United States v. Raphyal Crawford

REINHARDT, Circuit Judge:

Pursuant to a mandatory condition of Raphyal Crawford’s parole, FBI agents entered Crawford’s home to conduct a “parole search” on July 27, 2000. The agents conducted the search despite the fact that they expected to find absolutely no evidence of a crime on the premises, because they thought it would help pressure Crawford into talking about his role in an unsolved robbery committed two years before. Less than two hours later, Crawford confessed to participating in the robbery.

We hold that the search of Crawford’s home without any reasonable suspicion, although pursuant to a parole condition authorizing such searches, violated the Fourth Amendment. Because Crawford’s confession resulted from the suspicionless search of his residence, we reverse the district court’s decision denying his motion to suppress and remand to allow him to withdraw his guilty plea. FED. R. CRIM. P. 11(a)(2).

I. BACKGROUND

Sometime in 1998, FBI Special Agent David Bowdich was assigned to investigate a series of bank robberies that occurred in San Diego in 1997 and 1998, including the February 10, 1998, robbery of a Bank of America on Ulrich Street. Approximately two years later, Bowdich received information from an unnamed source that a person known as “Ralphie Rabbit” was a participant in the Ulrich Street robbery. Bowdich was led to believe that “Ralphie Rabbit” was an alias used by Raphyal Crawford.

Bowdich conducted a background investigation on Crawford and learned that Crawford was on state parole. Bowdich also learned that Crawford had signed what is commonly referred to as a “Fourth Waiver”1 as a condition of his parole. Specifically, the “Fourth Waiver” document contained the following parole conditions:

You and your residence and any property under your control may be searched without a warrant by an agent of the Department of Corrections or any law enforcement officer.
You agree 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.

Bowdich testified that it is a common practice for law enforcement officers to use “Fourth Waivers” as a “kind of tool to talk” to a suspect about crimes. In order to talk to Crawford about the robbery, Bowdich contacted Crawford’s parole agent, Carl Berner, and obtained his per*703mission2 to conduct a parole search of the residence where Crawford was living with his sister.3

Bowdich repeatedly testified that when he conducted the parole search on July 27, 2000, he did not expect to find any evidence linking Crawford to the Ulrich Street robbery or to any other criminal conduct, although he “hoped” he might find something that would show that Crawford was currently engaging in criminal activity. The robbery had occurred more than two years earlier; little physical evidence from the incident remained unaccounted for; Crawford was living in a different residence at the time of the robbery; and in the interim, he had been imprisoned in another state for an unrelated offense. Under these circumstances, Bowdich acknowledged that he did not expect the search to reveal any physical evidence of the bank robbery. Rather, he intended to use the parole search solely as a “tool to see” Crawford and to induce him to talk.

In the early morning of July 27, Bow-dich met four other law enforcement officials, including Detective Michael Gutierrez, and knocked on the door of Crawford’s residence. When Crawford’s sister Abdullah answered the door, Bow-dich stated that the officers were there to conduct a parole search and Abdullah told them that Crawford was in the bedroom asleep with his eighteen-month-old daughter. Bowdich, Gutierrez, and at least one other officer went into the bedroom and found Crawford and his young child on the bed. Both Crawford and Gutierrez stated that the officers had their weapons drawn when they went into the bedroom.

The officers told Crawford that they were conducting a parole search, and escorted him to the living room, where he remained seated on a couch, under “investigatory detention,” through the course of the search. While the search was being conducted, he was not permitted to move, even to get a glass of water.

The search itself may have lasted as long as 50 minutes.4 As expected, no evidence of any criminal activity was discovered. However, as planned, Bowdich used the time to initiate a conversation with Crawford. The discussion began with some “chit chat” that was designed “to put Mr. Crawford at ease” and dispel the “me-*704versus-you” atmosphere. Then Bowdich told Crawford that, although he was not under arrest, he would really like him to talk about “an old bank robbery case.” Crawford initially stated that he did not know anything about the robbery, but Bowdich suspected that he knew more than he was letting on, and thought that Crawford would talk if he was placed in the “right environment.” Bowdich testified that he wanted to eliminate the distractions, bring Crawford to an area where Bowdich was in charge of the scene, and eliminate the possibility that he would end “up in some hearing later where the defense is alleging that I’ve got five officers milling around, and that could be a coercive atmosphere.” In order to make it “as clean as possible,” Bowdich asked Crawford whether he would be “more comfortable” talking to the officers at the FBI office. Crawford agreed to the suggested alternative and was escorted — with an officer on each side — to Bowdich’s vehicle, which was parked outside.5 Crawford was not handcuffed at any point. However, Crawford’s sister testified that she assumed he was under arrest based on the length of the search, the way he was detained in the living room, and the fact that he was leaving with officers walking on both sides of him. Moreover, despite the fact that Crawford’s own car was parked in front of the house, he was transported to the FBI office in Bowdich’s unmarked car.

Bowdich drove to the FBI offices, which were located about 20 minutes from the residence. Detective Gutierrez sat with Crawford in the back seat of the car. Bowdich and Gutierrez continued the “chit chat” with Crawford during the drive, as a means of putting him at ease.

After arriving at the FBI offices, Bow-dich and Gutierrez took Crawford into an interview room, and closed the door “for [ ] privacy.” They then told Crawford that he was not in custody and could leave at any time. Bowdich began to read Crawford his Miranda rights, because he wanted to “make the case cleaner.” Crawford, however, interrupted in protest before Bowdich finished the second line of the warnings. Both agents again told Crawford that he was not in custody and could leave. Bowdich never completed the Miranda warnings.

The interrogation continued for approximately an hour to an hour and a half.6 No weapons were drawn. Eventually, Crawford confessed that he was a participant in the February 10, 1998, robbery and admitted to having used a gun during the crime.

*705On January 16, 2001, Crawford was indicted for armed bank robbery in violation of 18 U.S.C. §§ 2113(a) and (d) and for knowingly using and carrying a firearm in the commission of the robbery in violation of 18 U.S.C. §§ 924(c)(1) and (2).7 Crawford moved to suppress the statements that he made to the law enforcement officers on July 27. After three evidentiary hearings, the district court denied Crawford’s motion on both Fourth and Fifth Amendment grounds. It specifically held that the initial detention of Crawford in his residence pursuant to the parole search was unlawful under United States v. Knights, 219 F.3d 1138 (9th Cir.2000);8 however, under Dunaway v. New York, 442 U.S. 200, 99 S.Ct. 2248, 60 L.Ed.2d 824 (1979), and Brown v. Illinois, 422 U.S. 590, 95 S.Ct. 2254, 45 L.Ed.2d 416 (1975), the court found sufficient attenuating factors between detention and confession to purge the taint, rendering the subsequent statement admissible. The district court also rejected Crawford’s Fifth Amendment claim that he was held in custody during the interrogation and therefore entitled to full Miranda warnings. Finally, the court determined that Crawford’s confession was not involuntary.

Thereafter, Crawford entered a conditional guilty plea to the charged counts pursuant to Federal Rule of Criminal Procedure 11(a)(2). He reserved for appeal the denial of his motion to suppress.9 We review denials of motions to suppress de novo. See United States v. Murillo, 255 F.3d 1169, 1174 (9th Cir.2001), cert. denied, 535 U.S. 948, 122 S.Ct. 1342, 152 L.Ed.2d 245 (2002); United States v. Percy, 250 F.3d 720, 725 (9th Cir.2001); cert. denied, 534 U.S. 1009, 122 S.Ct. 493, 151 L.Ed.2d 405.

II. DISCUSSION

A. The Parole Search

The Fourth Amendment provides that “[t]he right of the people to be secure in their persons, houses, papers, and effects, against unreasonable searches and seizures, shall not be violated.” U.S. CONST, amend. IV. We review for clear error the district court’s underlying factual findings in a Fourth Amendment challenge, and we review de novo the lawfulness of a search or seizure. See United States v. Dorais, 241 F.3d 1124, 1128 (9th Cir.2001); United States v. Hudson, 100 F.3d 1409, 1414 (9th Cir.1996).

A search does not infringe the Fourth Amendment if it is “reasonable,” which we “measure[ ] in objective terms by examining the totality of the circumstances.” Ohio v. Robinette, 519 U.S. 33, 39, 117 S.Ct. 417, 136 L.Ed.2d 347 (1996). See also id. (“We have long held that the ‘touchstone of the Fourth Amendment is reasonableness.’ ”) (quoting Florida v. Jimeno, 500 U.S. 248, 250, 111 S.Ct. 1801, 114 L.Ed.2d 297 (1991)). Following the Supreme Court’s recent decision in United *706States v. Knights, 534 U.S. 112, 122 S.Ct. 587, 151 L.Ed.2d 497 (2001), in order to determine if a parole search is objectively reasonable, we are required to balance the privacy interests of a parolee against the government’s interest in the search.10 Knights, 534 U.S. at 118, 122 S.Ct. 587.

