Opinion of the Court by Kruger, J.

Responding to a report of suspicious activity in the area, a police officer unlawfully detained a bystander who had no apparent connection to the report. The officer ran a records search and learned that the bystander, Duvanh Anthony McWilliams, was on parole and subject to warrantless, suspicionless parole searches. The officer proceeded to search McWilliams and his vehicle, where the officer found an unloaded gun, ammunition, drugs, and drug paraphernalia. As a general rule, evidence seized as a result of an unlawful search or seizure is inadmissible against the defendant in a subsequent prosecution. But the law permits use of the evidence when the causal connection “between the lawless conduct of the police and the discovery of the challenged evidence has ‘become so attenuated as to dissipate the taint.’ ” (Wong Sun v. United States (1963) 371 U.S. 471, 487.) Here, the Court of Appeal held that the officer’s discovery of McWilliams’s parole search condition sufficiently attenuated the connection between the unlawful detention and the contraband found in McWilliams’s vehicle. The Court of Appeal relied on cases allowing the admission of evidence seized incident to arrest on a valid warrant, where the warrant was discovered during an unlawful investigatory stop. (Utah v. Strieff (2016) 579 U.S. 232 (Strieff); People v. Brendlin (2008) 45 Cal.4th 262 (Brendlin).) We now reverse. Unlike an arrest on an outstanding warrant, a parole search is not a ministerial act dictated by judicial mandate (Strieff, supra, 579 U.S. at p. 240), but a matter of discretion. We conclude the officer’s discretionary decision to conduct the parole search did not sufficiently attenuate the connection between the officer’s initial unlawful decision to detain McWilliams and the discovery of contraband. The evidence therefore was not admissible against him. I. Early one evening in January 2017, Officer Matthew Croucher of the San Jose Police Department responded to a report of a possible vehicle burglary in a business parking lot. When he arrived on the scene, a security guard told him she had seen two “suspicious individuals on bikes” shining flashlights into parked cars. Officer Croucher drove through the parking lot but did not see anything noteworthy. He then drove through an adjacent parking lot. There he found about four or five parked cars, one of which was occupied. The occupant of the car was McWilliams, who was fully reclined in the passenger seat. McWilliams did not appear to be sleeping, just “hanging out.” Officer Croucher waited for backup, then approached the vehicle and instructed McWilliams to exit. At the suppression hearing, Croucher testified this was for safety reasons; it was his usual practice “with most car stops that [he] do[es], or most suspicious vehicles that [he] come[s] across.” After McWilliams exited the vehicle, Croucher asked for his identification and permitted McWilliams to retrieve the identification from the vehicle. Croucher then performed a records check and learned that McWilliams was “on active and searchable [California Department of Corrections] parole.” Croucher proceeded to search both McWilliams and the vehicle, from which he seized a firearm, drugs, and drug paraphernalia. McWilliams was charged with multiple drug and weapons offenses. He filed a motion to suppress the evidence found in his vehicle. (See Pen. Code, § 1538.5.) He argued, among other things, that the evidence should be excluded as the fruits of a detention conducted in violation of the Fourth Amendment to the United States Constitution. The trial court denied the motion, concluding the detention was lawful. The court reasoned that Officer Croucher had reasonable suspicion to detain McWilliams based on the security guard’s 911 call reporting suspicious activity in the area. McWilliams pleaded guilty to three counts of the criminal information and received a negotiated sentence of seven years in state prison. McWilliams appealed the denial of the suppression motion. A divided Court of Appeal affirmed. (People v. McWilliams (Mar. 8, 2021, H045525) [nonpub. opn.].) Unlike the trial court, the Court of Appeal concluded the officer lacked reasonable suspicion to detain McWilliams, since McWilliams neither matched the security guard’s description of the individuals involved in the suspicious activity nor was engaged in any conduct “suggestive of criminal activity.” But the appellate court nonetheless upheld the trial court’s denial of the suppression motion, reasoning that the officer’s discovery of the parole search condition sufficiently attenuated the connection between the officer’s unlawful detention of McWilliams and the evidence seized during the search. The court likened the discovery of the parole search condition to the discovery of the arrest warrants in Strieff, supra, 579 U.S. 232 and Brendlin, supra, 45 Cal.4th 262, which the courts held to be sufficiently attenuating. Like the arrest warrants in those cases, the parole search condition here “predated the detention, was not subject to interpretation, and supplied entirely independent legal authorization for the search.” The court further concluded the evidence was admissible because the unlawful detention was not “pretextual, in bad faith, or part of recurrent police misconduct.” Justice Danner concurred in part and dissented in part. She agreed with the majority that the initial detention was unlawful, but disagreed that the evidence seized from McWilliams’s vehicle was nonetheless admissible. In her view, the discovery of the parole search condition was not an intervening circumstance that dissipated the taint of the unlawful detention, since Officer Croucher had no obligation to perform a search upon discovering the parole search condition but instead exercised his discretion to do so. Justice Danner also disagreed with the majority’s conclusion that “the officer’s actions here do not raise a broader issue of police misconduct,” given Officer Croucher’s thin rationale for detaining McWilliams; his testimony that it is his routine practice to order people out of suspicious vehicles; and the “growing recognition that seemingly small constitutional violations can add up to problems of significant national dimensions.” (People v. McWilliams, supra, H045525 (conc. & dis. opn. of Danner, J.), citing, inter alia, Strieff, supra, 579 U.S. at p. 254 (dis. opn. of Sotomayor, J.) [“it is no secret that people of color are disproportionate victims” of unlawful, suspicionless stops].) The disagreement between the justices of the Court of Appeal in this case mirrors a similar division among Courts of Appeal that have considered the attenuating effect of a probation (as opposed to parole) search condition. (Compare People v. Durant (2012) 205 Cal.App.4th 57, 66 [concluding that the “illegality in the initial traffic detention was attenuated by appellant’s probation search condition”] with People v. Bates (2013) 222 Cal.App.4th 60, 71 (Bates) [declining to adopt Durant’s reasoning and reaching the opposite conclusion on the facts].) We granted review to consider the proper application of the attenuation doctrine to the officer’s discovery of the parole search condition in this case.1

II.

