Opinion
CHIN, J. We granted review in this case to determine whether federal constitutional principles require the suppression of evidence discovered by a state parole officer and police during a search they conducted without a warrant under the erroneous belief that defendant Gary Wayne Willis was on parole and subject to a warrantless search condition. On the facts of this case, we agree with the Court of Appeal that the so-called good faith exception to the exclusionary rule does not apply. Because the Court of [26]Appeal found the evidence admissible and affirmed defendant’s conviction under a legal theory that the Attorney General concedes is erroneous, we reverse the Court of Appeal’s judgment.
Factual Background
On March 27, 1996, while working out of the Bakersfield Police Department as part of the Kern County Narcotics Enforcement Team, Officer Joseph Mullins received a telephone call from an employee of a Bakersfield motel. The motel employee advised Mullins of “a high level of phone and foot traffic” involving room 221, which was registered to defendant. This information was significant to Mullins because he knew from experience that narcotics dealers commonly conduct transactions at rented motel rooms.
Mullins checked “department records,” which indicated that defendant had several prior arrests and/or convictions involving narcotics. Mullins also checked “the local criminal justice information system,” which indicated that defendant was required to register as a sex offender. Finally, Mullins “checked the parole book,” or “parole listing,” “in the Bakersfield Police Department,” which indicated that defendant was on parole. According to Mullins, the “parole book” was “provided to the Police Department every month.” The “listing” Mullins checked on March 27 was dated either March 6 or March 16. Mullins then conveyed all of this information to Diane Mora, a state parole officer from the California Department of Corrections (CDC), and showed her “the parole list.” Mora told Mullins “the list indicated [that defendant] was on active parole,” and she “directed” Mullins “to make a search” of defendant’s motel room.
Mora and Mullins then went to the motel “to conduct a parole search,” accompanied by Detective Hood of the Kern County Sheriffs Department and Officer Silvius of the Bakersfield Police Department. At the motel, Mullins confirmed from motel records that room 221 was registered to defendant and that there were “several phone calls in and out of that motel room.” Hood knocked on the door of room 221. Defendant asked who was there. Hood replied, “[I]t’s Bill.” Defendant responded, “Bill who, fuck you.” Hood replied, “[ijt’s the police, open the door, please.” Neither Hood nor anyone else announced their purpose. Defendant then opened the door.
When the door opened, Mullins saw “a large sheath knife” on defendant’s belt, a hypodermic syringe on a dresser in the room (which he later determined was empty), and a woman named Kathleen Moye. Accompanied by Mullins and Silvius, Mora and Hood entered the room and announced their intention to conduct a parole search. Defendant did not invite them in or give [27]them permission to search. He informed them he had been discharged from parole in June 1995, and that they could not do a parole search. He directed their attention to a certificate of discharge from the CDC, which correctly showed that he had, in fact, been discharged from parole nine months earlier, on June 29, 1995.
Because “the parole listing indicated [defendant] was on parole,” Mullins did not consider the certificate to be “conclusive of [defendant’s] parole status.” Mullins asked Mora to use the telephone in the motel office to check defendant’s parole status and he escorted defendant to a walkway outside of the room.1 Mullins felt that Mora was the proper person for this task because she had been a parole officer for several years and was “better acquainted with the workings of the system and how to confirm through the [CDC] with their [szc] records in Sacramento the true status.”
While Mullins and defendant were outside, Silvius, who had remained in the room with Hood and Moye, advised Mullins that Moye appeared to be under the influence of narcotics. Consistent with this information, Moye said that she had “used this afternoon,” and she identified a briefcase in the room that she said contained “a speed pipe.” Mullins then announced that he had enough information to obtain a search warrant and asked defendant to consent to a search “to save us the time and trouble of obtaining a search warrant.” According to Mullins, defendant eventually admitted the briefcase contained methamphetamine and consented to a search of both the room and the briefcase. After defendant and Mullins reentered the room, Silvius broke the briefcase’s combination lock, opened the briefcase, and inside found narcotics, syringes, spoons and a set of scales. Defendant was then arrested.
Defendant was subsequently charged by information with possession of a controlled substance for purpose of sale (Health & Saf. Code, § 11378) and misdemeanor possession of drug paraphernalia (Health & Saf. Code, § 11364).2 Defendant moved to suppress the evidence discovered in his motel room, arguing that the warrantless search violated the Fourth Amendment to the United States Constitution. In opposition, the prosecution argued that the search was valid for numerous reasons, primarily consent. Alternatively, the prosecution argued that even if the search was not valid, because the police “acted in good faith reliance on Parole’s representations,” the [28]evidence should not be excluded. After argument, the trial court denied the suppression motion by a minute order that stated no reasons. A jury subsequently convicted defendant of possessing a controlled substance for sale and misdemeanor possession of drug paraphernalia.
On appeal, the Court of Appeal held that the initial entry into the motel room was unconstitutional because the officers did not have a search warrant, defendant was not on parole, and he did not consent to the entry. It also rejected the Attorney General’s argument that even though the search was constitutionally invalid, the exclusionary rule does not apply because the police relied in good faith on the information that defendant was on parole. The court reasoned that the inaccurate information regarding defendant’s parole status was “attributable to the police executing the search,” because Mora was an “adjunct of the law enforcement team” in that she “actively participated in the search” and, as a state parole officer, is a “peace officer” under California law.
Nevertheless, the Court of Appeal affirmed the denial of the suppression motion, reasoning that the unconstitutional entry did not taint the police’s subsequent reasonable actions that actually led to discovery of the evidence. According to the court, the police “were authorized to ‘freeze’ the motel room” after the illegal entry while Mora investigated defendant’s parole status, they obtained “additional information” from their interactions with defendant and Mora during the freeze that “amounted to probable cause” to obtain a search warrant, “were authorized to secure the room to prevent destruction of evidence until” they obtained a warrant, and received consent to search the room and the briefcase before they could take steps to obtain a warrant. Thus, the court concluded, the evidence “was not the fruit of an unlawful parole search, but instead of prudent lawful police work.”
Both defendant and the People petitioned for rehearing. In the People’s petition, the Attorney General repeatedly conceded that the police acted unconstitutionally in entering the motel room. However, he argued that the court erred in holding that the police’s good faith reliance on the information regarding defendant’s parole status did not render the exclusionary rule inapplicable. Defendant, in his petition, argued in part that the court’s “freezing” theory was both procedurally improper—because the Attorney General had never raised it—and substantively incorrect. The court denied the petitions.
Defendant and the People petitioned for review. We granted both petitions.
[29]Discussion
Federal constitutional standards generally govern our review of claims that evidence is inadmissible because it was obtained during an unlawful search. (Cal. Const., art. I, § 28, subd. (d); People v. Woods (1999) 21 Cal.4th 668, 674 [88 Cal.Rptr.2d 88, 981 P.2d 1019].) The Attorney General concedes that under those standards, the initial entry into defendant’s motel room “was unconstitutional” because there was no search warrant, no valid parole condition in effect, and no other applicable exception to the warrant requirement to justify the warrantless search. The Attorney General also concedes that the evidence recovered during the unconstitutional search is not admissible under the Court of Appeal’s “freezing” theory. He explains that because “a Fourth Amendment violation occurred when the officers entered [the] motel room,” “it is irrelevant” that the observations they made after the illegal entry “during the freezing period” may have established probable cause to support issuance of a search warrant. The Attorney General’s explanation is consistent with existing law. (See Murray v. United States (1988) 487 U.S. 533, 540 [108 S.Ct. 2529, 2535, 101 L.Ed.2d 472] [observations made after unlawful warrantless entry “cannot be used to establish probable cause” for a search warrant]; Burrows v. Superior Court (1974) 13 Cal.3d 238, 251 [118 Cal.Rptr. 166, 529 P.2d 590] [consent to search given “immediately following an illegal entry or search” is invalid because it “is inseparable from the unlawful conduct”]; People v. Roberts (1956) 47 Cal.2d 374, 377 [303 P.2d 721] [search warrant based on observations made after unlawful entry is invalid].) Accordingly, we need not and do not examine either of these questions further.