1. Crawford’s Privacy Interest in His Own Home

An individual’s “capacity to claim the protection of the Fourth Amendment depends ... upon whether the person who claims the protection of the Amendment has a legitimate expectation of privacy in the invaded place.” Minnesota v. Carter, 525 U.S. 83, 88, 119 S.Ct. 469, 142 L.Ed.2d 373 (1988) (quoting Rakas v. Illinois, 439 U.S. 128, 143, 99 S.Ct. 421, 58 L.Ed.2d 387 (1978)); United States v. Singleton, 987 F.2d 1444, 1449 (9th Cir.1993); see also United States v. Sarkisian, 197 F.3d 966, 986 (9th Cir.1999) (affirming that a defendant has “standing” to contest a Fourth Amendment violation if he manifests a subjective expectation of privacy in the area searched and the expectation is one that society is prepared to recognize as objectively reasonable).11 In other words, an individual must have “a sufficient connection to the invaded place to assert the protection of the [Fjourth [Ajmendment.” United States v. Davis, 932 F.2d 752, 757 (9th Cir.1991); see also Carter, 525 U.S. at 99, 119 S.Ct. 469 (Kennedy, J., concurring) (“Fourth Amendment rights are personal, and when a person objects to the search of a place and invokes the exclusionary rule, he or she must have the requisite connection to that place.”).

If an individual cannot demonstrate a connection to the invaded place sufficient *707to invoke the Fourth Amendment,12 we need proceed no further. See, e.g., Carter, 525 U.S. at 91, 119 S.Ct. 469. Otherwise, we must determine the extent of the individual’s reasonable expectation of privacy to decide whether the government’s interference with his privacy rights was reasonable.

In this case, the district court concluded that Crawford had an objectively reasonable subjective expectation of privacy in his home.13 We agree.

Crawford’s personal connection to his home is more than sufficient to afford him Fourth Amendment protection against an uninvited search. Indeed, Crawford’s reasonable expectation of privacy must be strongest in his own home. The home is the “prototypical ... area of protected privacy.” Kyllo v. United States, 533 U.S. 27, 34, 121 S.Ct. 2038, 150 L.Ed.2d 94 (2001); cf. Minnesota v. Olson, 495 U.S. 91, 110 S.Ct. 1684, 109 L.Ed.2d 85 (1990) (holding that even an overnight guest has a reasonable expectation of privacy in the premises). The Supreme Court has unambiguously insisted that an individual’s privacy interest in his home must be protected:

The Fourth Amendment protects the individual’s privacy in a variety of settings. In none is the zone of privacy more clearly defined than when bounded by the unambiguous physical dimensions of an individual’s home — a zone that finds its roots in clear and specific constitutional terms: ‘The right of the people to be secure in their ... houses ... shall not be violated.’ That language unequivocally establishes the proposition that ‘[a]t the very core [of the Fourth *708Amendment] stands the right of a man to retreat into his own home and there be free from unreasonable governmental intrusion.’

Payton v. New York, 445 U.S. 573, 589-90, 100 S.Ct. 1371, 63 L.Ed.2d 639 (1980) (quoting Silverman v. United States, 365 U.S. 505, 511, 81 S.Ct. 679, 5 L.Ed.2d 734 (1961)). See also Kyllo, 533 U.S. at 44, 121 S.Ct. 2038 (Stevens, J., dissenting) (“To be sure, the homeowner has a reasonable expectation of privacy concerning what takes place within the home.... ”); United States v. Johnson, 457 U.S. 537, 552 n. 13, 102 S.Ct. 2579, 73 L.Ed.2d 202 (1982) (“At least since Boyd v. United States, 116 U.S. 616, 630, 6 S.Ct. 524, 29 L.Ed. 746[] (1886), the Court ha[s] acknowledged that the Fourth Amendment accords special protection to the home.”); United States v. Martinez-Fuerte, 428 U.S. 543, 562, 96 S.Ct. 3074, 49 L.Ed.2d 1116 (1976) (“[T]he sanctity of private dwellings [is] ordinarily afforded the most stringent Fourth Amendment protection.”); L.A. Police Protective League v. Gates, 907 F.2d 879, 884 (9th Cir.1990) (“Nowhere is the protective force of the fourth amendment more powerful than it is when the sanctity of the home is involved.... The sanctity of a person’s home, perhaps our last real retreat in this technological age, lies at the very core of the rights which animate the amendment.”). If an individual has a reasonable expectation of privacy anywhere, he must surely reasonably expect privacy in his own home.

The dissent, placing great weight on recent decisions of the California Supreme Court, claims that Crawford’s status as a parolee deprives him of any reasonable expectation of privacy anywhere, even in his own bedroom. As the district court properly noted, however, federal law — not California,law — governs the extent of the protection that the Fourth Amendment provides.14 See Ooley, 116 F.3d at 372; Davis, 932 F.2d at 758. Accordingly, we must look to federal law to determine the expectation of privacy that Crawford reasonably possesses; once this threshold is established, a state may not define away the constitutional protection.

Under federal law, Crawford’s expectation of privacy in his own home is not wholly defeated by virtue of his parole status. As the Supreme Court has recognized, “A probationer’s home, like anyone else’s, is protected by the Fourth Amendment’s requirement that searches be ‘reasonable.’” Griffin v. Wisconsin, 483 U.S. 868, 873, 107 S.Ct. 3164, 97 L.Ed.2d 709 (1987).15 To find otherwise would be to equate a parolee’s home with a prisoner’s cell — a comparison that the Supreme Court has unequivocally rejected. Compare Hudson v. Palmer, 468 U.S. 517, 525-26, 104 S.Ct. 3194, 82 L.Ed.2d 393 (1984) (finding that the Fourth Amendment “does not apply within the confines of a prison *709cell”); with Morrissey v. Brewer, 408 U.S. 471, 482, 92 S.Ct. 2593, 33 L.Ed.2d 484 (1972) (“Though the State properly subjects [a parolee] to many restrictions not applicable to other citizens, his condition is very different from that of confinement in a prison.”). As we stated en banc over a generation ago:

Moreover, the theory upon which courts have usually relied to justify stripping parolees of Fourth Amendment protection has been widely criticized. Commentators have repeatedly criticized the notion that the status of parolees is legally comparable to that of prisoners in actual custody as being logically inconsistent and ignoring reality.... [T]he Supreme Court has specifically rejected the theory that parole officers have unfettered discretion in dealing with parolees, and refused to attach so broad a significance to the “custody” theory.
It is thus too late in the day to assert that searches of parolees by their parole officers present no Fourth Amendment issues. Rather, such searches may be held illegal and the evidence obtained therefrom suppressed unless they pass muster under the Fourth Amendment test of reasonableness.

Latta v. Fitzharris, 521 F.2d 246, 248-49 (9th Cir.1975) (en banc) (citations omitted).16 See also Sepulveda v. Ramirez, 967 F.2d 1413, 1416 (9th Cir.1992) (“[T]he constitutional rights of parolees are even more extensive than those of inmates.”).17

*710It is true that Crawford’s parole status reduces the “expectation of privacy ... that society is prepared to recognize as reasonable.” Bond v. United States, 529 U.S. 334, 338, 120 S.Ct. 1462, 146 L.Ed.2d 365 (2000) (citations and internal quotation marks omitted). However, a reduced expectation of privacy is substantially different from an extinguished expectation of privacy. Although parolees are subject to specific constraints on their privacy that “would not be constitutional if applied to the public at large,” Griffin, 483 U.S. at 875, 107 S.Ct. 3164, their privacy interests are not eliminated entirely. Indeed, the purposes of parole require a reasonable amount of privacy. Parole represents an interim state between custody and freedom, “critical to successful reintegration of the offender into society and to positive citizenship.” CAL. .PENAL CODE § 3000. It would be unreasonable to expect a parolee to negotiate the transition into the life of a normal citizen without some measure of the privacy that normal citizens take for granted.

The Supreme Court’s most recent discussion of parole searches confirms that a parolee has an objectively reasonable expectation of privacy in his home. In United States v. Knights, 534 U.S. 112, 122 S.Ct. 587, 151 L.Ed.2d 497 (2001), the Supreme Court confronted a search of the home of a probationer subject to a probation search condition. In examining the totality of the circumstances, the Court found the probationer’s privacy interest to be “significantly diminished” — but not extinguished. Knights, 534 U.S. at 118-19, 122 S.Ct. 587. Similarly, we find in this case that Crawford had a diminished but still objectively reasonable expectation of privacy in his home.