The Fourth Amendment to the United States Constitution protects the “right of the people to be secure in their persons, houses, papers, and effects, against unreasonable searches and seizures.”2 The right is primarily enforced through the

1 Although all parties assume a similar analysis would apply to both parole and probation searches, our analysis here focuses on parole searches like the search at issue in this case. In this court, the Attorney General has taken the view that the Court of Appeal erred in concluding the discovery of McWilliams’s parole status sufficiently attenuated the connection between the unlawful detention and the subsequent search. The Santa Clara District Attorney’s Office, which handled this matter in the trial court, has filed an amicus curiae brief supporting the judgment of the Court of Appeal. 2 The California Constitution similarly protects the “right of the people to be secure in their persons, houses, papers, and effects against unreasonable seizures and searches.” (Cal. Const., art. I, § 13.) But under the so-called truth-in-evidence provision of the state Constitution, “ ‘issues relating to the suppression of evidence derived from governmental searches and seizures are reviewed under federal constitutional exclusionary rule, “a deterrent sanction that bars the prosecution from introducing evidence obtained by way of a Fourth Amendment violation.” (Davis v. United States (2011) 564 U.S. 229, 231–232 (Davis); see Strieff, supra, 579 U.S. at p. 237.) Where it applies, the exclusionary rule forbids admission of both the “ ‘primary evidence obtained as a direct result of an illegal search or seizure’ ” and “ ‘evidence later discovered and found to be derivative of an illegality’ ” — familiarly known as the “ ‘ “fruit of the poisonous tree.” ’ ” (Strieff, at p. 237, quoting Segura v. United States (1984) 468

U.S. 796, 804.)

The exclusionary rule does not, however, apply in every case involving a Fourth Amendment violation. Balancing the benefits of the exclusionary remedy against its costs, the United States Supreme Court has fashioned various exceptions to the exclusionary rule, including the so-called attenuation doctrine. (Strieff, supra, 579 U.S. at pp. 237–238; see Davis, supra, 564 U.S. at p. 237.) The attenuation doctrine holds that, notwithstanding the exclusionary rule, “[e]vidence is admissible when the connection between unconstitutional police conduct and the evidence is remote or has been interrupted by some intervening circumstance, so that ‘the interest protected by the constitutional guarantee that has been violated would not be served by suppression of the evidence obtained.’ ” (Strieff, at p. 238, quoting Hudson v. Michigan (2006) 547 U.S. 586, 593 (Hudson).) In conducting the attenuation inquiry, courts are

standards.’ ” (People v. Macabeo (2016) 1 Cal.5th 1206, 1212, quoting People v. Troyer (2011) 51 Cal.4th 599, 605; see Cal. Const., art. I, § 28, subd. (f)(2).) We accordingly focus on federal constitutional standards in our analysis in this case. guided by three factors first set out in Brown v. Illinois (1975) 422 U.S. 590, 603–604 (Brown): (1) the “temporal proximity” between the unlawful conduct and the discovery of evidence; (2) the “presence of intervening circumstances”; and (3) the “purpose and flagrancy of the official misconduct.” (See Strieff, at p. 239.) Once the defendant establishes a Fourth Amendment violation, the prosecution bears the burden of establishing admissibility under this exception to the exclusionary rule. (Brown, at p. 604.) In Brendlin, supra, 45 Cal.4th 262, this court considered how the attenuation doctrine applies when an officer unlawfully seizes an individual and then discovers that the individual has an outstanding arrest warrant. In that case, a sheriff’s deputy had unlawfully stopped a vehicle to investigate expired registration tabs, on a hunch that the temporary operating permit in the window might belong to a different vehicle. The deputy asked the occupants to identify themselves, ran a records check, and discovered that the passenger, Brendlin, had an outstanding no-bail arrest warrant. The officer arrested Brendlin and searched him incident to the arrest, finding drugs and drug paraphernalia. (Id. at pp. 265–266.) We analyzed the three Brown factors to determine whether the incriminating evidence was admissible notwithstanding the unlawful stop. We acknowledged that the first factor, temporal proximity, weighed against attenuation because “only a few minutes elapsed” between the unlawful stop and the search. (Brendlin, supra, 45 Cal.4th at p. 270.) But we placed little weight on this factor, noting that the timeline is typical of cases in which an investigatory stop leads to the discovery of an outstanding arrest warrant, and in such circumstances this factor is “ ‘outweighed by the others.’ ” (Ibid.) As for the second factor, we concluded that an arrest on a valid warrant and subsequent search incident to arrest is an intervening circumstance that “tends to dissipate the taint caused by an illegal traffic stop.” (Id. at p. 271.) We explained that “[a] warrant is not reasonably subject to interpretation or abuse,” and that “the no-bail warrant [t]here supplied legal authorization to arrest defendant that was completely independent of the circumstances that led the officer to initiate the traffic stop.” (Ibid.) Finally, as to the third factor, we found the deputy’s misconduct was neither purposeful nor flagrant. The deputy testified that, in his experience, temporary stickers on a vehicle with expired registration “sometimes belonged to a different vehicle or had been falsified.” (Ibid.) Although the deputy had insufficient grounds to make the stop, “the insufficiency was not so obvious as to make one question [his] good faith in pursuing an investigation of what he believed to be a suspicious registration, nor [did] the record show that he had a design and purpose to effect the stop ‘in the hope that something [else] might turn up.’ ” (Ibid., quoting Brown, supra, 422 U.S. at p. 605.) Nearly a decade later, the United States Supreme Court granted review in Strieff to address the same issue as Brendlin. (Strieff, supra, 579 U.S. at p. 237.) In Strieff, a police detective investigated an anonymous tip that a certain residence was the site of narcotics activity. After observing visitors enter the house and leave after a few minutes, the detective came to believe that the house occupants were dealing drugs. The detective, Officer Fackrell, stopped one visitor, Strieff, as he exited the house. Strieff turned over his identification to Fackrell upon request and Fackrell learned that Strieff had an outstanding arrest warrant for a traffic violation. Fackrell arrested Strieff, searched him incident to the arrest, and found methamphetamine and drug paraphernalia. (Id. at pp. 235– 236.) Assuming with the parties that the investigatory stop was unlawful, the high court determined that the discovery of the arrest warrant sufficiently attenuated the connection between the stop and the ensuing discovery of drug-related evidence. (Strieff, supra, 579 U.S. at p. 239.) As to the first Brown factor, the court concluded that the close temporal proximity between the stop and the search favored suppression. (Ibid.) But the high court found the second factor, the presence of intervening circumstances, “strongly favor[ed] the State.” (Id. at p. 240.) The court explained: “In this case, the warrant was valid, it predated Officer Fackrell’s investigation, and it was entirely unconnected with the stop. And once Officer Fackrell discovered the warrant, he had an obligation to arrest Strieff. ‘A warrant is a judicial mandate to an officer to conduct a search or make an arrest, and the officer has a sworn duty to carry out its provisions.’ [Citation.] Officer Fackrell’s arrest of Strieff thus was a ministerial act that was independently compelled by the pre-existing warrant. And once Officer Fackrell was authorized to arrest Strieff, it was undisputedly lawful to search Strieff as an incident of his arrest to protect Officer Fackrell’s safety.” (Id. at pp. 240–241, citing Arizona v. Gant (2009) 556 U.S. 332, 339.) Finally, the court concluded that the third Brown factor also “strongly favor[ed] the State” because the officer “was at most negligent” in conducting the unlawful investigatory stop. (Strieff, at p. 241.) The court explained that there was “no indication that this unlawful stop was part of any systemic or recurrent police misconduct”; rather, “all the evidence suggests that the stop was an isolated instance of negligence that occurred in connection with a bona fide investigation of a suspected drug house.” (Id. at p. 242.) The court accordingly held the evidence seized in the search incident to Strieff’s arrest was admissible notwithstanding the initial unlawful stop. (Id. at p. 243.)

