprudence unimpeded by federal interference, and yet will preserve the integrity of federal law." Ibid.

Justice GINSBURG would overrule Michigan v. Long, supra, because she believes that the rule of that case "impedes the States' ability to serve as laboratories for testing solutions to novel legal problems." Post, at __.2 The opinion in Long describes the 60-year history of the Court's differing approaches to the determination whether the judgment of the highest court of a State rested on federal or nonfederal grounds. 463 U.S., at 1038-1040, 103 S.Ct. at 3474-3476. When we were in doubt, on some occasions we dismissed the writ of certiorari; on other occasions we vacated the judgment of the state court and remanded so that it might clarify the basis for its decision. See ibid. The latter approach did not always achieve the desired result and burdened the state courts with additional work. Ibid.

We believe that Michigan v. Long properly serves its purpose and should not be disturbed. Under it, state courts are absolutely free to interpret state constitutional provisions to accord greater protection to individual rights than do similar provisions of the United States Constitution. They also are free to serve as experimental laboratories, in the sense that Justice Brandeis used that term in his dissenting opinion in New State Ice Co. v. Liebmann, 285 U.S. 262, 311, 52 S.Ct. 371, 386, 76 L.Ed. 747 (1932) (urging that the Court not impose federal constitutional restraints on the efforts of a State to "serve as a laboratory"). Under our decision today, the State of Arizona remains free to seek whatever solutions it chooses to problems of law enforcement posed by the advent of computerization.3 Indeed, it is freer to do so because it is disabused of its erroneous view of what the United States Constitution requires.

State courts, in appropriate cases, are not merely free to they are bound to—interpret the United States Constitution. In doing so, they are not free from the final authority of this Court. This principle was enunciated in Cohens v. Virginia, 19 U.S. (6 Wheat.) 264, 5 L.Ed. 257 (1821), and presumably Justice GINSBURG does not quarrel with it.4 In Minnesota v. National Tea Co., 309 U.S. 551, 60 S.Ct. 676, 84 L.Ed. 920 (1940), we recognized that our authority as final arbiter of the United States Constitution could be eroded by a lack of clarity in state-court decisions.

"It is fundamental that state courts be left free and unfettered by us in interpreting their state constitutions. But it is equally important that ambiguous or obscure adjudications by state courts do not stand as barriers to a determination by this Court of the validity under the federal constitution of state action. Intelligent exercise of our appellate powers compels us to ask for the elimination of the obscurities and ambiguities from the opinions in such cases. . . . For no other course assures that important federal issues, such as have been argued here, will reach this Court for adjudication; that state courts will not be the final arbiters of important issues under the federal constitution; and that we will not encroach on the constitutional jurisdiction of the states." Id., at 557, 60 S.Ct., at 679.

We therefore adhere to the standard adopted in Michigan v. Long, supra.

Applying that standard here, we conclude that we have jurisdiction. In reversing the Court of Appeals, the Arizona Supreme Court stated that "[w]hile it may be inappropriate to invoke the exclusionary rule where a magistrate has issued a facially valid warrant (a discretionary judicial function) based on an erroneous evaluation of the facts, the law, or both, Leon, 468 U.S. 897 [104 S.Ct. 3405, 82 L.Ed.2d 677] . . . (1984), it is useful and proper to do so where negligent record keeping (a purely clerical function) results in an unlawful arrest." 177 Ariz., at 204, 866 P.2d, at 872. Thus, the Arizona Supreme Court's decision to suppress the evidence was based squarely upon its interpretation of federal law. See ibid. Nor did it offer a plain statement that its references to federal law were "being used only for the purpose of guidance, and d[id] not themselves compel the result that [it] reached." Long, supra, 463 U.S., at 1041, 103 S.Ct., at 3476.

The Fourth Amendment states that "[t]he right of the people to be secure in their persons, houses, papers, and effects, against unreasonable searches and seizures, shall not be violated, and no Warrants shall issue, but upon probable cause, supported by Oath or affirmation, and particularly describing the place to be searched, and the persons or things to be seized." U.S. Const. We have recognized, however, that the Fourth Amendment contains no provision expressly precluding the use of evidence obtained in violation of its commands. See United States v. Leon, 468 U.S. 897, 906, 104 S.Ct. 3405, 3411-3412, 82 L.Ed.2d 677 (1984). "The wrong condemned by the [Fourth] Amendment is 'fully accomplished' by the unlawful search or seizure itself," ibid. (quoting United States v. Calandra, 414 U.S. 338, 354, 94 S.Ct. 613, 623, 38 L.Ed.2d 561 (1974)), and the use of the fruits of a past unlawful search or seizure " 'work[s] no new Fourth Amendment wrong,' " Leon, supra, 468 U.S., at 906, 104 S.Ct., at 3411 (quoting Calandra, supra, 414 U.S., at 354, 94 S.Ct., at 623).

"The question whether the exclusionary rule's remedy is appropriate in a particular context has long been regarded as an issue separate from the question whether the Fourth Amendment rights of the party seeking to invoke the rule were violated by police conduct." Illinois v. Gates, 462 U.S. 213, 223, 103 S.Ct. 2317, 2324, 76 L.Ed.2d 527 (1983); see also United States v. Havens, 446 U.S. 620, 627-628, 100 S.Ct. 1912, 1916-17, 64 L.Ed.2d 559 (1980); Stone v. Powell, 428 U.S. 465, 486-487, 96 S.Ct. 3037, 3048-3049, 49 L.Ed.2d 1067 (1976); Calandra, supra, 414 U.S., at 348, 94 S.Ct., at 620. The exclusionary rule operates as a judicially created remedy designed to safeguard against future violations of Fourth Amendment rights through the rule's general deterrent effect. Leon, supra, 468 U.S., at 906, 104 S.Ct., at 3411-3412; Calandra, supra, 414 U.S., at 348, 94 S.Ct., at 620. As with any remedial device, the rule's application has been restricted to those instances where its remedial objectives are thought most efficaciously served. Leon, supra, 468 U.S., at 908, 104 S.Ct., at 3412-3413; Calandra, supra, 414 U.S., at 348, 94 S.Ct., at 620. Where "the exclusionary rule does not result in appreciable deterrence, then, clearly, its use . . . is unwarranted." United States v. Janis, 428 U.S. 433, 454, 96 S.Ct. 3021, 3032, 49 L.Ed.2d 1046 (1976).