2. Balancing Crawford’s Privacy Against the Government’s Intrusion

In Knights, to determine if the probation search was reasonable, the Court weighed the probationer’s diminished privacy interest against the government’s interest in preventing and punishing crimes committed by probationers lapsing into recidivism. Id. at 119, 122 S.Ct. 587. The Court emphasized that reasonableness was to be determined by examining the totality of the circumstances, “with the probation search condition being a salient circumstance.”18 Id. at 118, 122 S.Ct. 587.

Normally, of course, the search of a home is only reasonable for Fourth Amendment purposes if it is conducted pursuant to a warrant grounded in probable cause. See, e.g., Payton, 445 U.S. at 586, 100 S.Ct. 1371; cf. Illinois v. McArthur, 531 U.S. 326, 330, 121 S.Ct. 946, 148 L.Ed.2d 838 (2001) (noting the same for seizures). Under the circumstances of a parole search pursuant to a “Fourth Waiver” condition, however, the Knights Court found that a search could be reasonable without strict adherence to the ordinary *711probable cause requirements.19 See Knights, 534 U.S. at 119-21, 122 S.Ct. 587. Instead, some lesser amount of individualized suspicion suffices because “the balance of governmental and private interests makes such a standard reasonable.” Id. at 119, 122 S.Ct. 587. The Court noted that “[t]he degree of individualized suspicion required of a search is a determination of when there is a sufficiently high probability that criminal conduct is occurring to make the intrusion on the individual’s privacy interest reasonable.” Id (citing United States v. Cortez, 449 U.S. 411, 418, 101 S.Ct. 690, 66 L.Ed.2d 621 (1981)) (emphasis added). After acknowledging that both parties in the case before it had conceded the presence of reasonable suspicion, the Court found that reasonable suspicion provided a sufficiently high probability of criminal conduct to render the overall balance reasonable and the search valid.

Knights explicitly refused to consider whether, in the case of probationers, the constitutional requirement of reasonableness could be satisfied without individualized suspicion. See id. at 119 n. 6, 122 S.Ct. 587; see also Pa. Bd. of Prob. & Parole v. Scott, 524 U.S. 357, 362 n. 3, 118 S.Ct. 2014, 141 L.Ed.2d 344 (1998) (reserving the same question in the case of parolees). However, it is implicit in the Court’s statements in Knights and Griffin that probation and parole searches are limited by some reasonable and legally protectable privacy interest. See Knights, 534 U.S. at 119, 122 S.Ct. 587 (“When an officer has reasonable suspicion that a probationer subject to a search condition is engaged in criminal activity, there is enough likelihood that criminal conduct is occurring that an intrusion on the probationer’s significantly diminished privacy interests is reasonable.”) (emphasis added); Griffin, 483 U.S. at 875, 107 S.Ct. 3164 (“Supervision [of parolees] is a ‘special need’ of the State permitting a degree of impingement upon privacy that would not be constitutional if applied to the public at large. That permissible degree is not unlimited, however ....”) (emphasis added). By mandating that we balance the government’s interests against the privacy interests of a probationer or parolee, and by declaring the individual’s privacy interests to be diminished but not extinguished, see supra section II.A.1, the Supreme Court has made it clear that in the case of searches pursuant to parole conditions, the ordinary search requirements are to be relaxed but not eliminated.

Moreover, the fact that a parole search invades the home must weigh heavily in the “totality of the circumstances” that determines whether such a search is reasonable. See Knights, 534 U.S. at 118, 122 S.Ct. 587; see also Carter, 525 U.S. at 88, 119 S.Ct. 469 (“[T]he extent to which the Fourth Amendment protects people may depend upon where those people are.”).20 As noted above, supra section *712II.A.1, the Supreme Court considers the home sacrosanct, and permits government searches of the home only pursuant to enhanced procedural safeguards. Neither the Supreme Court nor this court has ever approved a suspicionless search of a home for a law enforcement purpose. To do so here would represent a substantial incursion into previously inviolate constitutional territory.

The Supreme Court’s “special needs” jurisprudence, cited with such enthusiasm by the dissent, does not support a different conclusion. The dissent correctly characterizes parole as a “special need” of the state, but incorrectly concludes that upon invocation of that phrase, Fourth Amendment protections vanish. Griffin itself stated otherwise. The Court found that “special needs” associated with the probation and parole system may “justify departures from the usual warrant and probable-cause requirements,” Griffin, 483 U.S. at 874, 107 S.Ct. 3164 (emphasis added), but unmistakably held that the permissible impingement on a parolee’s privacy is not unlimited. Id. at 875, 107 S.Ct. 3164. Certainly, nothing in Griffin purports to authorize substantial invasions of a parolee’s privacy without any suspicion of individual wrongdoing whatsoever.21

The Court’s “special needs” cases since Griffin — and our own cases following the “special needs” line — similarly provide no support for the dissent’s absolutist position. These cases reveal that each of two factors relevant to the search at issue independently bars this type of search and precludes it from coming within the “closely guarded” set of “special needs” cases authorizing suspicionless searches, Ferguson v. City of Charleston, 532 U.S. 67, 84, 121 S.Ct. 1281, 149 L.Ed.2d 205 (2001). Those factors are: (1) that the search was of Crawford’s home, and (2) that the government’s practice was designed to discover evidence of crime for future prosecution. As to the first factor, in a few exceptional “special needs” cases, searches not founded on any degree of individualized suspicion have been approved — but these searches have not involved the home. See Bd. of Educ. v. Earls, 536 U.S. 822, 122 S.Ct. 2559, 153 L.Ed.2d 735 (2002) (drug tests for extra-curriculars at school); United States v. Gonzalez, 300 F.3d 1048 (9th Cir.2002) (searches of employee backpacks to prevent loss of inventory); Vernonia Sch. Dist. 47J v. Acton, 515 U.S. 646, 115 S.Ct. 2386, 132 L.Ed.2d 564 (1995) (drug tests of athletes at school); Mich. Dept. of State Police v. Sitz, 496 U.S. 444, 110 S.Ct. 2481, 110 L.Ed.2d 412 (1990) (highway sobriety checkpoints); Skinner v. Ry. Labor Executives’ Ass’n, 489 U.S. 602, 109 S.Ct. 1402, 103 L.Ed.2d 639 (1989) (railroad employees’ drug tests at work); Nat’l Treasury Employees Union v. Von Raab, 489 U.S. 656, 109 S.Ct. 1384, 103 L.Ed.2d 685 (1989) (customs employees’ drug tests at work); New York v. Burger, 482 U.S. 691, *713107 S.Ct. 2636, 96 L.Ed.2d 601 (1987) (purely administrative search of regulated business); see also McMorris v. Alioto, 567 F.2d 897 (9th Cir.1978) (purely administrative search in public buildings); United States v. Martinez-Fuerte, 428 U.S. 543, 96 S.Ct. 3074, 49 L.Ed.2d 1116 (1976) (fixed checkpoint routine border search).22 In its enthusiastic embrace of decisions making possible the invasive searching of “[laterally hundreds of thousands of suspicion-free, conviction-free citizens of our nation,” see post at 735, the dissent completely ignores the significance of the fact that here, unlike in every suspicionless search approved above, law enforcement officials burst into a person’s home, where the protective force of the Fourth Amendment is at its most powerful. See supra section II.A.1.

Second, the Supreme Court has recently emphasized that it has never approved a suspicionless “special needs” search conducted for criminal law enforcement purposes. In Ferguson, the Court struck down a program collecting and screening urine from pregnant mothers, without individualized suspicion of drug use, in order to preserve evidence of cocaine abuse for later prosecution. See 532 U.S. at 72-73, 85-86, 121 S.Ct. 1281. The factor rendering that program unconstitutional could not have been clearer:

The immediate objective of the searches was to generate evidence for law enforcement purposes ... [footnote:] We italicize those words lest our reasoning be misunderstood. In none of our previous special needs cases have we upheld the collection of evidence for criminal law enforcement purposes.