III.

Here, as in Brendlin and Strieff, an officer conducted a concededly unlawful seizure: The Court of Appeal in this case concluded, and all parties now agree, that Officer Croucher violated the Fourth Amendment when he ordered McWilliams out of his vehicle with no basis to suspect McWilliams of involvement in any criminal activity. And here, as in Brendlin and Strieff, the officer conducted a records check after that unlawful detention. But unlike in Brendlin and Strieff, the records check did not turn up an outstanding warrant for arrest. Rather, it revealed that McWilliams was on parole and subject to a parole condition authorizing warrantless, suspicionless searches of his person and his vehicle. (See Pen. Code, § 3067, subd. (b)(3).) The question now before us is whether the evidence discovered in the ensuing search should have been suppressed under the exclusionary rule, or whether the evidence was properly admitted because the discovery of McWilliams’s parole search condition sufficiently attenuated the connection between the unlawful detention and the search. To answer the question, we consider how the three Brown factors apply in this distinct context. A. The first Brown factor, temporal proximity between the unlawful detention and the search, requires no extended discussion. As Strieff explains, the high court has “declined to find that this factor favors attenuation unless ‘substantial time’ elapses between an unlawful act and when the evidence is obtained.” (Strieff, supra, 579 U.S. at p. 239.) All agree that here, because no “ ‘substantial time’ ” separated Officer Croucher’s initial decision to detain McWilliams and his subsequent decision to search him and his vehicle, this factor weighs against a finding of attenuation. (Ibid.) The central subject of disagreement in this case concerns whether, and to what extent, the discovery of a parole search condition disrupts the causal connection between the unlawful stop and the discovery of evidence. This is the concern of the second Brown factor, the presence of intervening circumstances. (See Strieff, supra, 579 U.S. at p. 238.) McWilliams argues the discovery of a parole search condition, unlike the discovery of an arrest warrant, can never qualify as such an intervening circumstance. He argues it is therefore unnecessary to engage in the Brown attenuation analysis at all — but if we do, we should assign no attenuating significance to the discovery of the parole search condition under the second Brown factor. In the alternative, McWilliams argues, and the Attorney General agrees, that discovery of a parole search condition does trigger the Brown attenuation analysis, but on its own does very little to attenuate the taint of illegality under the second prong of Brown. The District Attorney of Santa Clara County, participating as an amicus curiae in support of the judgment of the Court of Appeal, disagrees with McWilliams on both counts; in the District Attorney’s view, discovery of a parole search condition is at least as attenuating a circumstance as the discovery of an arrest warrant, and strongly supports finding attenuation in this case and others like it. At the outset, we decline McWilliams’s invitation to hold that discovery of a parole search condition can never qualify as an intervening circumstance that tends to attenuate the link between an unlawful stop and an ensuing search. Many of the arguments McWilliams makes in support of this proposed categorical rule are squarely foreclosed by precedent. McWilliams argues, for instance, that the discovery of the parole search condition is itself the “fruit” of the illegal detention, and for that reason cannot attenuate the primary taint. We rejected much the same argument in Brendlin. We explained that, under long-standing high court precedent, exclusion “ ‘may not be premised on the mere fact that a constitutional violation was a “but-for” cause of obtaining evidence.’ ” (Brendlin, supra, 45 Cal.4th at p. 268, quoting Hudson, supra, 547 U.S. at p. 592.) Discovery of an arrest warrant, we explained, is not “ ‘ “ ‘fruit of the poisonous tree’ simply because it would not have come to light but for the illegal actions of the police.” ’ ” (Brendlin, at p. 268; accord, Strieff, supra, 579 U.S. at p. 235.) The same is true when a records check reveals a parole search condition rather than a warrant. McWilliams also attempts to analogize the discovery of a parole search condition following an unlawful detention to the discovery of contraband that comes into plain view because of an unlawful search or seizure. In the latter situation, where a police officer’s illegal conduct “caused the evidence to be placed in plain view” — for instance, where an unlawful order to exit a car leads to the observation of contraband on the driver’s person — the police may not rely on the plain view doctrine to justify a warrantless seizure. (U.S. v. Davis (10th Cir. 1996) 94 F.3d 1465, 1470.) But as McWilliams’s own cases explain, the rationale for this rule is specific to the plain view doctrine, whose “ ‘first and most fundamental prerequisite to reliance upon plain view as a basis for a warrantless seizure . . . is that “the initial intrusion which brings the police within plain view of such an article” is itself lawful.’ ” (Ibid. [“ ‘The [plain view] doctrine serves to supplement the prior justification — whether it be a warrant for another object, hot pursuit, search incident to lawful arrest, or some other legitimate reason for being present unconnected with a search directed against the accused.’ ”].) This is not a plain view case, and plain view cases do not help McWilliams here. McWilliams’s invocation of the conditions on valid parole searches, as set forth in People v. Sanders (2003) 31 Cal.4th 318, 333 (Sanders), is likewise unavailing. Sanders holds that, to conduct a valid parole search, an officer must be aware of the search condition at the time; it is not enough for the officer to learn of the condition after the search is done. McWilliams suggests that by parity of reasoning, the discovery of a parole search condition should not operate to validate a prior unauthorized detention. But for essentially the same reasons we explained in Brendlin, the rule of Sanders is inapposite here. (See Brendlin, supra, 45 Cal.4th at p. 273.) No one here contends the post-detention discovery of a parole search condition should retroactively validate the initial decision to detain McWilliams. The illegality of the detention is undisputed. The question is whether the mid-detention discovery of a parole search condition is an intervening circumstance that justifies making an exception to the exclusionary rule for the evidence turned up in that search. That is a question Sanders does not answer. Ultimately, however, we need not and do not decide here whether or under what circumstances discovery of a parole search condition could ever sufficiently dissipate the taint from an initial unlawful detention. It suffices for us to conclude that the discovery of the parole search condition had no considerable attenuating effect under the circumstances of this case. In reaching a contrary conclusion, the Court of Appeal emphasized the evident similarities between a valid arrest warrant and a parole search condition: Like an arrest warrant, a parole search condition necessarily predates the detention and is authorized by state law independent of the detention. (See Pen. Code, § 3067.) The court’s reliance on these features was understandable because these are the very same features we mentioned in Brendlin when we concluded that the discovery of a valid arrest warrant was an intervening circumstance that attenuated the causal chain between the unlawful stop and the incriminating evidence. (Brendlin, supra, 45 Cal.4th at p. 271.) But despite their similarities, the arrest warrants at issue in Brendlin and Strieff differ from parole search conditions in a critical respect: As judicial mandates to take a suspect into custody, the warrants not only authorized, but compelled, further action by the officer. Although we did not mention this point explicitly in Brendlin, the court emphasized it in Strieff. After laying out a set of general observations about the lack of connection between the unlawful stop and the existence of the arrest warrant, the court went on to explain that “once Officer Fackrell discovered the warrant, he had an obligation to arrest Strieff. ‘A warrant is a judicial mandate to an officer to conduct a search or make an arrest, and the officer has a sworn duty to carry out its provisions.’ [Citation.] Officer Fackrell’s arrest of Strieff thus was a ministerial act that was independently compelled by the pre-existing warrant.” (Strieff, supra, 579 U.S. at p. 240.)3 By contrast to the arrest warrant in Strieff, a parole search condition merely authorizes a suspicionless search of the parolee for purposes of monitoring the parolee’s rehabilitation and compliance with the terms of parole. It is not a judicial mandate, nor does it compel further action of any sort. Whether to take further action is largely within law enforcement’s discretion; the search of a parolee is generally permissible, so long as the search is not arbitrary, capricious, or harassing. (People v. Reyes (1998) 19 Cal.4th 743, 754 (Reyes).) We agree with other courts that have held that the absence of compulsion to continue the interaction after an initial unlawful detention makes a difference in the attenuation analysis. (See State v. Christian (Kan. 2019) 445 P.3d 183, 190 [distinguishing Strieff and finding insufficient attenuation where officer could choose whether to arrest defendant for having no proof of insurance].) As a general rule, the law recognizes that an intervening circumstance can break the chain of causation when that circumstance involves “an act of a third person or other force which by its intervention prevents the actor from being liable for harm” the actor played a substantial role in bringing about. (Rest.2d Torts, § 440; see Soule v. General Motors Corp. (1994) 8 Cal.4th 548, 573, fn. 9.) The corollary is that, as a general matter, defendant’s “own