However, the Attorney General argues that the evidence is admissible under what is commonly known as the good faith exception to the exclusionary rule, which the United States Supreme Court announced and applied in a trilogy of cases, United States v. Leon (1984) 468 U.S. 897 [104 S.Ct. 3405, 82 L.Ed.2d 677] (Leon), Illinois v. Krull (1987) 480 U.S. 340, 355 [107 S.Ct. 1160, 1170, 94 L.Ed.2d 364] (Krull), and Arizona v. Evans (1995) 514 U.S. 1 [115 S.Ct. 1185, 131 L.Ed.2d 34] (Evans)3 In these cases, the high court explained that the exclusionary rule does not, and cannot, cure the constitutional violation, which is fully accomplished by the illegal search itself. (Evans, supra, 514 U.S. at p. 10 [115 S.Ct. at pp. 1190-1191]; Leon, supra, 468 U.S at p. 906 [104 S.Ct. at pp. 3411-3412].) [30]Rather, 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. [Citations.]” (Evans, supra, 514 U.S. at p. 10 [115 S.Ct. at p. 1191].) Thus, its “‘prime purpose’” is to “ ‘effectuate’ ” the Fourth Amendment’s guarantee against unreasonable searches or seizures by “deter[ring] future unlawful police conduct.” (Krull, supra, 480 U.S. at p. 347 [107 S.Ct. at pp. 1165-1166].) Moreover, because the exclusionary rule is a “remedial device,” its application is “restricted to those situations in which its remedial purpose is effectively advanced.” (Ibid.) Thus, application of the exclusionary rule “ ‘is unwarranted’ ” where it would “ ‘not result in appreciable deterrence.’ ” (Evans, supra, 514 U.S. at p. 11 [115 S.Ct. at p. 1191].)
In Leon, the high court held that where police officers act in objectively reasonable reliance on a search warrant that is issued by a detached and neutral magistrate but is later found to be invalid for lack of probable cause, the deterrent effect of exclusion is insufficient to warrant the exclusionary rule’s application. (Leon, supra, 468 U.S. at p. 900 [104 S.Ct. at p. 3409].) In reaching this conclusion, the court considered exclusion’s potential effect first on judicial officers who issue warrants, and then on police officers who execute warrants and on the policies of their departments. (Id. at pp. 916-918 [104 S.Ct. at pp. 3417-3418].) Regarding the former, the court concluded that for three reasons, the potential behavioral effect on judicial officers is insufficient to justify exclusion. (Id. at p. 916 [104 S.Ct. at p. 3417].) “First, the exclusionary rule is designed to deter police misconduct rather than to punish the errors of judges and magistrates. Second, there exists no evidence suggesting that judges and magistrates are inclined to ignore or subvert the Fourth Amendment or that lawlessness among these actors requires application of the extreme sanction of exclusion, [f] Third, and most important, [there is] no basis ... for believing that exclusion . . . will have a significant deterrent effect on the issuing judge or magistrate. Many of the factors that indicate that the exclusionary rule cannot provide an effective ‘special’ or ‘general’ deterrent for individual offending law enforcement officers apply as well to judges or magistrates. And, to the extent that the rule is thought to operate as a ‘systemic’ deterrent on a wider audience, it clearly can have no such effect on individuals empowered to issue search warrants. Judges and magistrates are not adjuncts to the law enforcement team; as neutral judicial officers, they have no stake in the outcome of particular criminal prosecutions. . . . Imposition of the exclusionary sanction is not necessary meaningfully to inform [them] of their errors, and . . . admitting evidence obtained pursuant to a warrant while at the same time declaring that the warrant was somehow defective will [not] in any way reduce [their] professional incentives to comply with the Fourth Amendment, encourage them to repeat their mistakes, or lead to the granting of all [31]colorable warrant requests.” (Id. at pp. 916-917 [104 S.Ct. at pp. 3417-3418], ins. omitted.)
Regarding exclusion’s potential effect on individual law enforcement officers and the policies of their departments, the high court explained generally that the deterrence rationale for the exclusionary rule “ ‘necessarily assumes that the police have engaged in willful, or at the very least negligent, conduct ....’” (Leon, supra, 468 U.S. at p. 919 [104 S.Ct. at pp. 3418-3419].) Thus, exclusion is proper “ ‘only if it can be said that the law enforcement officer had knowledge, or may properly be charged with knowledge, that the search was unconstitutional. . . .’” (Ibid.) Given these underlying principles, the court concluded that exclusion will not further the exclusionary rule’s ends where “an officer acting with objective good faith has obtained a search warrant from a judge or magistrate and acted within its scope. In most such cases, there is no police illegality and thus nothing to deter. It is the magistrate’s responsibility to determine whether the officer’s allegations establish probable cause and, if so, to issue a warrant comporting in form with the requirements of the Fourth Amendment. In the ordinary case, an officer cannot be expected to question the magistrate’s probable-cause determination or his judgment that the form of the warrant is technically sufficient. ‘[0]nce the warrant issues, there is literally nothing more the policeman can do in seeking to comply with the law.’ [Citation.] Penalizing the officer for the magistrate’s error, rather than his own, cannot logically contribute to the deterrence of Fourth Amendment violations.” (Id. at pp. 920-921 [104 S.Ct. at p. 3419], fns. omitted.) Thus, “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.” (Id. at p. 922 [104 S.Ct. at p. 3420].)
However, suppression remains appropriate where an officer’s reliance on a search warrant was not “objectively reasonable,” i.e., the officer had “no reasonable grounds for believing that the warrant was properly issued.” (Leon, supra, 468 U.S. at pp. 922-923 [104 S.Ct. at p. 3420].) “ ‘Grounding the [good faith exception] in objective reasonableness . . . retains the value of the exclusionary rule as an incentive for the law enforcement profession as a whole to conduct themselves in accord with the Fourth Amendment.’ [Citations.]” (Id. at p. 919, fn. 20 [104 S.Ct. at p. 3419].) Thus, in determining whether the good faith exception applies, “[i]t is necessary to consider the objective reasonableness, not only of the officers who eventually executed a warrant, but also of the officers who originally obtained it or who provided information material to the probable-cause determination.” (Id. at p. 923, fn. 24 [104 S.Ct. at p. 3420].) For example, the court cautioned, [32]“[njothing” in Leon “suggests” that an officer may use “a ‘bare bones’ affidavit” to obtain a warrant “and then rely on colleagues who are ignorant of the circumstances under which the warrant was obtained to conduct the search. [Citation.]” (Ibid.) Moreover, the good faith exception does not apply “where the issuing magistrate wholly abandoned his judicial role,” where the affidavit was “ ‘so lacking in indicia of probable cause as to render official belief in its existence entirely unreasonable,’ ” or where the warrant was “so facially deficient—i. e., in failing to particularize the place to be searched or the things to be seized—that the executing officers cannot reasonably presume it to be valid.” (Id. at p. 923 [104 S.Ct. at p. 3421].) Thus, courts must determine “on a case-by-case basis” whether the circumstances of an invalid search pursuant to a warrant require the exclusionary rule’s application. (Id. at p. 918 [104 S.Ct. at p. 3418].) “[T]he government has the burden of establishing ‘objectively reasonable’ reliance [citation] . . . .” (People v. Camarella (1991) 54 Cal.3d 592, 596 [286 Cal.Rptr. 780, 818 P.2d 63] (Camarella).)
Three years after Leon, the high court in Krull used a similar analysis in finding the good faith exception applicable where “officers act in objectively reasonable reliance upon a statute authorizing warrantless administrative searches,” and the statute is later “found to violate the Fourth Amendment.” (Krull, supra, 480 U.S. at p. 342 [107 S.Ct. at p. 1163].) The court first reasoned that excluding evidence obtained under these circumstances “would have as little deterrent effect on the officer’s actions” as would excluding evidence obtained in objectively reasonable reliance on a search warrant. (Id. at p. 349 [107 S.Ct. at p. 1167].) “Unless a statute is clearly unconstitutional, an officer cannot be expected to question the judgment of the legislature that passed the law. If the statute is subsequently declared unconstitutional, excluding evidence obtained pursuant to it prior to such a judicial declaration will not deter future Fourth Amendment violations by an officer who has simply fulfilled his responsibility to enforce the statute as written. . . . ‘Penalizing the officer for the [legislature’s] error, rather than his own, cannot logically contribute to the deterrence of Fourth Amendment violations.’ [Citation.]” (Id. at pp. 349-350 [107 S.Ct. at p. 1167].)