In Leon, we applied these principles to the context of a police search in which the officers had acted in objectively reasonable reliance on a search warrant, issued by a neutral and detached Magistrate, that later was determined to be invalid. 468 U.S., at 905, 104 S.Ct., at 3411. On the basis of three factors, we determined that there was no sound reason to apply the exclusionary rule as a means of deterring misconduct on the part of judicial officers who are responsible for issuing warrants. See Illinois v. Krull, 480 U.S. 340, 348, 107 S.Ct. 1160, 1166, 94 L.Ed.2d 364 (1987) (analyzing Leon, supra ). First, we noted that the exclusionary rule was historically designed " 'to deter police misconduct rather than to punish the errors of judges and magistrates.' " Krull, supra, 480 U.S., at 348, 107 S.Ct., at 1166 (quoting Leon, supra, 468 U.S., at 916, 104 S.Ct., at 3417). Second, there was " 'no evidence suggesting that judges and magistrates are inclined to ignore or subvert the Fourth Amendment or that lawlessness among these actors requires the application of the extreme sanction of exclusion.' " Krull, supra, 480 U.S., at 348, 107 S.Ct., at 1166 (quoting Leon, supra, 468 U.S., at 916, 104 S.Ct., at 3417). Third, and of greatest importance, there was no basis for believing that exclusion of evidence seized pursuant to a warrant would have a significant deterrent effect on the issuing judge or magistrate. Krull, supra, 480 U.S., at 348, 107 S.Ct., at 1166.

The Leon Court then examined whether application of the exclusionary rule could be expected to alter the behavior of the law enforcement officers. We concluded:

"[W]here the officer's conduct is objectively reasonable, 'excluding the evidence will not further the ends of the exclusionary rule in any appreciable way; for it is painfully apparent that . . . the officer is acting as a reasonable officer would and should act in similar circumstances. Excluding the evidence can in no way affect his future conduct unless it is to make him less willing to do his duty.' " Leon, supra, 468 U.S., at 919-920, 104 S.Ct., at 3419 (quoting Stone v. Powell, supra, 428 U.S., at 539-540, 96 S.Ct., at 3073 (White, J., dissenting)).

See also Massachusetts v. Sheppard, 468 U.S. 981, 990-991, 104 S.Ct. 3424, 3428-3429, 82 L.Ed.2d 737 (1984) ("[S]uppressing evidence because the judge failed to make all the necessary clerical corrections despite his assurances that such changes would be made will not serve the deterrent function that the exclusionary rule was designed to achieve"). Thus, we held that the "marginal or nonexistent benefits produced by suppressing evidence obtained in objectively reasonable reliance on a subsequently invalidated search warrant cannot justify the substantial costs of exclusion." Leon, supra, 468 U.S., at 922, 104 S.Ct., at 3420.

Respondent relies on United States v. Hensley, 469 U.S. 221, 105 S.Ct. 675, 83 L.Ed.2d 604 (1985), and argues that the evidence seized incident to his arrest should be suppressed because he was the victim of a Fourth Amendment violation. Brief for Respondent 10-12, 21-22. In Hensley, the Court determined that evidence uncovered as a result of a Terry stop was admissible because the officers who made the stop acted in objectively reasonable reliance on a flyer that had been issued by officers of another police department who possessed a reasonable suspicion to justify a Terry stop. 469 U.S., at 231, 105 S.Ct., at 681-682. Because the Hensley Court determined that there had been no Fourth Amendment violation, id., at 236, 105 S.Ct., at 684, the Court never considered whether the seized evidence should have been excluded. Hensley does not contradict our earlier pronouncements that "[t]he question whether the exclusionary rule's remedy is appropriate in a particular context has long been regarded as an issue separate from the question whether the Fourth Amendment rights of the party seeking to invoke the rule were violated by police conduct." Gates, supra, 462 U.S., at 223, 103 S.Ct., at 2324; see also Stone v. Powell, supra, 428 U.S., at 486-487, 96 S.Ct., at 3048-3049; Calandra, 414 U.S., at 348, 94 S.Ct., at 620.

Respondent also argues that Whiteley v. Warden, Wyoming State Penitentiary, 401 U.S. 560, 91 S.Ct. 1031, 28 L.Ed.2d 306 (1971), compels exclusion of the evidence. In Whiteley, the Court determined that the Fourth Amendment had been violated when police officers arrested Whiteley and recovered inculpatory evidence based upon a radio report that two suspects had been involved in two robberies. Id., at 568-569, 91 S.Ct., at 1037-1038. Although the "police were entitled to act on the strength of the radio bulletin," the Court determined that there had been a Fourth Amendment violation because the initial complaint, upon which the arrest warrant and subsequent radio bulletin were based, was insufficient to support an independent judicial assessment of probable cause. Id., at 568, 91 S.Ct., at 1037. The Court concluded that "an otherwise illegal arrest cannot be insulated from challenge by the decision of the instigating officer to rely on fellow officers to make the arrest." Ibid. Because the "arrest violated [Whiteley's] constitutional rights under the Fourth and Fourteenth Amendments; the evidence secured as an incident thereto should have been excluded from his trial. Mapp v. Ohio, 367 U.S. 643 [81 S.Ct. 1684, 6 L.Ed.2d 1081] (1961)." Whiteley, supra, 401 U.S., at 568-569, 91 S.Ct., at 1037.