3 The District Attorney suggests that the arrest warrant in Strieff — which was for a traffic violation — would not have required custodial arrest under California law. Be that as it may, the high court’s analysis presumed that arrest was compelled under the Utah arrest warrant at issue in the case. conduct cannot be an intervening cause sufficient to defeat a finding of causation. ‘A superseding cause is something culpable that intervenes . . . , some action of a third party that makes the plaintiff’s injury an unforeseeable consequence of the defendant’s negligence.’ ” (Whitlock v. Brueggemann (7th Cir. 2012) 682 F.3d 567, 584; see also, e.g., Von der Heide v. Com., Dept. of Transp. (Pa. 1998) 718 A.2d 286, 289.) The rule of Strieff comports with this ordinary understanding of causation principles: As Strieff conceived of it, the discovery of the arrest warrant represented a form of compulsion by a third party magistrate that left the officer with no effective choice but to carry out an arrest. (Strieff, supra, 579 U.S. at p. 240.) This case, however, involves no such third party compulsion; the discovery of McWilliams’s parole status merely gave the detaining officer the discretion to conduct a warrantless, suspicionless search. At least absent other factors, the detaining officer’s discovery of a parole search condition, and subsequent decision to conduct a parole search, does comparatively little to disrupt the causal chain. This conclusion about the relative attenuating force of parole search conditions fits with the core concern underlying the Brown attenuation analysis. As Brown made clear, the factors must be applied in such a way as to ensure the officer is not “unduly exploit[ing]” an unlawful search or seizure to produce incriminating evidence. (Brown, supra, 422 U.S. at p. 603.) The existence of a valid arrest warrant significantly alleviates such exploitation concerns because the warrant represents a “ ‘judicial mandate’ ” to take further action (Strieff, supra, 579 U.S. at p. 240), that “is not reasonably subject to interpretation or abuse” (Brendlin, supra, 45 Cal.4th at p. 271, citing Hudson, supra, 547 U.S. at p. 595 and U.S. v. Green (7th Cir. 1997) 111 F.3d 515, 522). Because a parole search condition, by contrast, entails no such compulsion but instead invokes law enforcement discretion, it raises greater concerns about the course of events connecting an unlawful stop to an ensuing search. There is a danger that an officer who has unlawfully stopped a bystander without reasonable suspicion will regard the discovery of a parole search condition as a license to continue pursuing a baseless hunch, rather than fairly considering whether a search is appropriate to assess the individual’s rehabilitation and monitor “his transition from inmate to free citizen.” (Reyes, supra, 19 Cal.4th at p. 752.) In other words, in the hands of the very same officer who conducted an illegal stop, there is a risk that the discretion to conduct a parole search will lead to the exploitation of that illegal conduct, rather than severing the causal connection between the stop and the search. McWilliams and the Attorney General agree that because the choice to conduct a parole search was within Officer Croucher’s discretion, rather than a matter of compulsion, the discovery of the parole search condition does little to attenuate the connection between Officer Croucher’s unlawful detention of McWilliams and the evidence at issue. But the District Attorney, acting as amicus curiae in support of the Court of Appeal’s judgment, argues that the absence of compulsion to conduct a parole search ought to be irrelevant to the analysis. We are unpersuaded. In support of the argument, the District Attorney observes that there was also discretion at play in Brendlin and Strieff: Although the officers in those cases may have been under judicial compulsion to take the suspects into custody after they discovered outstanding arrest warrants, the District Attorney points out, the officers were not compelled to search the suspects incident to arrest and instead chose to do so as a matter of discretion. True enough, but the argument mistakes the point. We do not hold that any element of discretion necessarily defeats a claim of attenuation; we simply conclude, rather, that the absence of compulsion naturally weakens the claim. In Strieff, the court explained that once Officer Fackrell discovered an outstanding warrant for Strieff’s arrest, the discovery “broke the causal chain between the unconstitutional stop and the discovery of evidence by compelling Officer Fackrell to arrest Strieff.” (Strieff, supra, 579 U.S. at p. 242.) Once that causal chain was broken by the compulsion to arrest, the ultimate decision to conduct a search incident to arrest was attributable to the legally required arrest itself — “to protect Officer Fackrell’s safety” as he carried out the arrest — and not the initial unlawful decision to stop Strieff. (Id. at p. 241, citing Arizona v. Gant, supra, 556 U.S. at p. 339.) Here, there was no comparable compulsion for Officer Croucher to take any particular action regarding McWilliams, and thus no comparable force breaking the causal chain between Croucher’s unconstitutional detention of McWilliams and his discretionary decision to search. The District Attorney next argues that an