Nor, the court reasoned in Krull, is exclusion justified by its potential effect on legislators. (Krull, supra, 480 U.S. at p. 350 [107 S.Ct. at p. 1167].) First, “legislators, like judicial officers, are not the focus of the [exclusionary] rule,” which is “aimed at deterring police misconduct.” (Ibid.) Second, there is no “evidence to suggest that legislators ‘are inclined to ignore or subvert the Fourth Amendment.’ [Citation.] Although legislators are not ‘neutral judicial officers,’ as are judges and magistrates [citation], neither are they ‘adjuncts to the law enforcement team.’ [Citation.]” (Id. at pp. 350-351 [33][107 S.Ct. at p. 1167].) In performing their duty to establish a criminal justice system, “legislators’ deliberations of necessity are significantly different from the hurried judgment of a law enforcement officer ‘engaged in the often competitive enterprise of ferreting out crime.’ [Citation.]” (Id. at p. 351 [107 S.Ct. at p. 1167].) Moreover, they must “take an oath to support the Federal Constitution” and courts must presume that they act constitutionally. (Ibid) Finally, there is “no reason to believe that applying the exclusionary rule will have” a significant deterrent effect on legislators. (Id. at p. 352 [107 S.Ct. at p. 1168].) “Legislators enact statutes for broad, programmatic purposes, not for the purpose of procuring evidence in particular criminal investigations. Thus, it is logical to assume that the greatest deterrent to the enactment of unconstitutional statutes ... is the power of the courts to invalidate such statutes. Invalidating a statute informs the legislature of its constitutional error, affects the admissibility of all evidence obtained subsequent to the constitutional ruling, and often results in the legislature’s enacting a modified and constitutional version of the statute, as happened in this very case. There is nothing to indicate that applying the exclusionary rule to evidence seized pursuant to the statute prior to the declaration of its invalidity will act as a significant, additional deterrent.” (Ibid) In any event, any “incremental deterrent” effect of exclusion in this context, when “weighed against the ‘substantial social costs’ ” of exclusion, is insufficient to justify the exclusionary rule’s application. (Ibid)
As in Leon, the court in Krull stressed that the good faith exception does not apply if the officer does not act reasonably, and that “the standard of reasonableness . . . is an objective one; [it] does not turn on the subjective good faith of individual officers. [Citation.]” (Krull, supra, 480 U.S. at p. 355 [107 S.Ct. at p. 1170].) Thus, “a law enforcement officer [cannot] be said to have acted in good-faith reliance upon a statute if its provisions are such that a reasonable officer should have known that the statute was unconstitutional. [Citation.]” (Ibid) Nor can “[a] statute . . . support objectively reasonable reliance if, in passing the statute, the legislature wholly abandoned its responsibility to enact constitutional laws.” (Ibid.)
More recently, in Evans, the high court discussed the good faith exception as applied to court employees. There, upon entering the defendant’s name into a computer terminal during a traffic stop, a police officer received notice of an outstanding arrest warrant. During the ensuing arrest, the officer found marijuana in the defendant’s car. The police then reported the arrest to the justice court, which advised that the arrest warrant had been quashed 17 days before the arrest. The defendant later moved to suppress the evidence, arguing that the good faith exception did not apply because police error, rather than judicial error, caused the invalid arrest. (Evans, supra, 514 U.S. [34]at p. 4 [115 S.Ct. at pp. 1187-1188].) Testimony at the suppression hearing indicated that the error in the police computer may have been caused by a court clerk in failing to inform the sheriffs office that the warrant had been quashed, rather than by a records clerk in the sheriffs office. (Id. at p. 5 [115 S.Ct. at p. 1188].) The trial court made no factual finding on this issue, holding that exclusion was required whether the error was caused by the court clerk or by the police. (Ibid.) The Arizona Supreme Court agreed with the trial court and “rejected” a “ ‘distinction . . . between clerical errors committed by law enforcement personnel and similar mistakes by court employees.’ [Citation.]” (Id. at p. 6 [115 S.Ct. at p. 1189].)
The high court in Evans held that the refusal of the Arizona courts to distinguish between errors of the police and errors of the court was “contrary to the reasoning” of Leon and Krull. (Evans, supra, 514 U.S. at p. 14 [115 S.Ct. at p. 1193].) The court explained that under those cases, “[i]f 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.” (Ibid.) Under these circumstances, exclusion “could not be expected to alter the behavior of the arresting officer,” because he “ ‘[was] bound to arrest’ ” and “ ‘would [have been] derelict in his duty if he’ ” had not. (Id. at p. 15 [115 S.Ct. at p. 1193].) Nor, the high court reasoned, could exclusion be justified by its effect on court clerks. “First, ... the exclusionary rule was historically designed as a means of deterring police misconduct, not mistakes by court employees. [Citations.] Second, [the defendant] offer[ed] 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. [Citations.] 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. [Citation.] [f] 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 [citation], they have no stake in the outcome of particular criminal prosecutions. [Citations.] 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. [Citations.]” (Id. at pp. 14-15 [115 S.Ct. at p. 1193].) Because the Arizona courts ordered suppression without determining whether police or court employees were responsible for the computer error, the high court therefore reversed the judgment and remanded for further proceedings. (Id. at p. 16 [115 S.Ct. at p. 1194].)
[35]The Attorney General correctly contends that under these cases, application of the exclusionary rule depends on the source of the error or misconduct that led to the unconstitutional search and whether, in light of that source, the deterrent effect of exclusion is sufficient to warrant that sanction. (See Krull, supra, 480 U.S. at p. 360, in. 17 [107 S.Ct. at pp. 1172-1173] [whether exclusionary rule applies “in a particular context depends significantly upon the actors who are making the relevant decision that the rule is designed to influence”].) As to the source of the error here, the Attorney General states that the “sparse record” leaves “the precise duties and responsibilities of the person or persons responsible” for the error “unknown.” However, he argues the evidence “strongly suggests that the source of the erroneous information was the Department of Corrections.” In his initial briefing in the Court of Appeal, the Attorney General focused mainly on Mora, arguing that “it makes no sense to apply the exclusionary rule . . . where the negligence or mistake was due to a parole agent’s mistaken direction to police to conduct a search.” As we have noted, the Court of Appeal disagreed, finding the good faith exception inapplicable because Mora was “an adjunct of the law enforcement team.” The Attorney General now attributes the error to an “anonymous employee” of the CDC, “presumably an unsworn data entry clerk” who was “responsible for maintaining and updating the [parole] list.” In making this argument, the Attorney General cites Mullins’s testimony and several Penal Code provisions relating to the CDC’s authority and duties.
We agree with the Attorney General that the “sparse record” is inconclusive regarding the source of the error in this case. The testimony the Attorney General cites is of little help. Mullins testified that the “parole book” he checked was “provided to the Police Department every month.” However, Mullins did not indicate who provided the parole book; his vague statement does not point to the CDC as the source any more than it points to some police agency other than the Bakersfield Police Department. Mullins also testified that Mora told him “the list indicated [that defendant] was on active parole, and directed [him] to make a search of [defendant’s] motel room,” and that he believed Mora was the proper person to determine defendant’s parole status because she was “better acquainted with the workings of the system.” Again, this testimony does not identify the source of the parole listing or even exclude some other police agency as that source.
The statutes the Attorney General cites are of marginal help at best. Section 3056, which states merely that “[prisoners on parole shall remain under the legal custody of’ the CDC, provides little support for the conclusion that the CDC prepared the parole listing Mora and Mullins reviewed. Section 3058.5 requires the CDC to “provide within 10 days, upon request, [36]to the chief of police of a city or the sheriff of the county, information available to the department. . . concerning persons then on parole who are or may be residing or temporarily domiciled in that city or county.” However, the record contains no evidence that the Bakersfield Police Department made an information request pursuant to section 3058.5 or that the parole listing Mullins and Mora reviewed was prepared and provided by the CDC in response to such a request. Section 3003 requires the CDC to provide “local law enforcement agencies” with available information “regarding a paroled inmate who is released in their jurisdictions.” However, section 3003 did not impose this requirement statewide until after the search in this case, so it is irrelevant to determining the source of the error here. (See Stats. 1997, ch. 680, § 2.)
In any event, even if the record suggested that the CDC prepared the parole list, it does not indicate who was responsible for the parole list’s error regarding defendant’s parole status or how the error occurred. The record contains no evidence suggesting that a data entry clerk, rather than a parole officer, prepared the parole list, or that the error in the list was caused by the person who prepared it, rather than by a parole officer who failed to update defendant’s file or forward the information to the appropriate person. The Attorney General recognized this ambiguity in his rehearing petition in the Court of Appeal, asserting that the source of the error here was “the anonymous parole agent (or, more likely, the unsworn data entry clerk) who failed to update the parole book sent to police departments.” Indeed, the record does not even foreclose the possibility that Mora herself prepared the list, or that she failed to update the records in June 1995 when defendant was discharged from parole and then simply forgot about his discharge when she ordered and conducted the search of his motel room nine months later in March 1996.4 Thus, we agree with the Attorney General that “the sparse record” here fails to show “the precise duties and responsibilities of the person or persons responsible” for the error.