Although Whiteley clearly retains relevance in determining whether police officers have violated the Fourth Amendment, see Hensley, supra, 469 U.S., at 230-231, 105 S.Ct., at 681-682, its precedential value regarding application of the exclusionary rule is dubious. In Whiteley, the Court treated identification of a Fourth Amendment violation as synonymous with application of the exclusionary rule to evidence secured incident to that violation. 401 U.S., at 568-569, 91 S.Ct., at 1037-38. Subsequent case law has rejected this reflexive application of the exclusionary rule. Cf. Krull, supra; Sheppard, supra; United States v. Leon, 468 U.S. 897, 104 S.Ct. 3405, 82 L.Ed.2d 677 (1984); Calandra, supra. These later cases have emphasized that the issue of exclusion is separate from whether the Fourth Amendment has been violated, see e.g., Leon, supra, 468 U.S., at 906, 104 S.Ct., at 3411-3412, and exclusion is appropriate only if the remedial objectives of the rule are thought most efficaciously served, see Calandra, supra, 414 U.S., at 348, 94 S.Ct., at 620.

Our approach is consistent with the dissenting Justices' position in Illinois v. Krull, our only major case since Leon and Sheppard involving the good-faith exception to the exclusionary rule. In that case, the Court found that the good-faith exception applies when an officer conducts a search in objectively reasonable reliance on the constitutionality of a statute that subsequently is declared unconstitutional. Krull, 480 U.S., at 346, 107 S.Ct., at 1165. Even the dissenting Justices in Krull agreed that Leon provided the proper framework for analyzing whether the exclusionary rule applied; they simply thought that "application of Leon's stated rationales le[d] to a contrary result." 480 U.S., at 362, 107 S.Ct., at 1173 (O'CONNOR, J., dissenting). In sum, respondent does not persuade us to abandon the Leon framework.

Applying the reasoning of Leon to the facts of this case, we conclude that the decision of the Arizona Supreme Court must be reversed. The Arizona Supreme Court determined that it could not "support the distinction drawn . . . between clerical errors committed by law enforcement personnel and similar mistakes by court employees," 177 Ariz., at 203, 866 P.2d, at 871, and that "even assuming . . . that responsibility for the error rested with the justice court, it does not follow that the exclusionary rule should be inapplicable to these facts," ibid.

This holding is contrary to the reasoning of Leon, supra; Massachusetts v. Sheppard, 468 U.S. 981, 104 S.Ct. 3424, 82 L.Ed.2d 737 (1984); and, Krull, supra. If court employees were responsible for the erroneous computer record, the exclusion of evidence at trial would not sufficiently deter future errors so as to warrant such a severe sanction. First, as we noted in Leon, the exclusionary rule was historically designed as a means of deterring police misconduct, not mistakes by court employees. See Leon, supra, 468 U.S., at 916, 104 S.Ct., at 3417; see also Krull, supra, 480 U.S., at 350, 107 S.Ct., at 1167. Second, respondent offers no evidence that court employees are inclined to ignore or subvert the Fourth Amendment or that lawlessness among these actors requires application of the extreme sanction of exclusion. See Leon, supra, 468 U.S., at 916, 104 S.Ct., 3417, and n. 14; see also Krull, supra, 480 U.S., at 350-351, 107 S.Ct., at 1167-1168. To the contrary, the Chief Clerk of the Justice Court testified at the suppression hearing that this type of error occurred once every three or four years. App. 37.

Finally, and most important, there is no basis for believing that application of the exclusionary rule in these circumstances will have a significant effect on court employees responsible for informing the police that a warrant has been quashed. Because court clerks are not adjuncts to the law enforcement team engaged in the often competitive enterprise of ferreting out crime, see Johnson v. United States, 333 U.S. 10, 14, 68 S.Ct. 367, 369, 92 L.Ed. 436 (1948), they have no stake in the outcome of particular criminal prosecutions. Cf. Leon, supra, 468 U.S., at 917, 104 S.Ct., at 3417-3418; Krull, supra, 480 U.S., at 352, 107 S.Ct., at 1168. The threat of exclusion of evidence could not be expected to deter such individuals from failing to inform police officials that a warrant had been quashed. Cf. Leon, supra, 468 U.S., at 917, 104 S.Ct., at 3417-3418; Krull, supra, 480 U.S., at 352, 107 S.Ct., at 1168.

If it were indeed a court clerk who was responsible for the erroneous entry on the police computer, application of the exclusionary rule also could not be expected to alter the behavior of the arresting officer. As the trial court in this case stated: "I think the police officer [was] bound to arrest. I think he would [have been] derelict in his duty if he failed to arrest." App. 51. Cf. Leon, supra, 468 U.S., at 920, 104 S.Ct., at 3419 (" 'Excluding the evidence can in no way affect [the officer's] future conduct unless it is to make him less willing to do his duty.' " quoting Stone v. Powell, 428 U.S., at 540, 96 S.Ct., at 3073 (White, J., dissenting)). The Chief Clerk of the Justice Court testified that this type of error occurred "on[c]e every three or four years." App. 37. In fact, once the court clerks discovered the error, they immediately corrected it, id., at 30, and then proceeded to search their files to make sure that no similar mistakes had occurred, id., at 37. There is no indication that the arresting officer was not acting objectively reasonably when he relied upon the police computer record. Application of the Leon framework supports a categorical exception to the exclusionary rule for clerical errors of court employees. See Leon, 468 U.S., at 916-922, 104 S.Ct., at 3417-3420; Sheppard, supra, 468 U.S., at 990-991, 104 S.Ct., at 3428-3429.5

The judgment of the Supreme Court of Arizona is therefore reversed, and the case is remanded to that court for proceedings not inconsistent with this opinion.