officer’s discretion to conduct a search is constrained in various ways that limit the opportunities for “interpretation or abuse.” (Brendlin, supra, 45 Cal.4th at p. 271.) The authorized scope of a parole search is ordinarily clear; the officer must be aware of the search condition before conducting the search (Sanders, supra, 31 Cal.4th at p. 333); and the officer may not conduct a search that is arbitrary, capricious, or harassing (Reyes, supra, 19 Cal.4th at p. 752). These constraints mean, for instance, that an officer may not conduct a particular parole search in a harassing manner, or for reasons unrelated to any legitimate penological purpose, such as personal animosity toward the parolee. (Ibid.; see In re Anthony S. (1992) 4 Cal.App.4th 1000, 1004.) These limitations on parole searches are certainly important, but they do not answer the Fourth Amendment question at hand. The issue before us does not concern the validity of the parole search, standing alone, but instead concerns whether a court must exclude evidence in response to an immediately preceding, concededly unconstitutional detention. And as already explained, despite the limitations governing parole searches, the law leaves an officer substantial discretion whether to conduct such searches, which weakens the case for finding a break in the causal chain connecting that unlawful detention and the discovery of evidence. Taking a different tack, the District Attorney argues that the discovery of a parole search condition must have at least as much attenuating force as the discovery of an arrest warrant because a parolee has a “significantly diminished expectation of privacy in comparison to a mere arrestee.” (See Samson v. California (2006) 547 U.S. 843, 852 [parolees “have severely diminished expectations of privacy by virtue of their status alone”].) This argument, too, confuses the issue. To be sure, parolees’ diminished expectation of privacy is the reason the Fourth Amendment generally permits suspicionless parole searches. (Id. at p. 847.) But the question before us does not involve the constitutionality of parole searches; it is whether the discovery of the parole search condition in this case sufficiently attenuated the taint stemming from an initial unconstitutional detention. To answer this question, we evaluate “the causal link between the government’s unlawful act and the discovery of evidence.” (Strieff, supra, 579 U.S. at p. 238.) For purposes of this causation analysis, McWilliams’s expectations of privacy as a parolee have no particular relevance. Finally, the District Attorney points to an out-of-state case as an example of the attenuating force of a parole search condition. (State v. Fenton (Idaho Ct.App. 2017) 413 P.3d 419.) There, the police officer who made an unlawful stop discovered the defendant was on parole and called the defendant’s probation officer, who then decided to conduct a search after he arrived on the scene. (Id. at pp. 421–423.) This case is unlike Fenton, however, in that the officer who initially stopped McWilliams and the officer who decided to conduct the parole search were one and the same. To decide this case, and most cases like it, we need not decide whether the third party probation officer’s decision to search in Fenton sufficiently attenuated the discovery of incriminating evidence from the initial unlawful stop. We likewise need not consider whether, as the Attorney General argues, the discovery of a parole search condition might sufficiently dissipate the taint of an initial unlawful stop when there is a substantial period of time between the discovery and a parole search. It suffices to conclude that in this case — where the same officer who conducted the illegal detention also decided, minutes later, to conduct a parole search that yielded incriminating evidence — the discovery of the parole search condition did relatively little to break the causal connection between the two events. B. We turn, then, to the third and final Brown factor, the flagrancy and purposefulness of police misconduct. While the first two factors identify forces — time and intervening circumstances — that may tend to attenuate the causal connection between the misconduct and the discovery of evidence, the focus of the third factor is different: It “ ‘is directly tied to the purpose of the exclusionary rule — deterring police misconduct.’ ” (Brendlin, supra, 45 Cal.4th at p. 271.) Police misconduct, the high court has said, is “most in need of deterrence . . . when it is purposeful or flagrant.” (Strieff, supra, 579 U.S. at p. 241; accord, Brendlin, at p. 271.) The greater the degree of purposefulness or flagrancy associated with the police misconduct, the greater the justification required to admit evidence obtained through the misconduct.4 To the extent McWilliams suggests that Officer Croucher’s decision to detain him without reasonable suspicion itself establishes purposeful or flagrant misconduct under the third Brown factor, the law is to the contrary. Every attenuation case involves an improper search or seizure, but not every impropriety rises to the level of purposeful or flagrant illegality. (Strieff, supra, 579 U.S. at pp. 242–243; see Brendlin, supra, 45 Cal.4th at p. 271.) But as McWilliams emphasizes, here Officer Croucher’s basis to suspect McWilliams of violating the law was