Id. at 83 & n. 20, 121 S.Ct. 1281 (emphasis in original); see also id. at 88, 121 S.Ct. 1281 (Kennedy, J., concurring) (“None of our special needs precedents has sanctioned the routine inclusion of law enforcement, both in the design of the policy and in using arrests, ... as an integral part of a program which seeks to achieve legitimate, civil objectives. The traditional warrant and probable-cause requirements are waived in our previous cases on the explicit assumption that the evidence obtained in the search is not intended to be used for law enforcement purposes.”); Earls, 122 S.Ct. at 2564, 2566 (permitting suspicion-less searches because drug “test results are not turned over to any law enforcement authority” for use in criminal proceedings); City of Indianapolis v. Edmond, 531 U.S. 32, 38, 121 S.Ct. 447, 148 L.Ed.2d 333 (2000) (“In none of these [sus-picionless search] cases, however, did we indicate approval of a checkpoint program whose primary purpose was to detect evidence of ordinary criminal wrongdoing.”). As with the common practice of using “Fourth Waiver” conditions to search the homes of parolees, “[t]he stark ... fact that characterizes this case is that [the Ferguson suspicionless screening] was designed to obtain evidence of criminal conduct by the [individuals searched] that would be turned over to the police and that could be admissible in subsequent criminal prosecutions.” Ferguson, 532 U.S. at 85-86, 121 S.Ct. 1281. When, as in this case, searches are so designed, the Supreme *714Court has never granted its approval in the absence of individualized suspicion.

Because permitting a suspicionless search of Crawford’s residence would offend both the protected status of the home and the bar against suspicionless searches for evidence of criminal conduct, we conclude that some individualized suspicion was required here. The appropriate standard in such cases is reasonable suspicion-the standard authorized by Knights and discussed in our past parole and probation decisions.23 See, e.g., United States v. Guagliardo, 278 F.3d 868, 873 (9th Cir.2002) (affirming the validity of a probation term authorizing a warrantless search at any time by any officer, “as long as the search was supported by reasonable suspicion”), cert. denied, — U.S. -, 123 S.Ct. 515, 154 L.Ed.2d 401; Davis, 932 F.2d at 758 (“The permissible bounds of a probation search are governed by a reasonable suspicion standard.”); United States v. Richardson, 849 F.2d 439, 442 (9th Cir.1988) (noting that the law permits “searches of probationers without warrants and only upon a showing of reasonable cause”);24 see also Moreno v. Baca, 2002 WL 338366, at *10 (C.D.Cal.2002) (“Given the holdings in [Knights and Griffin], the Court finds that at least reason*715able suspicion is required to justify the search [of the parolee].”).

The dissent proposes an alternative' to reasonable suspicion: the “arbitrary, capricious, or harassing” standard propounded by the California Supreme Court. See Reyes, 19 Cal.4th at 758-54, 80 Cal.Rptr.2d 734, 968 P.2d 445; post at 723. Reyes adopted this standard only after it found individualized suspicion wholly unnecessary in a parole search. We are not free to adopt the Reyes “suspicionless” standard. As we have explained supra, in light of Supreme Court precedent applying the Fourth Amendment, a degree of individualized suspicion is constitutionally required in order to conduct a search of the home of a parolee. The federal Constitution, of course, governs our decision here. See supra note 17. To the extent that the dissent follows California in proposing its standard as a substitute for a constitutionally required degree of individualized suspicion, we are compelled to reject that approach.

Nor would we accept the dissent’s proposed standard even if we were willing, contrary to Reyes, to characterize the standard as itself embodying a degree of individualized suspicion somewhere between reasonable suspicion and no suspicion at all. The Supreme Court has specifically cautioned against creating such new federal categories for measuring the constitutionality of government actions in Fourth Amendment cases.25 See, e.g., United States v. Montoya de Hernandez, 473 U.S. 531, 541, 105 S.Ct. 3304, 87 L.Ed.2d 381 (1985) (criticizing the creation of a new Fourth Amendment standard in addition to “reasonable suspicion” and “probable cause”); see also United States v. Payne, 181 F.3d 781, 788 n. 5 (6th Cir.1999) (applying the Montoya de Hernandez restriction to the parole search context).26 Accordingly, we would not do so here.

After examining the totality of the circumstances — including Crawford’s parole status, the parole condition, the location of the search, Crawford’s expectation of privacy in his own home, the state’s interest in rehabilitating parolees, and the interest of both the state and federal government in preventing and punishing recidivist crimes — we hold that a search of a parolee’s home pursuant to a parole condition is *716reasonable only if it is supported by reasonable suspicion.

In this case, it is clear from the record that the law enforcement officials responsible for searching Crawford’s residence did not have reasonable suspicion that he was engaged in continuing criminal activity or that evidence of the two-year-old bank robbery might be found in his home, even under a generous view of the “totality of the circumstances.” See United States v. Arvizu, 534 U.S. 266, 273, 122 S.Ct. 744, 151 L.Ed.2d 740 (2002) (“When discussing how reviewing courts should make reasonable-suspicion determinations, we have said repeatedly that they must look at the ‘totality of the circumstances’ of each case .... ”).27 Indeed, the government conceded as much in the district court and in its brief on appeal. Both Bowdich and Berner repeatedly testified that when Bowdich conducted the parole search, he did not expect to find any evidence linking Crawford to the bank robbery or to any other ongoing criminal activity. Bowdich further acknowledged that he had ample reason to believe that evidence of the robbery would not be on the premises. The district court specifically found this testimony credible.28 On this record, we hold that the law enforcement officials conducting the parole search of Crawford’s home on July 27, 2000, clearly did not have reasonable suspicion to believe that the search would disclose any evidence of criminal activity. Because there was no reasonable suspicion, we agree with the *717district court that the parole search was illegal.29

3 Consent

The government also contends that the parole search was permissible because Crawford completely waived all of his Fourth Amendment rights by signing the standard compulsory “Fourth Waiver” parole condition.30 The essence of this theory is that the state may preemptively force all parolees to “consent” in blanket fashion to searches that would be unreasonable under the Fourth Amendment.31 We hold that a compulsory parole condition may not serve as a consent to engage in otherwise unreasonable searches, and that Crawford therefore did not consent to the parole search of his home.

*718It is clear that “a search conducted pursuant to a valid consent is constitutionally permissible.” Schneckloth v. Bustamonte, 412 U.S. 218, 222, 93 S.Ct. 2041, 36 L.Ed.2d 854 (1973). However, the Supreme Court has consistently demanded that the consent given be “valid” and meet well-established requirements:

When a prosecutor seeks to rely upon consent to justify the lawfulness of a search, he has the burden of proving that the consent was, in fact, freely and voluntarily given. This burden cannot be discharged by showing no more than acquiescence to a claim of lawful authority.

Bumper v. North Carolina, 391 U.S. 543, 548-49, 88 S.Ct. 1788, 20 L.Ed.2d 797 (1968).

As the record reveals, before a prisoner becomes a parolee, and while still in custody, he is given a sheet of conditions that his prospective parole agent “has” him sign. The purported blanket waiver is among these conditions. If Crawford did not sign the conditions sheet, he would have been denied parole and returned immediately to prison custody. To call this choice — either waiver or certain incarceration — -“free and voluntary” would be to misconceive the concept of meaningful consent. Under these conditions, we cannot equate acceptance of the compulsory condition with the sweeping voluntary consent the government claims it represents. Cf. Smyth v. Lubbers, 398 F.Supp. 777, 788 (W.D.Mich.1975) (“[A] blanket authorization in an adhesion contract [waiving Fourth Amendment rights] is not the type of focused, deliberate, and immediate consent contemplated by the Constitution.”); WAYNE R. LAEAVE, 4 SEARCH AND SEIZURE: A TREATISE ON THE FOURTH AMENDMENT § 10.10 (3d ed. 1996) (“[T]o speak of consent in this context [of a signature to a condition of release] is to resort to a ‘manifest fiction,’ for the probationer who purportedly waives his rights by accepting such a condition has little genuine option to refuse, and the waiver cannot be said to be voluntary in any generally-accepted sense of the term.”) (citation and internal quotation marks omitted).

Moreover, if the mandatory parole condition were deemed to be a valid blanket consent, any constitutional protection for parolees would be rendered illusory — every state could force its parolees to sign the blanket waiver as a condition of parole, and every parolee’s constitutional rights would thereby instantly vanish. Indeed, under the government’s theory, all parolees could be forced to waive all constitutional rights, including the right to due process in revocation proceedings, or even the right to trial on any new offense allegedly committed during the parole period. We hold that parolees may not generally be forced as a threshold condition of their parole to surrender by blanket waiver their Fourth Amendment rights, including those so recently recognized by Knights.

Indeed, we have previously expressed unequivocally our disapproval of general waivers of Fourth Amendment rights as conditions of parole. See Toomey, 898 F.2d at 744 (“We do not approve of general waivers of fourth amendment rights as a condition of parole.”) (citations omitted); cf. Anobile v. Pelligrino, 303 F.3d 107, 123-25 (2d Cir.2002) (refusing, under the totality of the circumstances, to construe a purported blanket waiver as valid consent to otherwise unreasonable searches). We have no cause to question the vitality of that conclusion here.