It is so ordered.

Justice O'CONNOR, with whom Justice SOUTER and Justice BREYER join, concurring.

The evidence in this case strongly suggests that it was a court employee's departure from established record-keeping procedures that caused the record of respondent's arrest warrant to remain in the computer system after the warrant had been quashed. Prudently, then, the Court limits itself to the question whether a court employee's departure from such established procedures is the kind of error to which the exclusionary rule should apply. The Court holds that it is not such an error, and I agree with that conclusion and join the Court's opinion. The Court's holding reaffirms that the exclusionary rule imposes significant costs on society's law enforcement interests and thus should apply only where its deterrence purposes are "most efficaciously served," ante, at __.

In limiting itself to that single question, however, the Court does not hold that the court employee's mistake in this case was necessarily the only error that may have occurred and to which the exclusionary rule might apply. While the police were innocent of the court employee's mistake, they may or may not have acted reasonably in their reliance on the recordkeeping system itself. Surely it would not be reasonable for the police to rely, say, on a recordkeeping system, their own or some other agency's, that has no mechanism to ensure its accuracy over time and that routinely leads to false arrests, even years after the probable cause for any such arrest has ceased to exist (if it ever existed).

This is saying nothing new. We have said the same with respect to other information sources police use, informants being an obvious example. In Illinois v. Gates, 462 U.S. 213, 103 S.Ct. 2317, 76 L.Ed.2d 527 (1983), the Court indicated that where an informant provides information about certain criminal activities but does not specify the basis for his knowledge, a finding of probable cause based on that information will not be upheld unless the informant is "known for [his] unusual reliability." Id., at 233, 103 S.Ct., at 2329, citing United States v. Sellers, 483 F.2d 37, 40, n. 1 (CA5 1973) (involving informant who had provided accurate information "in more than one hundred instances in matters of investigation"); see generally 1 W. LaFave, Search and Seizure § 3.3(b) (2d ed. 1987 and Supp.1995). Certainly the reliability of recordkeeping systems deserves no less scrutiny than that of informants. Of course, the comparison to informants may be instructive the opposite way as well. So long as an informant's reliability does pass constitutional muster, a finding of probable cause may not be defeated by an after-the-fact showing that the information the informant provided was mistaken. See 2 id. § 3.5(d), at 21, n. 73 (citation omitted); see also 1 id. § 3.2(d), at 575 ("It is axiomatic that hindsight may not be employed in determining whether a prior arrest or search was made upon probable cause").

In recent years, we have witnessed the advent of powerful, computer-based recordkeeping systems that facilitate arrests in ways that have never before been possible. The police, of course, are entitled to enjoy the substantial advantages this technology confers. They may not, however, rely on it blindly. With the benefits of more efficient law enforcement mechanisms comes the burden of corresponding constitutional responsibilities. djQ Justice SOUTER, with whom Justice BREYER joins, concurring.

In joining the Court's opinion, I share Justice O'CONNOR's understanding of the narrow scope of what we hold today. To her concurrence, which I join as well, I add only that we do not answer another question that may reach us in due course, that is, how far, in dealing with fruits of computerized error, our very concept of deterrence by exclusion of evidence should extend to the government as a whole, not merely the police, on the ground that there would otherwise be no reasonable expectation of keeping the number of resulting false arrests within an acceptable minimum limit.

Justice STEVENS, dissenting.

Justice GINSBURG has written an important opinion explaining why the Court unwisely departed from settled law when it interpreted its own jurisdiction so expansively in Michigan v. Long, 463 U.S. 1032, 103 S.Ct. 3469, 77 L.Ed.2d 1201 (1983). I join her dissent and her conclusion that the writ of certiorari should be dismissed. Because the Court has addressed the merits, however, I add this comment on its holding.

The Court seems to assume that the Fourth Amendment—and particularly the exclusionary rule, which effectuates the Amendment's commands—has the limited purpose of deterring police misconduct. Both the constitutional text and the history of its adoption and interpretation identify a more majestic conception. The Amendment protects the fundamental "right of the people to be secure in their persons, houses, papers, and effects," against all official searches and seizures that are unreasonable. The Amendment is a constraint on the power of the sovereign, not merely on some of its agents. See Olmstead v. United States, 277 U.S. 438, 472-479, 48 S.Ct. 564, 570-573, 72 L.Ed. 944 (1928) (Brandeis, J., dissenting). The remedy for its violation imposes costs on that sovereign, motivating it to train all of its personnel to avoid future violations. See Stewart, The Road to Mapp v. Ohio and Beyond: The Origins, Development and Future of the Exclusionary Rule in Search-and-Seizure Cases, 83 Colum.L.Rev. 1365, 1400 (1983).

The exclusionary rule is not fairly characterized as an "extreme sanction," ante, at __. As Justice Stewart cogently explained, the implementation of this constitutionally mandated sanction merely places the Government in the same position as if it had not conducted the illegal search and seizure in the first place.1 Given the undisputed fact in this case that the Constitution prohibited the warrantless arrest of petitioner, there is nothing "extreme" about the Arizona Supreme Court's conclusion that the State should not be permitted to profit from its negligent misconduct.