4 Where neither of the first two Brown factors establishes sufficient attenuation, courts have held that evidence may be subject to suppression even absent flagrant or purposeful conduct. (See U.S. v. Garcia (9th Cir. 2020) 974 F.3d 1071, 1082 [“[E]ven accepting the district court’s finding that the officers acted in good faith, this fact alone is not enough to justify admission of the evidence.”]; U.S. v. Bocharnikov (9th Cir. 2020) 966 F.3d 1000, 1005 [same]; U.S. v. Perez-Esparza (9th Cir. 1979) 609 F.2d 1284, 1291 [“[T]he last factor is insufficient to overcome the lack of attenuation dictated by the first two factors.”].) Here, however, we find purposeful conduct and so have no occasion to consider how to weigh the third Brown factor under different circumstances. not merely insufficient — it was essentially nonexistent. The security guard in a business parking lot had reported suspicious activity involving two individuals riding bicycles and shining flashlights into cars. Officer Croucher found McWilliams alone and reclined inside a car, with no bicycle or flashlight in sight. Rather than approaching McWilliams to ask him for information, Officer Croucher instead ordered McWilliams out of the car for asserted safety concerns, thereby effecting a seizure of his person. Then, despite these asserted safety concerns, Officer Croucher allowed McWilliams to return to his car to retrieve his identification and used that identification to run a records check. Officer Croucher may not have acted in bad faith when he detained McWilliams. But a finding of purposefulness does not require a showing of bad faith; the law instructs that officers act purposefully for Brown purposes when they conduct “a suspicionless fishing expedition ‘in the hope that something [will] turn up.’ ” (Strieff, supra, 579 U.S. at p. 242, quoting Taylor v. Alabama (1982) 457 U.S. 687, 691; accord, e.g., Brendlin, supra, 45 Cal.4th at pp. 271–272; see Bates, supra, 222 Cal.App.4th at p. 71 [finding purposefulness where officer stopped a car “without any observation of possible wrongdoing,” based on “a hunch that [a suspect] might be in the vehicle”].) One post-Strieff appellate court, for example, has found purposefulness and flagrancy where an officer stopped the defendant in an attempt to identify a person connected to a shooting that occurred two days earlier, “on the basis of a photograph that provided little meaningful identifying information to the police besides the race” of the person, and despite the absence of any indication that the person in the photograph had committed any crime in the first place. (U.S. v. Walker (2d Cir. 2020) 965 F.3d 180, 183; see id. at pp. 184, 188.) Then, even after it became evident that the defendant was not the person depicted in the photograph, the officer conducted a records check that revealed a valid arrest warrant; conducted a search incident to arrest; and found incriminating evidence. Notwithstanding Strieff, the court concluded the evidence should have been suppressed based on consideration of the third Brown factor. (Walker, at p. 190.) The problem with the stop, the Second Circuit explained, was “not simply the lack of reasonable suspicion,” but “the extreme lack of reasonable suspicion.” (Id. at p. 189.) The court added that reliance on the photograph to make the stop necessarily “involved impermissible and manifest stereotyping, which cannot be characterized as merely negligent conduct.” (Id. at p. 190.) And even if the initial justification for the stop “were not woefully anemic,” the officer’s decision to run a records check, even after he verified the defendant was not the subject of the photograph, amounted to “a mere fishing expedition.” (Id. at pp. 189–190.) The Attorney General and District Attorney argue Officer Croucher’s conduct in this case is comparable to the conduct of the officer in Strieff, which the court viewed as neither purposeful nor flagrant but “at most negligent.” (Strieff, supra, 579 U.S. at p. 241.) We are unpersuaded by the comparison. In Strieff, the officer had observed Strieff exiting what the officer reasonably believed to be a drug house; his primary error was in failing to observe how long Strieff remained at the location, which meant he “lacked a sufficient basis to conclude that Strieff was a short-term visitor who may have been consummating a drug transaction.” (Strieff, at p. 241.) In this case, by contrast, McWilliams had no connection whatsoever with the reported suspicious activity that prompted Officer Croucher’s investigation; he was found alone, seated in a car rather than riding a bicycle, with nary a flashlight in sight. Nor do we view the officer’s conduct in this case as comparable to that in Brendlin, in which we found nothing in the record to indicate the deputy who instigated the stop was engaged in a mere fishing expedition. (Brendlin, supra, 45 Cal.4th, supra, at p. 271.) In Brendlin, the officer offered a justification for the suspicion of criminality that prompted the traffic stop: that, in his experience, cars bearing expired registration tabs and temporary stickers are frequently being operated illegally. Though the justification was insufficient to justify the stop, we explained that “the insufficiency was not so obvious as to make one question [the deputy]’s good faith in pursuing an investigation of what he believed to be a suspicious registration.” (Ibid.) Here, by contrast, Officer Croucher offered no basis — rooted in experience or otherwise — for believing McWilliams was involved in the suspicious parking-lot activity he had set out to investigate. Of course, neither is this case on all fours with Walker, where the investigating officer stopped an individual based on a perceived resemblance to a photograph of a person who, as far as the officer knew, had not committed any crime, and where the two individuals bore no resemblance to one another besides the color of their skin. But while Walker may have involved more flagrant misconduct than this case, it also involved a search conducted after the discovery of a valid arrest warrant; even so, and despite Strieff, the court in that case concluded that the discovery was not sufficiently attenuating under the circumstances of that case. This case, by contrast, concerns a search conducted after the discovery of a parole condition authorizing suspicionless searches — a discovery that, for reasons already discussed, has meaningfully less attenuating force than the discovery of a valid arrest warrant. Here we conclude that the officer’s decision to detain McWilliams merely because he was in the broad vicinity of reported suspicious activity was purposeful and further supports applying the exclusionary rule to deter this type of unconstitutional conduct. (See Brown, supra, 422 U.S. at p. 600; Strieff, supra, 579 U.S. at p. 241.) McWilliams, who is Black, also urges us to find purposefulness and flagrancy based on an inference that racial bias may have played a role in Officer Croucher’s decision to detain him. As McWilliams himself acknowledges, however, nothing in the factual record supports that inference. But he asks us to consider a more general point: that “seemingly small constitutional violations can add up to problems of significant national dimensions.” (People v. McWilliams, supra, H045525 (conc. & dis. opn. of Danner, J.), citing, inter alia, Strieff, supra, 579 U.S. at p. 254 (dis. opn. of Sotomayor, J.) [“it is no secret that people of color are disproportionate victims” of unlawful, suspicionless stops].) The Attorney General acknowledges this broader point, and agrees that courts must be mindful about rules that encourage officers to conduct stops “in an arbitrary manner” and “risk treating members of our communities as second-class citizens.” (Strieff, at p. 252 (dis. opn. of Sotomayor, J.).) As the Attorney General recognizes, a rule permitting officers to rely exclusively on discretionary parole search conditions to purge the taint of unconstitutional, suspicionless detentions would risk creating such incentives; ultimately, a more careful approach to the attenuation analysis “is in the interest of society” as well as “the individuals who experience the deprivation of their Fourth Amendment rights.” In sum, the People have not carried their burden of establishing the attenuation doctrine applies here. No substantial time passed between Officer Croucher’s illegal detention of McWilliams and his seizure of the evidence in this case. Officer Croucher’s subsequent discovery of McWilliams’s parole search condition, and his discretionary decision to conduct the parole search, did little to attenuate the connection between the unlawful stop and the evidence. And Officer Croucher’s decision to conduct the stop, without any evident basis to believe McWilliams was connected to the activity Officer Croucher set out to investigate, indicates a purposefulness that further justifies the exclusion of the evidence. We conclude the evidence Officer Croucher found after his illegal detention of McWilliams is not admissible.5