However, we do not agree with the Attorney General that defendant bore the burden of producing evidence on this question and, therefore, he is responsible for the record’s inadequacy. Where, as here, the prosecution invokes the good faith exception, the government has “the burden ... to prove that exclusion of the evidence is not necessary because of [that] exception.” (People v. Turnage (1994) 162 Ill.2d 299 [205 Ill.Dec. 118, 642 N.E.2d 1235, 1241] (Turnage)) Thus, “the government has the burden of [37]establishing ‘objectively reasonable’ reliance” under Leon. (Camarella, supra, 54 Cal.3d at p. 596.) Establishing that the source of the error acted objectively reasonably is part of that burden. (See Turnage, supra, 642 N.E.2d at p. 1241 [state failed to satisfy its burden where it relied only on arresting officer’s good faith and offered no evidence that those who sought invalid warrant or signaled its vitality acted objectively reasonably]; cf. Commonwealth v. Hecox (1993) 35 Mass.App. 277 [619 N.E.2d 339, 341-344] (Hecox) [government failed to satisfy its burden where it offered no evidence that police were not at fault in failing to correct their records]; Ott v. State (1992) 325 Md. 206, 223, fn. 5 [600 A.2d 111, 119] [government failed to satisfy its burden where it offered no evidence of the length of time reasonably necessary to update information in sheriffs computer, which depends in part on who had responsibility for keeping information current].)
The Attorney General errs in asserting that in Leon, Krull, and Evans, the high court “repeatedly hinted that it is the defendant who bears the burden of production” regarding the source of the error and his or her duties. The discussion the Attorney General cites from those cases relates not to whether a defendant has offered evidence to identify the source of the error, but to whether the record contains evidence that the identified source of the error is inclined to subvert or ignore the Fourth Amendment. (Evans, supra, 514 U.S. at pp. 14-15 [115 S.Ct. at p. 1193] [defendant “offers no evidence that court employees are [so] inclined”]; Krull, supra, 480 U.S. at p. 351 [107 S.Ct. at p. 1168] [“There is no evidence suggesting,” and “we are given no basis for believing that legislators are [so] inclined”]; Leon, supra, 468 U.S. at p. 916 [104 S.Ct at p. 3417] [“there exists no evidence suggesting that judges and magistrates are [so] inclined”].) Moreover, in establishing the good faith exception, the high court stated in Leon that “[w]hen officers have acted pursuant to a warrant, the prosecution should ordinarily be able to establish objective good faith without a substantial expenditure of judicial time.” {Leon, supra, 468 U.S. at p. 924 [104 S.Ct. at p. 3421], italics added.) Numerous courts, including this one, have cited this statement from Leon in holding that the government has the burden to prove facts warranting application of the good faith exception. (Camarella, supra, 54 Cal.3d at p. 596; U.S. v. Corral-Corral (10th Cir. 1990) 899 F.2d 927, 932; United States v. Maggitt (5th Cir. 1985) 778 F.2d 1029, 1034; U.S. v. Conner (N.D. Iowa 1996) 948 F.Supp. 821, 852; U.S. v. Turner (D.Vt. 1989) 713 F.Supp. 714, 721, fn. 6; Hoay v. State (2001) 75 Ark.App. 103 [55 S.W.3d 782, 785]; Turnage, supra, 642 N.E.2d at p. 1241.)
Our conclusion regarding the burden of proof is also consistent with the high court’s decision in Miranda v. Arizona (1966) 384 U.S. 436 [86 S.Ct. 1602, 16 L.Ed.2d 694, 10 A.L.R.3d 974]. There, the court held in part that if [38]a person indicates before or during a custodial interrogation that he or she wants to consult with an attorney before speaking to police, then “the interrogation must cease.” (Id. at p. 474 [86 S.Ct. at p. 1628]; see id. at pp. 444-445, 473-474 [86 S.Ct. at pp. 1612-1613, 1627-1628].) The court also held that “[i]f the interrogation continues without the presence of an attorney and a statement is taken, a heavy burden rests on the government to demonstrate that the defendant knowingly and intelligently waived his privilege against self-incrimination and his right to retained or appointed counsel. [Citation.]” (Id. at p. 475 [86 S.Ct. at p. 1628].) The court explained that this burden “is rightly on [the] shoulders” of the government because it “is responsible for establishing the isolated circumstances under which the interrogation takes place and has the only means of making available corroborated evidence of warnings given during incommunicado interrogation . . . .” (Ibid.; see also Brown v. Illinois (1975) 422 U.S. 590, 604 [95 S.Ct. 2254, 2262, 45 L.Ed.2d 416] [prosecution bears “burden of showing” that evidence is not the fruit of illegal arrest or search and is therefore admissible].) Similarly, in this case, because the government is responsible for establishing the circumstances under which the error and the warrantless search occurred and is in a much better position to obtain evidence regarding those circumstances, the burden rightly falls on its shoulders. (Cf. State v. Mance (1996) 62 Wash.App. 539 [918 P.2d 527, 530] [state has “burden of proving” that police’s delay in failing to correct their records “was reasonable,” because the relevant facts are peculiarly within their knowledge and control].) We therefore disagree with the Attorney General’s assertion that defendant bore the burden of producing evidence to identify the source of the error and his or her duties. The prosecution bore the burden of proof on this point, and it failed to sustain that burden. (Cf. United States v. Hoffman (9th Cir. 1979) 607 F.2d 280, 284 [government does not satisfy its burden of justifying warrantless search “by leading a court to speculate about what ‘may’ or ‘might’ have been the circumstances surrounding the warrantless search”].)
In any event, we conclude that the exclusionary rule applies in this case whether the source of the error was Mora, as the Attorney General contended in the Court of Appeal, or a CDC data entry clerk, as the Attorney General now speculates. As to Mora, we begin, as Leon, Krull, and Evans direct, by considering exclusion’s potential effect on the behavior of parole agents. According to the high court, the “most important” question is whether there is a basis to believe that exclusion under the circumstances here will have a significant effect on parole agents. (Evans, supra, 514 U.S. at p. 15 [115 S.Ct. at pp. 1193-1194]; Leon, supra, 468 U.S. at p. 916 [104 S.Ct. at p. 3417].) Under the high court’s analysis, a key consideration in answering this question is whether parole agents are “adjuncts to the law [39]enforcement team.” (Evans, supra, 514 U.S. at p. 15 [115 S.Ct. at pp. 1193-1194]; Krull, supra, 480 U.S. at pp. 350-351 [107 S.Ct. at pp. 1167-1168]; Leon, supra, 468 U.S. at p. 916 [104 S.Ct. at p. 3417].)
On the record here, we agree with the Court of Appeal that Mora was, in fact, an adjunct to the law enforcement team. Like the Court of Appeal, we first find it significant that Mora, as a CDC parole officer, is “a peace officerQ” under California law. (§ 830.5.) Her authority as a peace officer extends to “the rendering of mutual aid to any other law enforcement agency.” (§ 830.5, subd. (a)(5).) Thus, she often works hand in hand with police, as she did in this case. Her authority also “extend[s]” to parole conditions and transportation of any state parolee, escapes by any state inmates or wards, and “violations of any penal provisions of law which are discovered while performing” her “usual or authorized duties.” (§ 830.5, subd. (a)(4).) She may “carry firearms” under specified terms and conditions (§ 830.5), and may make arrests with or (under specified circumstances) without a warrant (§ 836). Under section 3067, a California parolee must agree “to be subject to search or seizure by a parole officer or other peace officer at any time of the day or night, with or without a search warrant and with or without cause.” (§ 3067, subd. (a); see also Cal. Code Regs., tit. 15, §2511, subd. (b).) According to the relevant legislative history, the Legislature passed section 3067 “to provide for search of parolees by law enforcement officers without cause” and to “give . . . local law enforcement officers the tools they need to adequately supervise . . . parolees.”5 (Sen. Com. on Crim. Proc., Rep. on Assem. Bill No. 2284 (1995-1996 Reg. Sess.) June 25, 1996, p. 4, italics added.) Under state regulation, Mora must “seize []” any “contraband or evidence of illegal activity” she discovers while “conducting” these searches and, under specified circumstances, may “delegate[]” her “authority to search or arrest a parolee ... to another law enforcement agency.” (Cal. Code Regs., tit. 15, § 3701.1, subds. (a), (d).) Before exercising her authority, she was required to “satisfactorily complete an introductory course of training prescribed by the Commission on Peace Officer Standards and Training.” (§ 832.) Finally, “while engaged in the performance of [her] duties . . . and for the purpose of carrying out the primary function of [her] employment,” Mora’s authority as a peace officer “extends to any place in the state.” (§ 830.5.) While transporting prisoners or apprehending escaped prisoners, she is a “peace officer” even when acting outside of California. (Cal. Code Regs., tit. 15, § 3291, subd. (c).)