Our holding on this issue is appropriately narrow. We find that, by virtue of a signature on a compulsory parole condition, a parolee does not, in advance and in blanket fashion, consent to a general waiv*719er of his rights under the Fourth Amendment.32

B. Attenuation

We next consider whether Crawford’s statements to the law enforcement officials were fruits of the illegal parole search and his “investigatory detention” during that search. Unless the taint from the illegal search and detention was sufficiently attenuated, Crawford’s later statements must be suppressed. Wong Sun v. United States, 371 U.S. 471, 83 S.Ct. 407, 9 L.Ed.2d 441 (1963). “We review de novo the mixed question of fact and law whether evidence deriving from an illegal search is sufficiently tainted to require suppression, because legal concepts must be applied and judgment exercised about the values that animate the Fourth Amendment.” United States v. Johns, 891 F.2d 243, 244 (9th Cir.1989).

The pivotal question in determining attenuation is “whether, granting establishment of the primary illegality, the evidence ... has been come at by exploitation of that illegality or instead by means sufficiently distinguishable to be purged of the primary taint.” Wong Sun, 371 U.S. at 487-88, 83 S.Ct. 407 (internal quotation marks omitted). In order to determine whether Crawford’s statement was “come at by exploitation of’ the illegal parole search and detention, we consider three factors: (1) the temporal proximity of the illegal search and detention to the statement; (2) the presence of any intervening circumstances; and, “particularly,” (3) the “purpose and flagrancy” of the official misconduct. Taylor v. Alabama, 457 U.S. 687, 690, 102 S.Ct. 2664, 73 L.Ed.2d 314 (1982); Dunaway v. New York, 442 U.S. 200, 218, 99 S.Ct. 2248, 60 L.Ed.2d 824 (1979); Brown v. Illinois, 422 U.S. 590, 603-04, 95 S.Ct. 2254, 45 L.Ed.2d 416 (1975). The “burden of showing admissibility rests, of course, on the prosecution.” Brown, 422 U.S. at 604, 95 S.Ct. 2254. The prosecution has not met that burden here.

1. Temporal Proximity

As the district court correctly noted, the first factor weighs heavily in favor of suppressing the statements. Only a twenty-minute drive to the FBI offices separated Crawford’s illegal detention and his interrogation. The district court was correct to find that so little elapsed time was insufficient to purge the taint. See, e.g., Brown, 422 U.S. at 604, 95 S.Ct. 2254 (less than two hours not sufficient to purge taint); Taylor, 457 U.S. at 691, 102 S.Ct. 2664 (six hours not sufficient); United States v. Perez-Esparza, 609 F.2d 1284, 1290 (9th Cir.1980) (three hours not sufficient); United States v. George, 883 F.2d 1407, 1416 (9th Cir.1989) (“As best we are aware, no court has weighed the first factor against a defendant when his inculpatory statement followed illegal police conduct by only a few hours.”). Although “[t]he lack of a significant intervening period of time does not, in itself, require that the evidence be suppressed for want of sufficient attenuation,” United States v. Wellins, 654 F.2d *720550, 555 (9th Cir.1981), it does “bear[] directly on the probability of taint,” United States v. Delgadillo-Velasquez, 856 F.2d 1292, 1300 (9th Cir.1988).

2. Intervening Circumstances

The district court concluded that the limited time between the illegal parole search and Crawford’s statements did not purge the taint, because it found the temporal proximity outweighed by the other two prongs of the attenuation test. Specifically, with respect to the second prong, the district court found the following intervening circumstances: (1) that Crawford was “street savvy” and thus must have known that the agents had found nothing in the search; (2) that Crawford was told several times that he was not under arrest and was free to leave; (3) that Crawford voluntarily chose the FBI office as the venue for the interrogation; and (4) that Crawford’s interruption of the Miranda warnings showed his exercise of some free will.

By focusing on Crawford’s state of mind rather than on intervening events, the district court misconceived the nature of the “intervening circumstances” analysis. The facts relied on by the district court are, like Miranda warnings, principally useful in determining whether a confession was voluntary for Fifth Amendment purposes. However, the determination that a statement is voluntary “for purposes of the Fifth Amendment is merely a threshold requirement for Fourth Amendment analysis. The reason for this approach is clear: [t]he exclusionary rule, ... when utilized to effectuate the Fourth Amendment, serves interests and policies that are distinct from those it serves under the Fifth Amendment,” Taylor, 457 U.S. at 690, 102 S.Ct. 2664 (citations and internal quotation marks omitted); see also Brown, 422 U.S. at 600-04, 95 S.Ct. 2254.

Fourth Amendment attenuation analysis focuses on the circumstances that serve the twin interests of the exclusionary rule: deterrence and judicial integrity. See, e.g., Brown, 422 U.S. at 599-600, 95 S.Ct. 2254. We look not at the defendant’s conduct, but rather at “intervening events of significance” that “render inapplicable the deterrence and judicial integrity purposes that justify excluding [a tainted] statement.” See United States v. Ricardo D., 912 F.2d 337, 343 (9th Cir.1990); Perez-Esparza, 609 F.2d at 1289. Intervening circumstances that militate in favor of attenuation must be sufficiently important to ensure that potentially tainted evidence was “come at by way of’ some process other than the exploitation of an illegal search. Wong Sun, 371 U.S. at 487-88, 83 S.Ct. 407. Examples include release from custody, an appearance before a magistrate, or consultation with an attorney, “such that we would be able to say that [a defendant’s decision to confess] was an ‘unconstrained, independent decision’ that was completely unrelated to [the] initial unlawful” violation. George, 883 F.2d at 1416; see also Perez-Esparza, 609 F.2d at 1289 (“In some cases, the intervening, completely self-motivated decision of a putative defendant to inculpate himself is so unforeseeable an event, from the arresting officer’s vantage point, that excluding the defendant’s statement would serve no deterrent purpose.”).

There were no such intervening events here. Crawford was taken, in continuous law enforcement presence, directly from the site of his illegal search and detention to the site of his interrogation. He did not speak to an attorney, or, indeed, any individual other than Bowdich and Gutierrez.33 *721No circumstance cited by the district court amounts to an event of the type or significance necessary to purge taint. As with the first factor, we find that the intervening circumstances prong weighs against a finding of attenuation. See Perez-Esparza, 609 F.2d at 1290 (“The deterrence rationale was not vitiated, as in Wong Sun, by a lengthy period away from police influences. Nor, for the same reasons, can we find that[the defendant’s] decision to speak was so independent of police pressures as to absolve the judicial system from the charge of savoring the forbidden fruits of unconstitutional conduct. The first two factors thus dictate a finding of no attenuation.”).

3. “Purpose and Flagrancy” of the Official Misconduct

The district court made a similar error in its analysis of the third, “particularly important,” prong. George, 883 F.2d at 1416. In holding that the third factor weighed in favor of attenuation, the district court credited the agents’ testimony that they did not use any implied promises in their interrogation. As above, however, this type of analysis “betrays a lingering confusion between ‘voluntariness’ for purposes of the Fifth Amendment and the ‘causal connection’ test established in Brown. Satisfying the Fifth Amendment is only the ‘threshold’ condition of the Fourth Amendment analysis required by Brown.” Dunaway, 442 U.S. at 219, 99 S.Ct. 2248.

Instead, under a proper analysis, this prong suggests suppression if the law enforcement officials conducted the illegal search with the purpose of extracting the evidence in question, or if they flagrantly broke the law in conducting the search. In reciting the Brown factors, courts usually choose a conjunctive phrasing (“purpose and flagrancy”), but the same courts then find in favor of taint if there is evidence of either improper purpose or flagrant illegality. See, e.g., Taylor, 457 U.S. at 693, 102 S.Ct. 2664 (only “purpose”); Dunaway, 442 U.S. at 218-19, 99 S.Ct. 2248 (only “purpose”); United States v. Jenkins, 938 F.2d 934, 941 (9th Cir.1991) (only “flagrancy”); George, 883 F.2d at 1416 (only “flagrancy”). We also find the disjunctive analysis more persuasive, and explicitly clarify that either improper purpose or flagrant illegality will support a determination that the third prong of the test weighs against attenuation. Although either element alone would suffice, we find both present here.