5 Although People v. Durant, supra, 205 Cal.App.4th 57, found attenuation on a different set of facts, we disapprove the opinion to the extent its reasoning is inconsistent with this opinion. IV. We reverse the judgment of the Court of Appeal and remand for further proceedings consistent with this opinion.

KRUGER, J.

We Concur:

GUERRERO, C. J.

CORRIGAN, J.

LIU, J.

GROBAN, J.

JENKINS, J.

CANTIL-SAKAUYE, J.*

* Retired Chief Justice of California, assigned by the Chief Justice pursuant to article VI, section 6 of the California Constitution. S268320

Concurring Opinion by Justice Liu

I agree with today’s opinion that Officer Croucher’s “discretionary decision to conduct the parole search [of defendant Duvanh McWilliams] did not sufficiently attenuate the connection between the officer’s initial unlawful decision to detain McWilliams and the discovery of contraband.” (Maj. opn., ante, at p. 2.) Our reasoning focuses on the discretionary nature of a search pursuant to a parole search condition. (Id. at pp. 14– 17.) I write separately to note that in such circumstances, an officer’s decision-making may be vulnerable to implicit biases that result in a heightened risk of exploitation of the unlawful detention. This reality is a proper consideration under the second factor of the attenuation doctrine set out in Brown v. Illinois (1975) 422 U.S. 590, 603–604. In analyzing whether the discovery of a parole search condition is an intervening circumstance under the second Brown factor, today’s opinion says: “By contrast to the arrest warrant in [Utah v. Strieff (2016) 579 U.S. 232], a parole search condition merely authorizes a suspicionless search of the parolee for purposes of monitoring the parolee’s rehabilitation and compliance with the terms of parole. It is not a judicial mandate, nor does it compel further action of any sort. Whether to take further action is largely within law enforcement’s discretion . . . .” (Maj. opn., ante, at p. 15.) The attenuation doctrine’s treatment of a search incident to arrest on an outstanding warrant recognizes the mandatory nature of the arrest and reasons that “the ultimate decision to conduct a search . . . [is] attributable to the legally required arrest itself — ‘to protect [the officer]’s safety’ as he carrie[s] out the arrest — and not the initial unlawful decision to stop [the defendant].” (Id. at p. 18, quoting Utah v. Strieff, at p. 241.) A search incident to such an arrest is, as the Attorney General said at oral argument, “something that operates independently of . . . [the officer’s] implicit biases.” The same cannot be said of an officer’s discretionary decision to conduct a search pursuant to a parole condition. Empirical studies have shown that “the conditions under which implicit biases translate most readily into discriminatory behavior are when people have wide discretion in making quick decisions with little accountability.” (Kang et al., Implicit Bias in the Courtroom (2012) 59 UCLA L.Rev. 1124, 1142; see id. at pp. 1142–1150 [citing studies]; Casey et al., Addressing Implicit Bias in the Courts (2013) 49 Ct.Rev. 64, 68 & fn. 38 [citing studies].) As Justice Danner noted in the Court of Appeal, the issue is not racism in the sense of intentional discrimination. It is the operation of “attitudes and stereotypes” that “are not consciously accessible through introspection” and “can function automatically.” (Kang et al., at p. 1129.) Research confirms what is no surprise as a matter of common sense: On-the-spot discretionary decisions are vulnerable to implicit bias because they are neither constrained by a clear rubric of relevant criteria nor preceded by extensive deliberation. Where a discretionary search is preceded by an unlawful detention, the very impulses that may have given rise to the initial detention may also contribute to an officer’s decision to conduct the search. Such impulses may include the well-documented unconscious “stereotype of Black Americans as violent and criminal.” (Eberhardt et al., Seeing Black: Race, Crime, and Visual Processing (2004) 87 J. Personality & Soc. Psychol. 