Both this court and the United States Supreme Court have looked to these provisions in considering the nature of California parole officers. Citing [40]section 830.5, we have explained that a parole officer’s “status ... as a peace officer” is one of the “sources” of a parole officer’s “authority to restrain [a] parolee.” (In re Law (1973) 10 Cal.3d 21, 24, fn. 2 [109 Cal.Rptr. 573, 513 P.2d 621].) We have also explained that the “purpose” of section 830.5 is “to authorize the named persons to exercise the statutory powers of a peace officer.” (Dyas v. Superior Court (1974) 11 Cal.3d 628, 635-636, fn. 3 [114 Cal.Rptr. 114, 522 P.2d 674].) In Cabell v. Chavez-Salido (1982) 454 U.S. 432 [102 S.Ct. 735, 70 L.Ed.2d 677], the high court reviewed some of these California statutes and found that “the unifying character of all categories of peace officers,” including “parole . . . officers” under section 830.5, “is their law enforcement function.” The court also reasoned that “[t]he general law enforcement character of all California ‘peace officers’ is underscored by the fact that all have the power to make arrests, § 836, and all receive a course of training in the exercise of their respective arrest powers and in the use of firearms. § 832.” (Cabell, supra, 454 U.S. at pp. 443-444 [102 S.Ct. at pp. 741-742].) Thus, based on California statutes, the high court concluded that persons classified under California law as peace officers, including parole agents, serve a “law enforcement function” and have a “general law enforcement character.” (Ibid)
Like the Court of Appeal, we also find it significant that Mora took an active role in the search in this case. As we have explained, Mullins did not make the decision to search; Mora authorized the search and directed Mullins to carry it out after he presented her with the parole list and the information he had received. Mora then went to defendant’s motel room with police to conduct a search for evidence of narcotics activity. She and the police officers were acting with a unity of purpose: investigating crime. Thus, as relevant to applying the exclusionary rule, Mora bears little resemblance to the neutral and detached judicial officers and court clerks in Leon and Evans. Nor does she resemble the legislators in Krull, who, the high court found, do not act “for the purpose of procuring evidence in particular criminal investigations.” (Krull, supra, 480 U.S. at pp. 352 [107 S.Ct. at p. 1168].) Unlike those actors, Mora is an adjunct to the law enforcement team when she, as a peace officer under California law, conducts or participates in a search, and the threat of exclusion can be expected to alter her behavior. (Cf. Leon, supra, 468 U.S. at pp. 914, 923 [104 S.Ct. at pp. 3416, 3421] [good faith exception does not apply where “the issuing magistrate wholly abandon[s] his judicial role” and becomes “ ‘adjunct law enforcement officer’ ” by acting as a member of search party that is essentially a police operation].)
The Attorney General asserts that this conclusion is inconsistent with the high court’s decision in Pennsylvania Bd. of Probation and Parole v. Scott [41](1998) 524 U.S. 357 [118 S.Ct. 2014, 141 L.Ed.2d 344] (Scott). There, the court held that the exclusionary rule does not apply in parole revocation hearings where police officers or parole officers conduct an illegal search. (Id. at pp. 362-369 [118 S.Ct. at pp. 2019-2023].) Specifically, the Attorney General quotes the following passage in the court’s opinion: “Parole agents, in contrast to police officers, are not ‘engaged in the often competitive enterprise of ferreting out crime,’ [citation]; instead, their primary concern is whether their parolees should remain free on parole. Thus, their relationship with parolees is more supervisory than adversarial. [Citation.] It is thus ‘unfair to assume that the parole officer bears hostility against the parolee that destroys his neutrality; realistically the failure of the parolee is in a sense a failure for his supervising officer.’ [Citation.]” (Id. at p. 368 [118 S.Ct. at p. 2022].)
Unlike the Attorney General, we find that Scott supports our conclusion. Immediately after the passage the Attorney General quotes, the high court observed that “in some instances parole officers may act like police officers and seek to uncover evidence of illegal activity . . . .” (Scott, supra, 524 U.S. at p. 369 [118 S.Ct. at p. 2022].) As we have explained, the record shows that Mora acted here in precisely that capacity. The high court then stated in Scott that when parole officers “act like police officers and seek to uncover evidence of illegal activity, they (like police officers) are undoubtedly aware that any unconstitutionally seized evidence that could lead to an indictment could be suppressed in a criminal trial.” (Scott, supra, 524 U.S. at p. 369 [118 S.Ct. at p. 2022], italics added.) Accordingly, the court continued, any evidence the parole officers in Scott uncovered during an illegal search “could have been inadmissible at trial if [the defendant] had been criminally prosecuted.” (Ibid., italics added.) This statement necessarily implies the court’s conclusion that as to parole agents who conduct searches, exclusion in a criminal trial would alter their behavior—i.e., deter them— enough to justify applying the exclusionary rule in a criminal trial, otherwise, under Leon, Evans, and Krull, which predated Scott, exclusion would be improper even in a criminal trial. Thus, Scott directly supports the conclusion that if Mora made the error that led to the illegal search of defendant’s motel room, the exclusionary rule applies in this criminal proceeding.
Indeed, our research indicates that both before and after Evans, courts have uniformly held that the exclusionary rule applies in a criminal proceeding where a parole officer obtains evidence during an unconstitutional search. For example, the federal Sixth Circuit Court of Appeals recently applied the exclusionary rule where parole officers, assisted by police whom the parole officers had invited along, conducted an unconstitutional search. [42](U.S. v. Payne (6th Cir. 1999) 181 F.3d 781, 784 (Payne).) Regarding the exclusionary rule’s deterrence objective, the court explained that the “zone of interest” of the parole officers who “primarily” conducted the search was “distinct from, but overlapped] with, an interest in enforcing the drug laws. Although the parole officer is interested in the parolee’s rehabilitation, the officer is also charged with monitoring compliance with various restrictions, including restrictions on the use of drugs. As in this case, parole officers often work closely with the police. Exempting evidence illegally obtained by a parole officer from the exclusionary rule would greatly increase the temptation to use the parole officer’s broad authority to circumvent the Fourth Amendment.” (Id. at p. 788.) Both this court and others have similarly applied the exclusionary rule in criminal proceedings where a parole officer conducts an unconstitutional search. (People v. Superior Court (Stevens) (1974) 12 Cal.3d 858, 860-861 [117 Cal.Rptr. 433, 528 P.2d 41]; Commonwealth v. Gayle (1996) [673 A.2d 927, 929-932]; State ex rel. Corgan v. King (1994) 1994 Okla. Crim. 7 [868 P.2d 743, 747-748]; Commonwealth v. Elliott (Ky.Ct.App. 1986) 714 S.W.2d 494; People v. Candelaria (1978) 63 A.D.2d 85 [406 N.Y.S.2d 783, 786].) We have not found, and the Attorney General has not cited, a single case holding to the contrary. Thus, exclusion is required here if we focus on Mora as the source of the error, as the Attorney General did in his briefing in the Court of Appeal.
The discussion in Payne suggests another distinction between this case and Leon, Evans, and Krull, in terms of the deterrent effect of applying the exclusionary rule. In Leon and Evans, the court reasoned that exclusion for errors by judges, magistrates, and court clerks cannot be expected to alter the behavior of police officers, who are in no position to question court directives. (Evans, supra, 514 U.S. at pp. 15-16 [115 S.Ct. at pp. 1193-1194]; Leon, supra, 468 U.S. at pp. 920-921 [104 S.Ct. at p. 3419].) As Leon more fully explains, an officer ordinarily “cannot be expected to question the magistrate’s probable-cause determination or his judgment that the form of the warrant is technically sufficient. ‘[0]nce the warrant issues, there is literally nothing more the policeman can do in seeking to comply with the law.’ [Citation.] Penalizing the officer for the magistrate’s error, rather than his own, cannot logically contribute to the deterrence of Fourth Amendment violations.” (Leon, supra, 468 U.S. at p. 921 [104 S.Ct. at p. 3419].) La.Krull, the court similarly reasoned that because arresting officers “cannot be expected to question the judgment of [a] legislature that passe[s] [a] law” authorizing warrantless searches, exclusion where an officer relies on such a statute cannot be expected to alter the officer’s behavior. (Krull, supra, 480 U.S. at p. 350 [107 S.Ct. at p. 1167].)