Bowdich testified extensively that the search was executed in order to induce Crawford to talk. In other words, the evidence that was ultimately obtained was not the mere byproduct of the search, but its primary objective. Moreover, by conducting the search solely to pressure Crawford into talking, Bowdich and his accompanying officers blatantly ignored then-existing Ninth Circuit law prohibiting “law enforcement/investigatory” parole searches.34 See supra note 10. The search and detention was not questionably *722illegal; it was flagrantly so.35 This combination calls for hornbook application of the exclusionary rule, which is designed to deter government officials from conducting illegal searches by suppressing the evidence that provoked the search. See, e.g., Brown, 422 U.S. at 605, 95 S.Ct. 2254 (refusing to purge taint where “[t]he arrest, both in design and in execution, was investigatory. The detectives embarked upon this expedition for evidence in the hope that something might turn up.”); Perez-Esparza, 609 F.2d at 1289 (“When police purposely effect an illegal arrest or detention in the hope that custodial interrogation will yield incriminating statements, the deterrence rationale for application of the exclusionary rule is especially compelling.”). In fact, we have found this factor “decisive most often in those cases where [sic] police officers ... took a suspect into custody hoping that an interrogation would yield incriminating statements.” George, 888 F.2d at 1416. We find that the third prong of the attenuation analysis therefore weighs heavily in favor of suppression.

4. Conclusion

We find that, in both design and effect, Crawford’s statements at the FBI offices were “come at by exploitation of’ the illegal parole search. Wong Sun, 371 U.S. at 487-88, 83 S.Ct. 407. The government has not met its burden of showing that the taint of the search was attenuated by any of the factors articulated in Brown or its progeny. Therefore, as a fruit of the illegal search, Crawford’s statements must be suppressed.

III. CONCLUSION

Because law enforcement officials conducted a search of Crawford’s home without reasonable suspicion to believe that they would uncover evidence of criminal activity, we hold that the search violated the Fourth Amendment. The fact that Crawford signed a blanket “Fourth Waiver” as a mandatory condition of his parole does not serve to waive the minimum constitutional protection of reasonable suspicion to which he and other parolees are entitled. Moreover, as neither time, intervening events, nor the officers’ motives purged the taint of the unconstitutional search, Crawford’s subsequent statements must be suppressed. We therefore reverse the district court’s denial of the motion to suppress and remand to allow Crawford to withdraw his guilty plea.

Every court action protecting the constitutional rights of individuals evokes vigorous objections from those who foresee disastrous consequences; the comments of our learned dissenting colleague demonstrate that this opinion is no exception. Surely, law enforcement could succeed in incarcerating a greater number of dangerous individuals if we dispensed entirely with the Fourth and Fifth Amendments, or even with the Sixth. However, our Found*723ers chose a Constitution that balances liberty and security, and that preserves to all individuals certain guarantees against the existence of a police state. We have sought to balance the relevant interests and thereby to honor our obligation to the Constitution. We very much doubt that what we have done will be understood or valued only by “the Richard Allen Davises of the underworld.”36 Post at 737.

REVERSED AND REMANDED.

. In referring to the parole condition as a "Fourth Waiver,” we adopt the government’s preferred nomenclature. For purposes of all but Section II.A.3 infra, we treat the "Fourth Waiver” precisely as the dissent urges that it be treated-as a mandatory condition of parole. For purposes of Section II.A.3, however, we treat the condition as a purported waiver in order to address the government's arguments in that respect.

. Both Bowdich and Bemer testified that Berner’s permission was not legally required for Bowdich, an FBI agent, to conduct the search; rather, Bowdich secured Berner’s agreement as a matter of courtesy.

. At the hearing on Crawford’s motion to suppress, the government relied exclusively on the "Fourth Waiver” as the basis for the search. Crawford’s counsel attempted to ask Bowdich about the source and reliability of the information connecting Crawford to the robbery, but the prosecutor objected and the court sustained the objection. Crawford's counsel pressed the issue, stating that he thought the "quantum of information which was available” to the investigators was relevant. The prosecutor replied that the information available to them was "not relevant for determining” the Fourth and Fifth Amendment issues. The court ultimately stated that it would assume for purposes of the motion that even though the government had information that Crawford was a participant in the robbery, there was not probable cause to arrest him. See infra note 28. There is no evidence in the record regarding the nature or source of the information that led Bowdich to suspect Crawford. More important, Bow-dich’s testimony makes it clear that he had no reason to suspect that he would find evidence relating to a crime at Crawford’s home, and thus had no basis other than the "Fourth Waiver” for the search. See id.

.Crawford and his sister thought that the search took about an hour, and police records indicate that the officers were in the residence from 8:20 a.m. until approximately 9:10 a.m.

. According to Crawford, he did not at first agree to leave. He told the officers that he did not want to go anywhere because he had an appointment related to his child support. The officers then told him that, because he was on parole, he could be detained in any event. Crawford stated that because he did not feel that he had a choice, he went with the officers.

According to Detective Gutierrez, Crawford was given a choice of either continuing the interview in his home or going to the FBI office to complete the interview. He said that he would rather go to the FBI office.

. There is, of course, some dispute over precisely what was said during the interrogation. Crawford testified that he did not feel that he was free to leave because whenever he indicated that he wanted to leave, the officers stalled by telling him that they had "just one more question.” The officers, however, stated that Crawford did not ask to leave during the interrogation.

Crawford also stated that Bowdich called the prosecutor during the meeting and then told Crawford that the prosecutor had agreed that if he confessed, he would not “do [any] time.” Both Bowdich and Gutierrez denied that they made any promises, express or implied, to Crawford before, during, or after the interrogation. Gutierrez did admit, however, that he told Crawford that it would be better for him if he "g[o]t it out” and told the truth.

. Crawford was one of four individuals indicted in a 26 count indictment pertaining to a series of bank robberies. Of these, only two counts, relating to the Ulrich Street robbery, alleged Crawford’s involvement.

. The district judge entered judgment before the Supreme Court reversed our Knights decision. See United States v. Knights, 534 U.S. 112, 122 S.Ct. 587, 151 L.Ed.2d 497 (2001). The Knights issue is discussed further in Section II.A., infra.

.In addition to the district court’s ruling on the taint from the illegal search, Crawford also appeals the district court's finding that his statements were voluntary and not taken in violation of Miranda. Finally, he appeals a two-level upward sentencing enhancement assessed for physical restraint of a security guard during the robbery. We reach neither the Miranda nor the sentencing issue.

. The district court found Crawford's parole search to be unreasonable, but did so using a doctrinal distinction rejected by the Supreme Court in Knights while this case was up on appeal. In the past, we had approved only probation searches “necessary to the performance of probation duties,” by developing a distinction between "probation searches” and "law enforcement/investigatory searches.” See United States v. Ooley, 116 F.3d 370, 372 (9th Cir.1997), overruled by Knights, 534 U.S. 112, 122 S.Ct. 587, 151 L.Ed.2d 497 (2001); cases cited in id. The distinction turned on the motivations of the personnel primarily responsible for the search. Searches independently instituted by probation officers were deemed valid probation searches, while searches conducted at the request of police trying to evade Fourth Amendment requirements ("subterfuge” searches using probation officers as "stalking horses”) were deemed invalid investigation searches. See, e.g., id.

In Knights, the Supreme Court refused to consider the subjective “actual motivations” of individual officers, and disapproved the distinction between a search for probation purposes and a search for investigation purposes. See Knights, 534 U.S. at 116-18, 121, 122 S.Ct. 587. Instead, Knights required an objective reasonableness analysis, conducted by "examining the totality of the circumstances, with [a] probation search condition being a salient circumstance.” Knights, 534 U.S. at 118, 122 S.Ct. 587.

. This threshold inquiry is often discussed in terms of an individual’s "standing” to invoke Fourth Amendment protections. See United States v. Nerber, 222 F.3d 597, 599 n. 1 (9th Cir.2000). "Standing” is also the label used by both parties in this case and the district court below. In order to avoid confusion with the wholly different doctrine defining Article III "standing,” see United States v. Gamez-Orduno, 235 F.3d 453, 458 n. 7 (9th Cir.2000), the Supreme Court has strongly cautioned against using that particular label in the Fourth Amendment context. See Carter, 525 U.S. at 87-88, 119 S.Ct. 469 ("The Minnesota courts analyzed whether respondents had a legitimate expectation of privacy under the rubric of 'standing' doctrine, an analysis that this Court expressly rejected 20 years ago.... [I]n determining whether a defendant is able to show the violation of his ... Fourth Amendment rights, the ‘definition of those rights is more properly placed within the purview of substantive Fourth Amendment law than within that of standing.'”) (quoting Rakas, 439 U.S. at 140, 99 S.Ct. 421).

. The extent of the connection required to invoke Fourth Amendment protection may depend on the type of privacy interest involved and the nature of the alleged invasion. The expectation of privacy required to protest a physical search, for example, may be different from that required to protest video surveillance. See Nerber, 222 F.3d at 601-03.