Even if one accepts deterrence as the sole rationale for the exclusionary rule, the Arizona Supreme Court's decision is correct on the merits. The majority's reliance on United States v. Leon, 468 U.S. 897, 104 S.Ct. 3405, 82 L.Ed.2d 677 (1984), is misplaced. The search in that case had been authorized by a presumptively valid warrant issued by a California Superior Court Judge. In contrast, this case involves a search pursuant to an arrest made when no warrant at all was outstanding against petitioner. The holding in Leon rested on the majority's doubt "that exclusion of evidence seized pursuant to a warrant will have a significant deterrent effect on the issuing judge or magistrate." Id., at 916, 104 S.Ct., at 3417. The reasoning in Leon assumed the existence of a warrant; it was, and remains, wholly inapplicable to warrantless searches and seizures.2

The Fourth Amendment's Warrant Clause provides the fundamental check on official invasions of the individual's right to privacy. E.g., Harris v. United States, 331 U.S. 145, 195-196, 67 S.Ct. 1098, 1118-1119, 91 L.Ed. 1399 (1947) (Jackson, J., dissenting); see generally Kamisar, Does (Did) (Should) the Exclusionary Rule Rest on a "Principled Basis" Rather Than an "Empirical Proposition"?, 16 Creighton L.Rev. 565, 571-579 (1983). Leon stands for the dubious but limited proposition that courts should not look behind the face of a warrant on which police have relied in good faith. The Leon Court's exemption of judges and magistrates from the deterrent ambit of the exclusionary rule rested, consistently with the emphasis on the warrant requirement, on those officials' constitutionally determined role in issuing warrants. See Leon, 468 U.S., at 915-917, 104 S.Ct. at 3416-3418. Taken on its own terms, Leon's logic does not extend to the time after the warrant has issued; nor does it extend to court clerks and functionaries, some of whom work in the same building with police officers and may have more regular and direct contact with police than with judges or magistrates.

The Phoenix Police Department was part of the chain of information that resulted in petitioner's unlawful, warrantless arrest. We should reasonably presume that law enforcement officials, who stand in the best position to monitor such errors as occurred here, can influence mundane communication procedures in order to prevent those errors. That presumption comports with the notion that the exclusionary rule exists to deter future police misconduct systemically. See, e.g., Stone v. Powell, 428 U.S. 465, 492, 96 S.Ct. 3037, 3051, 49 L.Ed.2d 1067 (1976); Dunaway v. New York, 442 U.S. 200, 221, 99 S.Ct. 2248, 2261, 60 L.Ed.2d 824 (1979) (STEVENS, J., concurring); see generally Kamisar, supra, at 659-662; Stewart, supra, at 1400. The deterrent purpose extends to law enforcement as a whole, not merely to "the arresting officer." Compare ante, at __, with Whiteley v. Warden, Wyoming State Penitentiary, 401 U.S. 560, 568, 91 S.Ct. 1031, 1037, 28 L.Ed.2d 306 (1971). Consequently, the Phoenix officers' good faith does not diminish the deterrent value of invalidating their arrest of petitioner.

The Court seeks to minimize the impact of its holding on the security of the citizen by referring to the testimony of the chief clerk of the East Phoenix Number One Justice Court that in her "particular court" this type of error occurred " 'maybe [once] every three or four years.' " See ante, at __. Apart from the fact that the clerk promptly contradicted herself,3 see post, at __, this is slim evidence on which to base a conclusion that computer error poses no appreciable threat to Fourth Amendment interests. For support, the Court cites a case from 1948. See ante, at __, citing Johnson v. United States, 333 U.S. 10, 68 S.Ct. 367, 92 L.Ed. 436 (1948). The Court overlooks the reality that computer technology has changed the nature of threats to citizens' privacy over the past half century. See post, at __. What has not changed is the reality that only that fraction of Fourth Amendment violations held to have resulted in unlawful arrests is ever noted and redressed. As Justice Jackson observed: "There may be, and I am convinced that there are, many unlawful searches . . . of innocent people which turn up nothing incriminating, in which no arrest is made, about which courts do nothing, and about which we never hear." Brinegar v. United States, 338 U.S. 160, 181, 69 S.Ct. 1302, 1313, 93 L.Ed. 1879 (1949) (dissenting opinion). Moreover, even if errors in computer records of warrants were rare, that would merely minimize the cost of enforcing the exclusionary rule in cases like this.

While I agree with Justice GINSBURG that premature adjudication of this issue is particularly unwise because we have much to learn about the consequences of computer error as well as the efficacy of other preventive measures, see post at 7-8, one consequence of the Court's holding seems immediately obvious. Its most serious impact will be on the otherwise innocent citizen who is stopped for a minor traffic infraction and is wrongfully arrested based on erroneous information in a computer data base. I assume the police officer who reasonably relies on the computer information would be immune from liability in a § 1983 action. Of course, the Court has held that respondeat superior is unavailable as a basis for imposing liability on her municipality. See Monell v. New York City Dept. of Social Services, 436 U.S. 658, 663-664, n. 7, 98 S.Ct. 2018, 2021-2022, n. 7, 56 L.Ed.2d 611 (1978). Thus, if courts are to have any power to discourage official error of this kind, it must be through application of the exclusionary rule.

The use of general warrants to search for evidence of violations of the Crown's revenue laws understandably outraged the authors of the Bill of Rights. See, e.g., Lo-Ji Sales, Inc. v. New York, 442 U.S. 319, 325, 99 S.Ct. 2319, 2323-2324, 60 L.Ed.2d 920 (1979); Weeks v. United States, 232 U.S. 383, 389-391, 34 S.Ct. 341, 343-344, 58 L.Ed. 652 (1914). " 'It is a power, that places the liberty of every man in the hands of every petty officer.' " James Otis, quoted in 2 Works of John Adams 524 (C. Adams, ed. 1850), in turn Illinois v. Krull, 480 U.S. 340, 363, 107 S.Ct. 1160, 1174, 94 L.Ed.2d 364 (1987) (O'CONNOR, J., dissenting). The offense to the dignity of the citizen who is arrested, handcuffed, and searched on a public street simply because some bureaucrat has failed to maintain an accurate computer data base strikes me as equally outrageous. In this case, of course, such an error led to the fortuitous detection of respondent's unlawful possession of marijuana, and the suppression of the fruit of the error would prevent the prosecution of his crime. That cost, however, must be weighed against the interest in protecting other, wholly innocent citizens from unwarranted indignity. In my judgment, the cost is amply offset by an appropriately "jealous regard for maintaining the integrity of individual rights." Mapp v. Ohio, 367 U.S. 643, 647, 81 S.Ct. 1684, 1687, 6 L.Ed.2d 1081 (1961). For this reason, as well as those set forth by Justice GINSBURG, I respectfully dissent.