By contrast, in this case, Mullins could have done more than simply rely on Mora’s review of the parole list. As the high court has explained, the [43]Fourth Amendment “was in large part a reaction to the general warrants and warrantless searches that had so alienated the colonists and had helped speed the movement for independence.” (Chimel v. California (1969) 395 U.S. 752, 761 [89 S.Ct. 2034, 2039, 23 L.Ed.2d 685].) Thus, a warrantless entry of the home is the “ ‘chief evil’ ” to which the Fourth Amendment is directed. (Welsh v. Wisconsin (1984) 466 U.S. 740, 748 [104 S.Ct. 2091, 2096-2097, 80 L.Ed.2d 732].) Nevertheless, Mora, who may not have even been defendant’s parole agent (see in. 4, ante), apparently did nothing more than look at the parole list and confirm that it said what Mullins already knew it said: that defendant was on parole. Essentially, Mora “ ‘serve[d] merely as a rubber stamp’ ” for Mullins. (Leon, supra, 468 U.S. at p. 914 [104 S.Ct. at p. 3416].) Although Mullins and Mora were not in the field dealing with an unfolding situation when Mullins first received the information about defendant and consulted Mora, neither made any further attempt to verify the information on the parole list—by, for example, reviewing defendant’s case file or checking the CDC’s Sacramento records—before going to the motel and entering defendant’s room to conduct a search knowing they had no warrant. Thus, exclusion here is more likely to alter police behavior than it was in the high court cases.6 As Leon explained, “ ‘refusing to admit evidence gained’ ” from “ ‘negligent^ conduct which has deprived the defendant of some right’ ” will “ ‘hope[fully] instill in those particular investigating officers, or in their future counterparts, a greater degree of care toward the rights of an accused.’ ” (Leon, supra, 468 U.S. at p. 919 [104 S.Ct. at p. 3418]; see also Evans, supra, 514 U.S. at p. 17 [115 S.Ct. at p. 1194] (cone. opn. of O’Connor, J.) [police must “have acted reasonably in their reliance on the recordkeeping system itself” (italics omitted.)].) On the other hand, as the Sixth Circuit reasoned in Payne, declining to apply the exclusionary rule “would greatly increase the temptation to use the parole officer’s broad authority to circumvent the Fourth Amendment.”7 (Payne, supra, 181 F.3d at p. 788.)
Our conclusion regarding the exclusionary rule’s applicability in this case is the same even if we assume, as the Attorney General now speculates, that [44]the error was the fault of a CDC data entry clerk responsible for maintaining and updating the parole list. Again, the most important consideration is whether there is a basis to believe that exclusion under these circumstances will have a significant effect on such clerks, which turns in part on whether they are adjuncts to the law enforcement team. As a matter of logic, CDC clerks responsible for preparing or updating the parole list are adjuncts to the law enforcement. As we have explained, in California, all parolees are subject by law to warrantless search by parole officers. (§ 3067; Cal. Code Regs., tit. 15, § 2511, subd. (b).) As we have also explained, in carrying out such searches, parole officers from the CDC are exercising a law enforcement function and are part of the law enforcement team. Logically, then, employees of the same department who support parole officers in carrying out this law enforcement function—by preparing and maintaining parole lists indicating who is on active parole and subject to warrantless search—must be considered to be adjuncts to the law enforcement team.
Moreover, by statute, CDC employees responsible for parole records play a similar role in supporting the work of other California law enforcement officers—including police—who are also authorized by law (§ 3067) to conduct warrantless searches of parolees. Since 1981, section 3058.5 has required the CDC to provide information about parolees to city police chiefs and county sheriffs on request. (See Stats. 1981, ch. 1111, § 5, p. 4340.) By 1986, the CDC was reporting that law enforcement agencies in most cities and in all counties were routinely requesting and receiving information under this statute. (Assem. Com. on Public Safety, Rep. on Assem. Bill No. 3110 (1985-1986 Reg. Sess.) Apr. 14, 1986, p. 2; Sen. Com. on Judiciary, Rep. on Assem. Bill No. 3110 (1985-1986 Reg. Sess.) p. 3.)
In 1997, the Legislature enacted section 3003, which greatly expanded the CDC’s duty to provide parole information to police and other law enforcement agencies. (See § 3003, subd. (e); Stats. 1997, ch. 680, § 2.) That statute directs the CDC to “release [] ... to local law enforcement agencies” available information, including date of parole and discharge, “regarding a paroled inmate who is released in their jurisdictions.” (§ 3003, subd. (e)(1); Stats. 1997, ch. 680, § 2.) It also specifies that the released information “shall come from the statewide parolee data base,” “shall be provided utilizing a computer-to-computer transfer in a format usable by a desktop computer system,” and “shall be continually available to local law enforcement agencies upon request.” (§ 3003, subd. (e)(2), (3); Stats. 1997, ch. 680, § 2.) The statute also provides that the CDC “shall be . . . primarily responsible for, and shall have control over, the program, resources, and staff implementing” the required computer transfer system, which is known as “the Law Enforcement Automated Data System (LEADS).” (§ 3003, [45]subd. (k); Stats. 1997, ch. 680, § 2.) In an uncodified section of the 1997 legislation that enacted these requirements, the Legislature declared its “intent ... to establish a statewide Law Enforcement Automated Data System (LEADS) to provide up-to-date information regarding parolees to local law enforcement agencies . . . .” (Stats. 1997, ch. 680, § 1.)
The Legislature passed section 3003 as the culmination of a pilot project for computer transfer of parole information from the CDC to local law enforcement agencies in San Bernardino County. (See Stats. 1997, ch. 680, § 1; Stats. 1994, ch. 904, § 1, p. 4552; Stats. 1993-1994, 1st Ex. Sess., ch. 56, §§ 1, 3, pp. 8787, 8791.) In 1995, when the Legislature considered expanding the project, it was told that the information the CDC was providing through the computer link was proving to be a valuable and useful law enforcement tool for investigating and fighting crime. (Sen. Rules Com., Off. of Sen. Floor Analyses, 3d reading analysis of Assem. Bill No. 752 (1995-1996 Reg. Sess.) as amended July 12, 1995; Sen. Com on Crim. Proc., Rep. on Assem. Bill No. 752 (1995-1996 Reg. Sess.) July 11, 1995, p. 4; Assem. Com. on Public Safety, Rep. on Assem. Bill No. 752 (1995-1996 Reg. Sess.) Apr. 4, 1995, pp. 1-2.) Two years later, when the project was made statewide, the CDC told the Legislature, among other things, that LEADS is “valuable to detectives” because it “allows [them] to search parolee information if a parolee is a prime target in criminal investigations, as they historically have been.” (Sen. Com. on Public Safety, Rep. on Assem. Bill No. 1275 (1997-1998 Reg. Sess.) July 12, 1997, p. 5.) The Legislature was also told that LEADS “enhances law enforcement’s ability to carry out the mandate of . . . earlier legislation [giving] priority to the safety of the community . . . .” (Id. at pp. 3-4.)
In passing these statutes, the Legislature has thus made clear its view that CDC employees who provide police with parole information are integral parts of the law enforcement team, and it has acted to recognize, formalize, and facilitate that relationship. These considerations reinforce our conclusion that CDC employees who prepare and maintain parole lists intended for distribution to police and other law enforcement officers—which indicate who is on parole and who may be searched without a warrant—are adjuncts to the law enforcement team and that exclusion’s deterrent effect is sufficient to justify applying the exclusionary rule.
Our conclusion is also supported by our decision in People v. Ramirez (1983) 34 Cal.3d 541 [194 Cal.Rptr. 454, 668 P.2d 761] (Ramirez). There, we found the exclusionary rule applicable where a police officer arrested and searched the defendant based on a check with “the police computer system” that showed an outstanding arrest warrant, and an inquiry after the search [46]showed that the warrant “had been recalled some six months earlier . . . (Id. at pp. 543-544.) In reaching this conclusion, we “reject[ed] the People’s argument that [the arresting officer’s] good faith reliance on information communicated to him through ‘official channels’ should validate the arrest and search.” (Id. at p. 544.) After reviewing numerous federal and state authorities, we concluded: “[L]aw enforcement officials are collectively responsible for keeping [“ ‘official’ ”] channels free of outdated, incomplete, and inaccurate warrant information. That the police now rely on elaborate computerized data processing systems to catalogue and dispatch incriminating information enhances rather than diminishes that responsibility, ffl] . . . [Thus], [t]he test, under these circumstances, is not merely the good faith of the individual officer in the field, but the good faith of law enforcement agencies of which he is a part.” (Id. at p. 552.) When “ ‘the police ... are at fault in permitting the records to remain uncorrected,’ ” they “ ‘may not rely upon incorrect or incomplete information . . . .’ [Citation.]” (Id. at pp. 545-546.) We then found that exclusion was required because the arrest that led to the search was “made on the basis of data [that] a law enforcement agency knew or should have known were in error because of inadequate or negligent record-keeping.” (Id. at p. 552.)