. We review the district court’s factual findings for clear error. See Nerber, 222 F.3d at 599, 603. The district court’s implicit factual finding that Crawford had a subjective expectation of privacy in his person and his home is not clearly erroneous.

The government makes much of two facts purportedly showing that, despite the district court’s finding, Crawford had no actual expectation of privacy. The first is Crawford’s assessment of his reaction to Bowdich’s assertion during the parole search that the officials could detain him simply because he was on parole. Crawford stated at trial that "I mean, I just, you know, just took for granted that, you know, I’m on parole, that I don’t have no rights at all, so I didn’t really say nothing.” This resigned submission to the officials’ asserted authority does not invalidate Crawford's expectation of privacy in his own home.

The second fact is Crawford's signature on the "Fourth Waiver,” a compulsory parole condition that purports to permit a search of his home, at any time, with or without a warrant, with or without cause. The government claims that Crawford could not have had any subjective expectation of privacy after signing the condition. As discussed in Section II.A.3, infra, the "Fourth Waiver” does not extend so far, and Crawford’s signature can only be deemed to reduce his expectation of privacy to, at most, the condition's lawful limits. Moreover, although Crawford may have known that he was potentially subject to police search, this court has "rejected the argument that a person lacks a subjective expectation of privacy simply because he ... could have expected the police to intrude on his privacy.” United States v. Sandoval, 200 F.3d 659, 660 (9th Cir.2000). Indeed, consistent with the testimony of Crawford’s parole officer that parolees are sometimes "very surprised” by searches even after signing "Fourth Waivers,” Crawford stated that although he knew what a parole search entailed, he was "shocked” when the officials invaded his home. The natural shock that Crawford felt upon being roused from his own bed by law enforcement officials with weapons drawn is not consistent with the government’s contention that Crawford actually had no expectation of privacy at all.

. We recognize that the California Supreme Court has apparently determined that California law does not recognize as legitimate the expectation of privacy of a parolee subject to a "properly imposed parole search condition,” People v. Reyes, 19 Cal.4th 743, 754, 80 Cal.Rptr.2d 734, 968 P.2d 445 (1998), although it does provide that a search of a parolee’s home may not be arbitrary or capricious. Id. Of course, a search of a California parolee must comport not only with California law, but — as in all jurisdictions within the United States and its territories — with the demands of the federal Constitution as well.

. "Nor do we see a constitutional difference between probation and parole for purposes of the fourth amendment.” United States v. Harper, 928 F.2d 894, 896 n. 1 (9th Cir.1991). Although we recognize that distinctions between probationers and parolees may be relevant in other contexts, because there is no relevant distinction here, we treat discussions regarding probationers as being applicable to parolees, and vice versa.

. The dissent points to isolated language in Latta regarding a “hunch” that appears not to have had the support of a majority of the en banc court and was specifically labeled dicta in a concurring opinion authored by Judge Choy. See Latta, 521 F.2d at 253 (Choy, J., concurring). That concurring opinion correctly noted that the proper standard for a probation search falls somewhat short of probable cause but nevertheless requires specific facts that give rise to a reasonable belief that a search will produce the evidence sought. Moreover, the "hunch” sentence is based entirely on the proposition, first rejected in another concurrence, authored by Judge Wright, see id. (Wright, J., concurring), and subsequently rejected by the Supreme Court in Knights, see supra note 10, that parole officers have a unique knowledge of the parolee that requires that their searches be governed by a different standard than those conducted by police officers.

None of the thirteen judges on the en banc court, however, expressed any disagreement with the conclusions stated in the excerpt quoted in the text, supra, except for the dissenters — who would have imposed an even stricter standard on parole searches (i.e., requiring a warrant and probable cause). See Latta, 521 F.2d at 254-59 (Hufstedler, J., dissenting).

. A state's assertion that it maintains legal custody of its parolees thus has limited impact on Fourth Amendment protections. The dissent’s failure to recognize this shows only that our learned colleague has been bewitched by the California Penal Code and the California Supreme Court at the expense of the supremacy of the federal Constitution. The misguided urge to treat California's assertion of legal custody as dispositive, and on that basis to equate parole with prison, reappears throughout the dissent. See, e.g., post at 724 (citing the custody provisions of CAL. PEN. CODE § 3056); at 724 ("[Ujnder California law, a parolee is in fact in the custody of the Department of Corrections ...."); 3161 (suggesting that the search condition was proper because Crawford was "still in custody”); at 728 (finding that “parolees as a class are different" because they are "still in constructive custody”); at 729 ("Although parole restrictions and conditions strictly speaking are not prison regulations, they are akin to that category.”); at 735 (distinguishing Edmond and Ferguson, in which the Supreme Court refused to approve suspicionless searches, in part because parolees are "in custody and serving out prison sentences”). In the face of unequivocal instruction from the Supreme Court — and our own court sitting en banc— that parole and imprisonment are to be sharply distinguished in this context, the dissent incorrectly clings to a model of parole that the federal courts long ago disavowed.

. The government contends that a parole search becomes reasonable when conducted pursuant to a compulsory "Fourth Waiver” condition, because the condition notifies the parolee of the possibility of suspicionless searches, and therefore nullifies a parolee’s otherwise legitimate expectation of privacy in his own home.. We reject this contention. The parole condition certainly limits a parolee’s expectation of privacy, but only as one "salient circumstance.” The Supreme Court has refused to allow this one factor to override all others. See also note 10, supra.

We therefore reject the argument that a blanket "Fourth Waiver” parole condition eliminates a parolee’s expectation of privacy so as to render any parole search reasonable. We address below, see infra section II.A.3, the different contention that the compulsory parole condition represents a parolee’s consent to be subjected to otherwise unreasonable searches.

. The Knights Court also concluded that the police need not obtain a warrant to search the home of a probationer subject to a “Fourth Waiver”- condition. See Knights, 534 U.S. at 121, 122 S.Ct. 587. The dissent is therefore correct that, after Knights, a search pursuant to a "Fourth Waiver” parole condition is not subject to the ordinary Fourth Amendment warrant requirement. See post at 732.

. In emphasizing the special status of the home, we do not, despite the dissent’s sugges-lion to the contrary, depart from the long-established principle that "the Fourth Amendment protects people, not places.” Katz v. United States, 389 U.S. 347, 351, 88 S.Ct. 507, 19 L.Ed.2d 576 (1967). It is not the building that concerns us, but the unique relationship between an individual and the place in which he resides. To a stranger, a particular structure may be of only passing interest, but to the person who lives there, it is his home — and therein, his privacy interests are subject *712to an especial constitutional protection. We also note that the text of the Fourth Amendment itself singles out the home, by protecting the "right of the people to be secure in their persons, houses, papers, and effects...." U.S. CONST, amend. IV (emphasis added). See supra section II.A.l, and particularly Kyllo, 533 U.S. at 34, 121 S.Ct. 2038 (finding the home to be the “prototypical ... area of protected privacy”); and Johnson, 457 U.S. at 552 n. 13, 102 S.Ct. 2579 ("At least since [1886,] the Court ha[s] acknowledged that the Fourth Amendment accords special protection to the home.”).

. Significantly, both the case that initiated the "special needs” line of cases and the case applying the doctrine to probationers upheld searches predicated on individualized suspicion. See New Jersey v. T.L.O., 469 U.S. 325, 345-47, 105 S.Ct. 733, 83 L.Ed.2d 720 (1985); Griffin, 483 U.S. at 876, 880 & n. 8, 107 S.Ct. 3164.

. In one case decided well before the “special needs” doctrine was announced, the Supreme Court did authorize the search of a residence without individualized suspicion: Camara v. Municipal Court, 387 U.S. 523, 87 S.Ct. 1727, 18 L.Ed.2d 930 (1967), approved residential building code inspections pursuant to a judicial warrant. The Camara Court, however, emphatically stated that such suspi-cionless intrusions were only valid because they were not "aimed .at the discovery of evidence of crime.” Camara, 387 U.S. at 537, 87 S.Ct. 1727.