Justice GINSBURG, with whom Justice STEVENS joins, dissenting.

This case portrays the increasing use of computer technology in law enforcement; it illustrates an evolving problem this Court need not, and in my judgment should not, resolve too hastily.1 The Arizona Supreme Court relied on "the principles of a free society" in reaching its decision. This Court reviews and reverses the Arizona decision on the assumption that Arizona's highest court sought assiduously to apply this Court's Fourth Amendment jurisprudence. The Court thus follows the presumption announced in Michigan v. Long, 463 U.S. 1032, 103 S.Ct. 3469, 77 L.Ed.2d 1201 (1983): If it is unclear whether a state court's decision rests on state or federal law, Long dictates the assumption that the state court relied on federal law. On the basis of that assumption, the Court asserts jurisdiction to review the decision of the Arizona Supreme Court.

The Long presumption, as I see it, impedes the States' ability to serve as laboratories for testing solutions to novel legal problems. I would apply the opposite presumption and assume that Arizona's Supreme Court has ruled for its own State and people, under its own constitutional recognition of individual security against unwarranted state intrusion. Accordingly, I would dismiss the writ of certiorari.

* Isaac Evans was arrested because a computer record erroneously identified an outstanding misdemeanor arrest warrant in his name. The Arizona Supreme Court's suppression of evidence obtained from this unlawful arrest did not rest on a close analysis of this Court's Fourth Amendment precedents. Indeed, the court found our most relevant decision, United States v. Leon, 468 U.S. 897, 104 S.Ct. 3405, 82 L.Ed.2d 677 (1984), "not helpful." State v. Evans, 177 Ariz. 201, 203, 866 P.2d 869, 871 (1994). Instead, the Arizona court emphasized its comprehension of the severe curtailment of personal liberty inherent in arrest warrants.

Specifically, the Arizona Supreme Court saw the growing use of computerized records in law enforcement as a development presenting new dangers to individual liberty; excluding evidence seized as a result of incorrect computer data, the Arizona court anticipated, would reduce the incidence of uncorrected records:

"The dissent laments the 'high costs' of the exclusionary rule, and suggests that its application here is 'purposeless' and provides 'no offsetting benefits.' Such an assertion ignores the fact that arrest warrants result in a denial of human liberty, and are therefore among the most important of legal documents. It is repugnant to the principles of a free society that a person should ever be taken into police custody because of a computer error precipitated by government carelessness. As automation increasingly invades modern life, the potential for Orwellian mischief grows. Under such circumstances, the exclusionary rule is a 'cost' we cannot afford to be without." Id., at 204, 866 P.2d, at 872.

Thus, the Arizona court did not consider this case to involve simply and only a court employee's slip in failing to communicate with the police, or a police officer's oversight in failing to record information received from a court employee. That court recognized a "potential for Orwellian mischief" in the government's increasing reliance on computer technology in law enforcement. The Arizona Supreme Court concluded that Leon's distinction between police conduct and judicial conduct loses force where, as here, the error derives not from a discretionary judicial function, but from inattentive recordkeeping. Application of an exclusionary rule in the circumstances Evans' case presents, the Arizona court said, "will hopefully serve to improve the efficiency of those who keep records in our criminal justice system." Ibid.

Invoking Long, this Court's majority presumes that the Arizona Supreme Court relied on federal law. Long instructs that a state-court opinion discussing both state and federal precedents shall be deemed to rely on federal law, absent a plain statement in the opinion that the decision rests on state law. Long, 463 U.S., at 1040-1042, 103 S.Ct., at 3476-3477.2 For reasons this case illustrates, I would choose the opposite plain statement rule. I would presume, absent a plain statement to the contrary, that a state court's decision of the kind here at issue rests on an independent state-law ground.3

Widespread reliance on computers to store and convey information generates, along with manifold benefits, new possibilities of error, due to both computer malfunctions and operator mistakes. Most germane to this case, computerization greatly amplifies an error's effect, and correspondingly intensifies the need for prompt correction; for inaccurate data can infect not only one agency, but the many agencies that share access to the data-base. The computerized databases of the FBI's National Crime Information Center (NCIC), to take a conspicuous example, contain over 23 million records, identifying, among other things, persons and vehicles sought by law enforcement agencies nationwide. See Hearings before the Subcommittee on the Departments of Commerce, Justice, and State, the Judiciary, and Related Agencies of the House Committee on Appropriations, 102d Cong., 2d Sess., pt. 2B, p. 467 (1992). NCIC information is available to approximately 71,000 federal, state, and local agencies. See Hearings before the Subcommittee on the Departments of Commerce, Justice, and State, the Judiciary, and Related Agencies of the House Committee on Appropriations, 103d Cong., 1st Sess., pt. 2A, p. 489 (1993). Thus, any mistake entered into the NCIC spreads nationwide in an instant.