876, 876; see Hetey & Eberhardt, Racial Disparities in Incarceration Increase Acceptance of Punitive Policies (2014) 25 Psychol. Sci. 1949.) Black individuals like McWilliams disproportionately bear the brunt of discretionary decisions by law enforcement. “Black Californians are significantly more likely to be stopped than white Californians, and experiences during stops and outcomes afterward also vary. . . . Black individuals are more than twice as likely to be searched as white individuals.” (Lofstrom et al., Racial Disparities in Law Enforcement Stops (Oct. 2021) p. 25.) Not only are Black people stopped and searched more often, but such searches are less likely to yield evidence or contraband. (Id. at pp. 14–16; see People v. Tacardon (2022) 14 Cal.5th 235, 264 (dis. opn. of Liu, J.) [citing Ayres & Borowsky, A Study of Racially Disparate Outcomes in the Los Angeles Police Department (Oct. 2008) pp. 5–8 [Black and Hispanic residents of Los Angeles, compared to Whites, were more likely to be stopped, frisked, searched, and arrested but significantly less likely to be found with weapons or drugs]; Gross & Barnes, Road Work: Racial Profiling and Drug Interdiction on the Highway (2002) 101 Mich. L.Rev. 651, 668 [searches of White drivers in Maryland reveal drugs 22% more often than searches of Black drivers and over 200% more often than searches of Hispanic drivers]; Note, Discrimination During Traffic Stops: How an Economic Account Justifying Racial Profiling Falls Short (2012) 87 N.Y.U. L.Rev. 1025, 1040 [searches of White drivers in Illinois reveal contraband over 50% more often than searches of non-White drivers]]; cf. Lofstrom et al., at p. 15 [contraband or evidence is found in 21.4 percent of searches overall]; Bar-Gill & Friedman, Taking Warrants Seriously (2012) 106 Nw. U. L.Rev. 1609, 1655 [“police find evidence in only about 10% to 20% of the total traffic searches”].) For every search of a Black person that yields contraband, there are far more — and disproportionately more — searches of Black people that turn up nothing. These practices are not only inefficient but also detrimental to building trust between minority communities and law enforcement. (People v. Tacardon, at p. 264 (dis. opn. of Liu, J.).) As today’s decision explains, “[t]here is a danger that an officer who has unlawfully stopped a bystander without reasonable suspicion will regard the discovery of a parole search condition as a license to continue pursuing a baseless hunch, rather than fairly considering whether a search is appropriate to assess the individual’s rehabilitation and monitor ‘his transition from inmate to free citizen.’ [Citation.] In other words, in the hands of the very same officer who conducted an illegal stop, there is a risk that the discretion to conduct a parole search will lead to the exploitation of that illegal conduct, rather than severing the causal connection between the stop and the search.” (Maj. opn., ante, at p. 17.) It is appropriate for courts to recognize, in applying the second factor of Brown’s attenuation inquiry, that this risk in a given case may be heightened by the operation of implicit biases, including the unconscious association between Blackness and criminality.

LIU, J.

4 See next page for addresses and telephone numbers for counsel who argued in Supreme Court.

Name of Opinion People v. McWilliams __________________________________________________________

Procedural Posture (see XX below) Original Appeal Original Proceeding Review Granted (published) Review Granted (unpublished) XX NP opn. filed 3/8/21 – 6th Dist. Rehearing Granted __________________________________________________________

Opinion No. S268320 Date Filed: February 23, 2023 __________________________________________________________

Court: Superior County: Santa Clara Judge: David A. Cena __________________________________________________________

Counsel:

William M. Robinson, under appointment by the Supreme Court, and Marc McKenna, under appointment by the Court of Appeal, for Defendant and Appellant.

Martin F. Schwarz, Public Defender (Orange), Laura Jose, Chief Public Defender, Adam Vining, Assistant Public Defender, and Abby Taylor, Deputy Public Defender, for Orange County Public Defender’s Office as Amicus Curiae on behalf of Defendant and Appellant.

Xavier Becerra and Rob Bonta, Attorneys General, Lance E. Winters, Chief Assistant Attorney General, Jeffrey M. Laurence, Assistant Attorney General, Catherine A. Rivlin, Karen Z. Bovarick, Seth K. Schalit and Amit Kurlekar, Deputy Attorneys General, for Plaintiff and Respondent.

Jeffrey F. Rosen, District Attorney (Santa Clara), and Jeff Rubin, Deputy District Attorney, for Santa Clara County District Attorney’s Office as Amicus Curiae on behalf of Plaintiff and Respondent. Counsel who argued in Supreme Court (not intended for publication with opinion):

William M. Robinson Sixth District Appellate Program 95 South Market Street, Suite 570 San Jose, CA 95113 (408) 241-6171

Amit Kurlekar Deputy Attorney General 455 Golden Gate Avenue, Suite 11000 San Francisco, CA 94102 (415) 510-3810