Ramirez fully supports application of the exclusionary rule in the case now before us. Under Ramirez, if, as the Attorney General contends, a CDC employee other than Mora—either a parole officer or a clerk—was responsible for the error here, then defendant’s parole discharge nine months before the search was within the CDC’s collective knowledge, and we cannot conclude that Mora acted in objective good faith in authorizing and conducting the warrantless search of defendant’s motel room. Similarly, because, as we have found on the record here, the relevant CDC employees were adjuncts to the law enforcement team, we cannot conclude that the police officers acted in objective good faith in assisting Mora in that warrantless search.
The Attorney General argues that “subsequent decisions of the United States Supreme Court have seriously undermined [Ramirez’s] validity.” Noting that Evans “placed great weight on the actual source of the error,” he asserts that Ramirez “is suspect” because it did not focus on this question and did not consider who was responsible for updating the records or notifying the police about recalled warrants; instead, it “simply assumed,” without “articulating] any [supporting] facts or evidence,” that the police alone were responsible for the error. “Indeed,” the Attorney General speculates, “since” Ramirez involved “a recalled bench warrant,” “there exists at least the possibility that court personnel may have played some role contributing to the inaccuracy of the police database.” The Attorney General also [47]asserts that under Evans, “police reliance upon erroneous information may establish a Fourth Amendment violation,” but “does not require strict application of the exclusionary rule,” and that Ramirez is “flaw[ed]” in that “it failed to appreciate this admittedly fine distinction.”
Unlike the Attorney General, we find nothing in Evans or in any other high court decision that undermines Ramirez’s application in the case now before us. As the Attorney General correctly suggests, Ramirez would be inconsistent with Evans if Ramirez held that exclusion is required where a judge or a court employee commits an error. However, there is no basis for the Attorney General’s speculation that Ramirez involved such an error, and our opinion there indicates otherwise. Analytically, we concluded in Ramirez that in determining an arrest’s validity, we must “examine^ the relationship between the conduct of the arresting officer and the underlying source of the probable cause determination.” (Ramirez, supra, 34 Cal.3d at p. 551.) We therefore disapproved a California Court of Appeal decision to the extent it “assert[ed] that courts must look only to the perceptions of the officer in the field, rather than trace his [or her] probable cause determination to its source in the law enforcement system . . . .” (Ramirez, supra, 34 Cal.3d at p. 550.) We also concluded that the governing test required inquiry into “the good faith of law enforcement agencies of which [the arresting officer] is a part.” (Id. at p. 552.) Given this analytical framework, we undoubtedly would have noted that a judge or court employee, rather than a law enforcement agency, made the error had the record contained any evidence to support that conclusion. Instead, we first specifically noted that the error appeared in “the police computer system.” (Id. at p. 543.) We then applied the exclusionary rule notwithstanding the good faith of the arresting officer, because he acted on data “a law enforcement agency knew or should have known were in error because of inadequate or negligent record-keeping.” (Id. at p. 552.) We also explained that the defendant’s arrest was invalid because “ ‘the police may not rely upon incorrect or incomplete information when they are at fault in permitting the records to remain uncorrected.’ [Citation.]” (Id. at pp. 545-546.) Thus, the Attorney General’s speculation about the source of the error in Ramirez is groundless.
In any event, even were the Attorney General correct that we merely assumed in Ramirez the police caused the error, the principles we applied there based on that understanding are fully consistent with Leon, Krull and Evans. As we have explained, those cases teach that for purposes of the exclusionary rule, we must distinguish between errors of law enforcement and those of judges, court employees, and legislators. In Ramirez, we drew precisely this distinction in rejecting the People’s reliance on Michigan v. DeFillippo (1979) 443 U.S. 31, 37-38 [99 S.Ct. 2627, 2632-2633, 61 [48]L.Ed.2d 343], which affirmed the validity of an arrest under an ordinance later declared unconstitutional. We found DeFillippo “distinguishable,” explaining that an arrest for violating an ordinance later found unconstitutional is “fundamentally different from” an arrest “made in reliance on the erroneous police communication that a warrant [is] outstanding.” (Ramirez, supra, 34 Cal.3d at p. 546.) In the former case, we reasoned, there is “no police misconduct to deter,” because the officer has a duty to enforce the ordinance until it is stricken. (Ibid., italics added.) Thus, we explained, “penalizing the police for legislative errors” would not “advance[]” the exclusionary rule’s goal of deterring “law enforcement misconduct. [Citation.]” (Ibid.)
The high court cases also teach that “the standard of reasonableness” for determining an officer’s good faith “is an objective one”—whether “ ‘the law enforcement officer had knowledge, or may properly be charged with knowledge, that the search was unconstitutional.’ ” (Leon, supra, 468 U.S. at p. 919 & fn. 20 [104 S.Ct. at p. 3419]; see also Krull, supra, 480 U.S. at pp. 348-349 [107 S.Ct. at pp. 1166-1167].) Leon further explains that in determining whether this standard was met, “[i]t is necessary to consider the objective reasonableness, not only of the officers who eventually executed a warrant, but also of the officers who originally obtained it or who provided information material to the probable-cause determination.” (Leon, supra, 468 U.S. at p. 923, fn. 24 [104 S.Ct. at p. 3420].) For example, the court cautioned, “[n]othing” in Leon “suggests” that an officer may use “a ‘bare bones’ affidavit” to obtain a warrant “and then rely on colleagues who are ignorant of the circumstances under which the warrant was obtained to conduct the search. [Citation.]” (Ibid.) Again, Ramirez is in accord, reasoning: “[A]n officer in the field may rely on information communicated to him by fellow officers to establish probable cause to arrest. [Citation.] However, if we impute to the arresting officer the collective knowledge of law enforcement agencies for the purpose of establishing probable cause, we must also charge him with knowledge of information exonerating a suspect formerly wanted in connection with a crime.” (Ramirez, supra, 34 Cal.3d at p. 547.)
Finally, Leon teaches that the exclusionary rule should not be applied where exclusion cannot be expected to serve “ ‘as an incentive for the law enforcement profession as a whole to conduct themselves in accord with the Fourth Amendment.’ [Citations.]” (Leon, supra, 468 U.S. at p. 919, fn. 20 [104 S.Ct. at p. 3419], italics added.) Again, Ramirez is consistent; we there concluded that if police are collectively at fault for an inaccurate record that results in an unconstitutional search, then exclusion “is consistent with the deterrence goal of the exclusionary rule.” (Ramirez, supra, 34 Cal.3d at [49]p. 547.) “[F]ocus[ing] not on the actions of the arresting officer but on the conduct of law enforcement generally,” we explained that “[suppressing the fruits of an arrest made on a recalled warrant will deter further misuse of the computerized criminal information systems and foster more diligent maintenance of accurate and current records.” (Ibid.) Thus, we disagree with the Attorney General that the high court cases undermine Ramirez.
Contrary to the Attorney General’s suggestion, the California appellate decisions he discusses are consistent with our conclusion. All of them recognize that notwithstanding the high court cases, Ramirez remains good law insofar as it holds that the good faith exception does not apply where law enforcement is collectively at fault for an inaccurate record that results in an unconstitutional search. (See In re Arron C. (1997) 59 Cal.App.4th 1365, 1369, 1372 [69 Cal.Rptr.2d 852] [citing Ramirez in reaffirming that the exclusionary rule applies “where a police officer conducts a search on the basis of faulty information from police sources,” and stating that a juvenile probation officer is an adjunct to the law enforcement team where he “becomes enmeshed in law enforcement activities” by “actively participat[ing] in a search”]; People v. Downing (1995) 33 Cal.App.4th 1641, 1654, fn. 19 [40 Cal.Rptr.2d 176] [Ramirez is “still precedential and not conflicting with Leon”]; Miranda v. Superior Court (1993) 13 Cal.App.4th 1628, 1636 [16 Cal.Rptr.2d 858] [Ramirez was not “eroded” by Leon and “remains precedential” after Leon and Krull where “error [is] generated by the police department itself’]; People v. Howard (1985) 162 Cal.App.3d 8, 20 [208 Cal.Rptr. 353] [same result required by Ramirez and by Leon, which “recognized] ... the ‘collective knowledge of law enforcement’ rationale”].)8
Our conclusion is also in accord with numerous decisions from other jurisdictions holding that Ramirez's collective knowledge principle is fully consistent with both the exclusionary rule’s deterrence objective and the high court’s decisions on the good faith exception. In State v. White (Fla. 1995) 660 So.2d 664, 667-668, the Florida Supreme Court applied the exclusionary rule where the police’s failure to update their records led to the defendant’s arrest on a warrant that had already been served. The court held that the good faith exception was “inapplicable” under these circumstances because “it was within the collective knowledge of the sheriffs office that the warrant was void” and “the arresting officers are charged with knowledge that they had no authority to arrest the defendant.” (Id. at p. 668.) The [50]court reasoned: “This type of police negligence fits squarely within the class of governmental action that the exclusionary rule was designed to deter .... Suppression of evidence seized pursuant to police computer errors will encourage law enforcement agencies to diligently maintain accurate and current computer records.” (Id. at p. 667, fn. omitted.)