Isaac Evans' arrest exemplifies the risks associated with computerization of arrest warrants. Though his arrest was in fact warrantless—the warrant once issued having been quashed over two weeks before the episode in suit—the computer reported otherwise. Evans' case is not idiosyncratic. Rogan v. Los Angeles, 668 F.Supp. 1384 (CD Cal.1987), similarly indicates the problem. There, the Los Angeles Police Department, in 1982, had entered into the NCIC computer an arrest warrant for a man suspected of robbery and murder. Because the suspect had been impersonating Terry Dean Rogan, the arrest warrant erroneously named Rogan. Compounding the error, the Los Angeles Police Department had failed to include a description of the suspect's physical characteristics. During the next two years, this incorrect and incomplete information caused Rogan to be arrested four times, three times at gunpoint, after stops for minor traffic infractions in Michigan and Oklahoma. See id., at 1387-1389.4 In another case of the same genre, the District Court observed:

"Because of the inaccurate listing in the NCIC computer, defendant was a 'marked man' for the five months prior to his arrest. . . . At any time . . . a routine check by the police could well result in defendant's arrest, booking, search and detention. . . . Moreover, this could happen anywhere in the United States where law enforcement officers had access to NCIC information. Defendant was subject to being deprived of his liberty at any time and without any legal basis." United States v. Mackey, 387 F.Supp. 1121, 1124 (Nev.1975).

In the instant case, the Court features testimony of the chief clerk of the justice court in East Phoenix to the effect that errors of the kind Evans encountered are reported only "on[c]e every three or four years." Ante, at __ (citing App. 37). But the same witness also recounted that, when the error concerning Evans came to light, an immediate check revealed that three other errors of the very same kind had occurred on "that same day." Ibid.; see ante, at __, and n. 3 (STEVENS, J., dissenting).

This Court and the Arizona Supreme Court hold diverse views on the question whether application of an exclusionary rule will reduce the incidence of erroneous computer data left without prompt correction. Observing that "court clerks are not adjuncts to the law enforcement team engaged in the often competitive enterprise of ferreting out crime," the Court reasons that "there is no basis for believing that application of the exclusionary rule in these circumstances will have a significant effect on court employees responsible for informing the police that a warrant has been quashed." Ante, at __. In the Court's view, exclusion of evidence, even if capable of deterring police officer errors, cannot deter the carelessness of other governmental actors.5 Whatever federal precedents may indicate—an issue on which I voice no opinion—the Court's conclusion is not the lesson inevitably to be drawn from logic or experience.

In this electronic age, particularly with respect to recordkeeping, court personnel and police officers are not neatly compartmentalized actors. Instead, they serve together to carry out the State's information-gathering objectives. Whether particular records are maintained by the police or the courts should not be dispositive where a single computer database can answer all calls. Not only is it artificial to distinguish between court clerk and police clerk slips; in practice, it may be difficult to pinpoint whether one official, e.g., a court employee, or another, e.g., a police officer, caused the error to exist or to persist. Applying an exclusionary rule as the Arizona court did may well supply a powerful incentive to the State to promote the prompt updating of computer records. That was the Arizona Supreme Court's hardly unreasonable expectation. The incentive to update promptly would be diminished if court-initiated records were exempt from the rule's sway.

The debate over the efficacy of an exclusionary rule reveals that deterrence is an empirical question, not a logical one. "It is one of the happy incidents of the federal system that a single courageous State may, if its citizens choose, serve as a laboratory; and try novel social and economic experiments without risk to the rest of the country." New State Ice Co. v. Liebmann, 285 U.S. 262, 311, 52 S.Ct. 371, 386, 76 L.Ed. 747 (1932) (Brandeis, J., dissenting). With that facet of our federalism in mind, this Court should select a jurisdictional presumption that encourages States to explore different means to secure respect for individual rights in modern times.

Historically, state laws were the source, and state courts the arbiters, of individual rights. Linde, First Things First: Rediscovering the States' Bills of Rights, 9 U.Balt.L.Rev. 379, 382 (1980). The drafters of the federal Bill of Rights looked to provisions in state constitutions as models. Id., at 381. Moreover, many States that adopted constitutions after 1789 modeled their bills of rights on pre-existing state constitutions, rather than on the federal Bill of Rights. Ibid. And before this Court recognized that the Fourteenth Amendment—which constrains actions by States—incorporates provisions of the federal Bill of Rights, state constitutional rights, as interpreted by state courts, imposed the primary constraints on state action. Brennan, State Constitutions and the Protection of Individual Rights, 90 Harv.L.Rev. 489, 501-502 (1977).

State courts interpreting state law remain particularly well situated to enforce individual rights against the States. Institutional constraints, it has been observed, may limit the ability of this Court to enforce the federal constitutional guarantees. Sager, Fair Measure: The Legal Status of Underenforced Constitutional Norms, 91 Harv.L.Rev. 1212, 1217-1218 (1978). Prime among the institutional constraints, this Court is reluctant to intrude too deeply into areas traditionally regulated by the States. This aspect of federalism does not touch or concern state courts interpreting state law.

Under Long, when state courts engage in the essential process of developing state constitutional law, they may insulate their decisions from this Court's review by means of a plain statement of intent to rest upon an independent state ground. The plain statement option does not, however, make pleas for reconsideration of the Long presumption much ado about nothing.6 Both on a practical and on a symbolic level, the presumption chosen matters.

The presumption is an imperfect barometer of state courts' intent. Although it is easy enough for a state court to say the requisite magic words, the court may not recognize that its opinion triggers Long's plain statement requirement. "[A]pplication of Long's presumption depends on a whole series of 'soft' requirements: the state decision must 'fairly appear' to rest 'primarily' on federal law or be 'interwoven' with federal law, and the independence of the state ground must be 'not clear' from the face of the state opinion. These are not self-applying concepts." P. Bator, D. Meltzer, P. Mishkin, & D. Shapiro, Hart and Wechsler's The Federal Courts and the Federal System 552 (3d ed. 1988) (hereinafter Hart and Wechsler); cf. Coleman v. Thompson, 501 U.S. 722, 735-740, 111 S.Ct. 2546, 2556-2560, 115 L.Ed.2d 640 (1991) (declining to apply Long presumption to summary dismissal order).

Can the highest court of a State satisfy Long's "plain statement" requirement in advance, through a blanket disclaimer? The New Hampshire Supreme Court, for example, has declared: "We hereby make clear that when this court cites federal or other State court opinions in construing provisions of the New Hampshire Constitution or statutes, we rely on those precedents merely for guidance and do not consider our results bound by those decisions." State v. Ball, 124 N.H. 226, 233, 471 A.2d 347, 352 (1983). See also State v. Kennedy, 295 Ore. 260, 267, 666 P.2d 1316, 1321 (1983) ("Lest there be any doubt about it, when this court cites federal opinions in interpreting a provision of Oregon law, it does so because it finds the views there expressed persuasive, not because it considers itself bound to do so by its understanding of federal doctrines."). This Court's stated reluctance to look beneath or beyond the very state-court opinion at issue in order to answer the jurisdictional question, see Long, 463 U.S., at 1040, 103 S.Ct., at 3476, may render such blanket declarations ineffective. Cf. Hart and Wechsler 553 ("[T]he Court's protestations—that its presumption shows greater respect for state courts than asking them to clarify their opinions—ring hollow: Long simply puts the burden of clarification on the state court in advance.").

Application of the Long presumption has increased the incidence of nondispositive U.S. Supreme Court determinations—instances in which state courts, on remand, have reinstated their prior judgments after clarifying their reliance on state grounds. Westling, Advisory Opinions and the "Constitutionally Required" Adequate and Independent State Grounds Doctrine, 63 Tulane L.Rev. 379, 389, and n. 47 (1988) (pre-Long, i.e., between January 1, 1978, and June 30, 1983, 14.3% of decisions (2 of 14) involving potentially adequate and independent state grounds were reinstated on state grounds upon remand; post-Long, i.e., between July 1, 1983, and January 1, 1988, 26.7% of such decisions (4 of 15) were reinstated on remand). Even if these reinstatements do not render the Supreme Court's opinion technically "advisory," see Hart and Wechsler 537, they do suggest that the Court unnecessarily spent its resources on cases better left, at the time in question, to state-court solution.

The Long presumption, in sum, departs from the traditional understanding that "every federal court is 'without jurisdiction' unless 'the contrary appears affirmatively from the record.' " Delaware v. Van Arsdall, 475 U.S. 673, 692, 106 S.Ct. 1431, 1442, 89 L.Ed.2d 674 (1986) (STEVENS, J., dissenting) (quoting King Bridge Co. v. Otoe County, 120 U.S. 225, 226, 7 S.Ct. 552, 30 L.Ed. 623 (1887)). And it is out of sync with the principle that this Court will avoid constitutional questions when an alternative basis of decision fairly presents itself. Ashwander v. TVA, 297 U.S. 288, 346-347, 56 S.Ct. 466, 482-483, 80 L.Ed. 688 (1936) (Brandeis, J., concurring). Most critically, as this case shows, the Long presumption interferes prematurely with state-court endeavors to explore different solutions to new problems facing modern society.

I recognize that "[s]ince Long, we repeatedly have followed [its] 'plain statement' requirement," Harris v. Reed, 489 U.S. 255, 261, n. 7, 109 S.Ct. 1038, 1042, n. 7, 103 L.Ed.2d 308 (1989), and that precedent ought not be overruled absent strong cause. But the Long ruling itself did

"a virtual about-face regarding the guidelines for determining the reviewability of state court decisions in situations where the state court opinion is not absolutely clear about the bases on which it rests. The traditional presumption was that the Court lacked jurisdiction unless its authority to review was clear on the face of the state court opinion. When faced with uncertainty, the Court in the past occasionally remanded such cases to the state court for clarification. But more commonly, the Court would deny jurisdiction where there was uncertainty." G. Gunther, Constitutional Law 56 (12th ed. 1991).

Restoring a main rule "deny[ing] jurisdiction where there [is] uncertainty," ibid., would stop this Court from asserting authority in matters belonging, or at least appropriately left, to the States' domain. Cf. Erie R. Co. v. Tompkins, 304 U.S. 64, 77-80, 58 S.Ct. 817, 822-823, 82 L.Ed. 1188 (1938). Recognizing that "adequate state grounds are independent unless it clearly appears otherwise," Long, 463 U.S., at 1066, 103 S.Ct., 3489 (STEVENS, J., dissenting),7 would also avoid premature settlement of important federal questions. The submission for the United States is telling in this regard. While filing in support of petitioner, the United States acknowledges the problem occasioned by "erroneous information contained in law enforcement computer-information systems," but does not see this case as a proper vehicle for a pathmarking opinion. The United States suggests that the Court "await a case in which relevant characteristics of such systems and the legal questions they pose can be thoroughly explored." Brief for United States as Amicus Curiae 13.

The Arizona Supreme Court found it "repugnant to the principles of a free society," 177 Ariz., at 204, 866 P.2d, at 872, to take a person "into police custody because of a computer error precipitated by government carelessness." Ibid. Few, I believe, would disagree. Whether, in order to guard against such errors, "the exclusionary rule is a 'cost' we cannot afford to be without," ibid., seems to me a question this Court should not rush to decide. The Court errs, as I see it, in presuming that Arizona rested its decision on federal grounds. I would abandon the Long presumption and dismiss the writ because the generally applicable obligation affirmatively to establish the Court's jurisdiction has not been satisfied.