In State v. Gough (1986) 35 Ohio App.3d 81 [519 N.E.2d 842, 846], an Ohio appellate court found the good faith exception inapplicable where police executed an arrest warrant issued on the basis of information that was incorrect due to “negligence, inaccuracies, or inadequacies in record-keeping procedure” by “law enforcement personnel at [a] jail.” The court explained that in Leon, the high court recognized and affirmed the collective knowledge principle Ramirez applied. (Id. at p. 845.) The court also explained that where that principle applies, “there is police conduct to deter.” (Ibid.) Finally, the court reasoned that failure to apply the exclusionary rule “would . . . encourage careless, perhaps deliberately neglectful, record keeping.” (Id. at p. 846.)
In State v. Mayorga (Tex.App. 1996) 938 S.W.2d 81, 83, a Texas Court of Appeals considered how the collective knowledge principle would apply where a police officer arrested the defendant based on incorrect radio information from a police dispatcher that there were outstanding arrest warrants for the defendant. After reviewing Evans, the court concluded that “the Leon analytical framework does not support a categorical exception to the federal exclusionary rule for mistakes made by police dispatchers. Unlike court clerks or judges, police dispatchers are in continuous radio contact with the officers on duty. They are adjuncts to the law enforcement team with a stake in the outcome of criminal prosecution. They directly provide the warrant information upon which an officer in the field depends to make an arrest; their misconduct or carelessness can be significantly affected by the threat of exclusion. Because we recognize the exclusionary rule as an important tool to help prevent impingement on Fourth Amendment rights, we decline to create another exception to the rule for errors caused by police personnel.” (Mayorga, at pp. 83-84.)
Finally, in Turnage, the Illinois Supreme Court found the good faith exception inapplicable despite the subjective good faith of the officer who arrested the defendant on a duplicative and invalid warrant. (Turnage, supra, 642 N.E.2d at p. 1241.) The court explained that “[t]he appropriate focus” under Leon “is not on the conduct of the arresting officer, but on the conduct of those who obtained the warrant and informed the arresting officer of its continued vitality.” (Ibid.) The court further explained that the invalid [51]warrant had been procured by “the State’s Attorney,” who is a “member[] of the law enforcement team . . . (Id. at p. 1240.) The court then found the good faith exception inapplicable because the government offered no evidence to show that “the State’s Attorney who sought issuance of the warrant or the sheriffs department that signaled its vitality harbored an objectively reasonable belief that the warrant was valid.” (Id. at p. 1241.) The court reasoned that suppression on these facts “would further the deterrent purposes of the exclusionary rule. Specifically, exclusion of evidence where the State’s Attorney or the sheriffs department may be charged with knowledge of the repetitive nature of a warrant will deter fishing expeditions and provide an incentive to keep accurate records.” (Ibid.) These cases all support the continued viability of Ramirez's collective knowledge analysis. (See also Hecox, supra, 619 N.E.2d at p. 342 [“[m]ost courts” hold that the good faith exception “does not apply” where incorrect information leading to invalid arrest is due to police error, in which case “there is police misconduct to deter”].)
In summary, for the reasons discussed, we agree with the Court of Appeal that the good faith exception to the exclusionary rule does not apply in this case.9
In her concurring opinion, Justice Brown argues we should reach this conclusion on the ground that once defendant stated he was not on parole and offered what he claimed was a discharge certificate, the officers could not have had an objectively reasonable belief their conduct was lawful. For several reasons, we disagree. Neither the Court of Appeal in its opinion, nor the parties in their petitions for review or briefs, addressed this theory, which, no less than our analysis, would establish a constitutional principle. By contrast, the Court of Appeal decided, and the parties have fully briefed and argued, the issue we have addressed. Moreover, Justice Brown’s theory is questionable under decisions of both this court and the United States Supreme Court. (See Hill v. California (1971) 401 U.S. 797, 799-805 [91 S.Ct. 1106, 1108-1111, 28 L.Ed.2d 484] [because false identifications are not uncommon, police had a reasonable, good faith belief that arrestee was the person they wanted, notwithstanding his claim he was someone else and his proffered identification]; Michel v. Smith (1922) 188 Cal. 199, 208 [205 P. 113] [police need not “accept the word of persons about to be arrested,” because such persons “make whatever statements . . . best serve[] [their] [52]purpose in escaping arrest”].) Justice Brown cites no case ordering suppression on similar facts or using a similar theory.10
Disposition
The judgment of the Court of Appeal is reversed and the matter is remanded for further proceedings consistent with this opinion.
George, C. J., Kennard, J., Werdegar, J., Moreno, J., and Moore, J.,* concurred.
Mullins indicated at the suppression hearing that defendant directed attention to the parole discharge certificate after they moved to the outside walkway, but the Court of Appeal’s opinion states otherwise and the Attorney General did not challenge that factual statement in a rehearing petition.
A11 further unspecified statutory references are to the Penal Code. The information also charged possession of a controlled substance (Health & Saf. Code, § 11377, subd. (a)). The trial court later dismissed that charge on the prosecution’s motion.
We have previously noted that the term “good faith exception” may be somewhat of a misnomer, because the exception focuses on the objective reasonableness of an officer’s conduct. (People v. Machupa (1994) 7 Cal.4th 614, 618, fn. 1 [29 Cal.Rptr.2d 775, 872 P.2d 114]; see Krull, supra, 480 U.S. at p. 355 [107 S.Ct. at p. 1170].) Nevertheless, we use the term because of its common acceptance by commentators and courts, including the high court itself. (See Krull, supra, 480 U.S. at p. 353 [107 S.Ct. at pp. 1168-1169].)
The record is unclear as to whether Mora was defendant’s parole agent. Defendant testified at the suppression hearing that when Mora and the police entered his room, they asserted that Mora was his “agent of record.” He also testified that in response, he denied that Mora had been his parole officer.
Section 3067 was enacted in August 1996. However, parole agreements have included equivalent search conditions since well before the search of the defendant’s room in March 1996. (See, e.g., People v. Burgener (1986) 41 Cal.3d 505, 528, fh. 10 [224 Cal.Rptr. 112, 714 P.2d 1251] [similar search condition in 1980 parole agreement].)
As we later more fully explain, since 1997, the Legislature has required the CDC to provide local law enforcement agencies with direct and continuous access to specified CDC parole information, including date of discharge, through computer link. (See § 3003, subd. (e).)
We also note that in Evans, the record contained evidence that the type of error involved there “occurred ‘on[c]e every three or four years,’ ” and that “once the court clerks discovered the error, they immediately corrected it [citation], and then proceeded to search their files to make sure that no similar mistakes had occurred [citation].” (Evans, supra, 514 U.S. at p. 15 [115 S.Ct. at p. 1194].) The record here contains no evidence regarding these matters. (Cf. id. at p. 17 [115 S.Ct. at p. 1194] (cone. opn. of O’Connor, J.) [“it would not be reasonable for the police to rely . . . 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 probable cause for any such arrest has ceased to exist (if it ever existed)” (italics omitted)].)
We express no opinion regarding the specific application of Ramirez or the specific holding in any of these cases, because it is unnecessary to do so here. We similarly express no opinion about the conclusion reached in any of the decisions we discuss that involved facts different from the facts before us. We consider those decisions only as they are relevant to our analytical approach here.
Given our conclusion, we need not, and do not, consider defendant’s argument that Leon’s good faith exception applies to an invalid warrantless search only if the error that led to the search is attributable solely to a court employee or the Legislature.
Nor does Justice Brown explain how our analysis is “ ‘oblivious or hostile to . . . common sense’ ” (cone. opn. of Brown, J., post, at p. 55), in requiring suppression where police and parole officers conducted a warrantless search of someone who, nine months before the search, was discharged from parole and “regain[ed] full Fourth Amendment protection.” (Cone. opn. of Brown, J., post, at p. 54.)
Associate Justice, Court of Appeal, Fourth Appellate District, Division Three, assigned by the Chief Justice pursuant to article VI, section 6, of the California Constitution.