. Most other courts agree that reasonable suspicion is the appropriate constitutional threshold for parole and probation searches. See, e.g., United States v. Giannetta, 909 F.2d 571, 576 (1st Cir.1990); United States v. Reyes, 283 F.3d 446, 462 (2d Cir.2002), cert. denied, - U.S. -, 123 S.Ct. 106, 154 L.Ed.2d 31 (2002); United States v. Bradley, 571 F.2d 787, 790 & 790 n. 4 (4th Cir.1978) (requiring "articulable grounds” for suspicion); United States v. Scott, 678 F.2d 32, 35 (5th Cir.1982); United States v. Payne, 181 F.3d 781, 786, 788 & 788 n. 5 (6th Cir.1999); Nixon v. State, 18 P.3d 631, 635-36 (Wyo. 2001); Commonwealth v. Williams, 547 Pa. 577, 692 A.2d 1031, 1036 (1997); State v. Beaudry, 282 Mont. 225, 937 P.2d 459, 461-62 (1997); State v. Velasquez, 672 P.2d 1254, 1260 (Utah 1983); State v. Drane, 828 So.2d 107, 112 (La.Ct.App.2002); Fox v. State, 272 Ga. 163, 527 S.E.2d 847, 850 (2000) (reasonable suspicion applies absent consent); State v. Devore, 134 Idaho 344, 2 P.3d 153, 156-57 (App.2000) (reasonable suspicion applies absent consent); Reyes, 19 Cal.4th at 760-61, 80 Cal.Rptr.2d 734, 968 P.2d 445 (Kennard, J., concurring and dissenting) (collecting cases from Hawaii, Massachusetts, Nevada, Oklahoma, and Washington); cf. United States v. Baker, 221 F.3d 438, 448 (3d Cir.2000) (noting reasonable suspicion requirement under Pennsylvania law); United States v. Cantley, 130 F.3d 1371, 1375 (10th Cir.1997) (same for Oklahoma); United States v. Lewis, 71 F.3d 358, 362 (10th Cir.1995) (same for Utah); State v. West, 185 Wis.2d 68, 517 N.W.2d 482, 485 (1994) (same for Wisconsin); Cherry v. State, 302 Ark. 462, 791 S.W.2d 354, 356-57 (1990) (same for Arkansas); State v. Ashley, 459 N.W.2d 828, 830 (S.D.1990) (same for South Dakota); Carswell v. State, 721 N.E.2d 1255, 1262-63 (Ind.Ct.App.1999) (same for Indiana); N.Y. CRIM. PRO. LAW § 410.50 (McKinney 1994); OHIO REV. CODE ANN. § 2967.131(C) (West 2000); OR. REV. STAT. § 137.540(1)(i) (2001); but see Owens v. Kelley, 681 F.2d 1362, 1368 (11th Cir.1982) (rejecting a reasonable suspicion requirement); People v. McCullough, 6 P.3d 774, 781 (Colo.2000) (collecting cases from Nebraska, New Hampshire, and North Dakota).

. See also United States v. Stokes, 292 F.3d 964, 967 (9th Cir.2002) (permitting a probation search because “[t]he standard of reasonable suspicion [embraced by a probation condition] was clearly met”), cert. denied, - U.S. -, 123 S.Ct. 398, 154 L.Ed.2d 321; United States v. Garcia-Cruz, 978 F.2d 537, 541 (9th Cir.1992) (approving a search governed by a California "reasonable suspicion” standard); Toomey v. Bunnell, 898 F.2d 741, 744 (9th Cir.1990) ("We do not approve of general waivers of fourth amendment rights as a condition of parole. However, parole searches may be conducted without a warrant under a reasonableness standard.”) (citations omitted); United States v. Duff, 831 F.2d 176, 179 (9th Cir.1987) (affirming a probation search, in part because the "probation officer had a reasonable suspicion that Duff might be using drugs”).

. We recognize the possibility that some might consider a search premised on a "hunch” — or some other scintilla of information — to be founded on individualized suspicion that is less than "reasonable suspicion” but that is still not "arbitrary or capricious.” Because we are not free to create yet another incremental Fourth Amendment standard— and because we would not do so even if we could — we do not decide whether a search based on a hunch is or is not something more than "arbitrary or capricious.”

. Indeed, we are aware of only one anomalous example of a standard less protective than reasonable suspicion: the minimal suspicion required for pat-downs and incremental canine “sniff tests” at the national border, see United States v. Couch, 688 F.2d 599 (9th Cir.1982); United States v. Taylor, 934 F.2d 218 (9th Cir.1991). We have consistently recognized that "the border search occupies a unique spot in[F]ourth [A]mendment jurisprudence,” United States v. Des Jardins, 747 F.2d 499, 502 (9th Cir.1984), vacated in part on other grounds, 772 F.2d 578 (9th Cir.1985), and that border cases involve an extraordinary set of interests far different from those applicable to searches of a person’s home. See, e.g., Montoya de Hernandez, 473 U.S. at 538, 105 S.Ct. 3304 ("[T]he Fourth Amendment's balance of reasonableness is qualitatively different at the international border than in the interior.”). Indeed, when a border search is "routine,” the government’s particular "concern for the protection of the integrity of the border,” id., permits even sus-picionless searches. We decline to import the unique border search precedents to create a new category of reasonableness that would govern parole searches of the home.

. In the previous paragraph, we determined the degree of individualized suspicion necessary for the search at issue to be reasonable under the Fourth Amendment by applying the "totality of the circumstances" analysis mandated by Knights. Here, we address whether reasonable suspicion — the standard we hold to be appropriate in such cases — actually supported the particular search of Crawford's home, by applying the distinct "totality of the circumstances” test mandated by Arvizu.

. Ironically, the government claims that the defendant conceded the existence of "reasonable suspicion” by conceding that it met the even higher standard of "probable cause,” but this argument is off the mark. The "probable cause” in question involved probable cause to believe that Crawford committed the two-year-old crime, not individualized suspicion that a search might disclose evidence of wrongdoing. The transcript of the district court hearing contains the following dialogue:

DEFENSE COUNSEL: And so it is clear from this record, jumps off the page, that [the parole search] was utilized for the purpose of setting the scene in order to get Mr. Crawford to admit his involvement ... because, up to that point in time, they did not have any evidence sufficient to establish even probable cause to believe that he had committed the robbery in question....
THE COURT: Well, ... we know that they certainly didn't want to disclose their evidence, but they had probable cause.
COUNSEL: I understand that, but I was prevented from going into that.
THE COURT: No, no, no. I just mean on the state of the record—
COUNSEL: All right. I will — I will concede that particular point, but I think that the inference is quite clear that they didn’t believe that they had sufficient testimony

This excerpt reveals only that defense counsel was willing to assume, for purposes of argument, that there may have been "probable cause” to believe that Crawford had been involved in the old robbery. Even that assumption is substantially undermined, however, by Bowdich’s own uncontroverted testimony that the officials did not seek a warrant for Crawford’s arrest at the time of the parole search because "[w]e didn't have anything over him....”

In any event, probable cause to believe that Crawford had participated in a bank robbery two and one half years earlier would not have justified a search of his residence. The individualized suspicion that is required to render a search reasonable is suspicion that the search may uncover evidence relating to a crime. As discussed supra, the record clearly reflects that the officers had no reasonable suspicion that a search of Crawford's home would disclose any such evidence.

. Although federal law governs our holding that the parole search in this case did not comply with the requirements of the Fourth Amendment, the dissent proclaims that we may not properly issue our opinion without California's participation in the case. See post at 736-737. The dissent cites 28 U.S.C. § 2403(b) and FED. R. APP. PRO. 44, both of which apply only to constitutional challenges to state statutes. Id, Here, of course, we do not even construe, much less question the constitutionality of, any state statute. Moreover, California's interests in conducting parole searches, and its legal position regarding the searches, are fully explained in the record on appeal and in decisions of the California Supreme Court, including People v. Reyes, 19 Cal.4th 743, 80 Cal.Rptr.2d 734, 968 P.2d 445 (1998). We see no benefit in now delaying the proceedings sua sponte to ask for further exposition of a position that is amply clear from the materials before us.

. The government does not make any claims with respect to Crawford's particular consent on the morning of July 27, 2000, nor could it, given that Crawford was roused from bed by law enforcement officers who, weapons drawn, informed him that a search was in progress. Rather, the government claims only that Crawford, like all parolees, consented in advance and in blanket fashion to otherwise unlawful searches when he signed the "Fourth Waiver” condition. '

The government cites a string of California cases to support its consent theory. However, we again note that federal law governs the standard for a valid consent to search under the Fourth Amendment. See Ooley, 116 F.3d at 372; Davis, 932 F.2d at 758.

. The dissent harshly criticizes our alleged misrepresentation of the nature of the "Fourth Waiver” condition as a consensual waiver rather than as a mandatory rule governing all parolees. See post 724-125. We intend no such deceit and offer no such erroneous characterization. See supra note 1. We acknowledge that the record does not show that California has chosen to construe its parole condition as a waiver, and we do not intimate that it would do so. Nor does it matter how California characterizes the condition. We fully agree with the dissent that "consent and waiver cannot be used to validate [mandatory] parole conditions,” post at 725; as we explain infra, no construction of the condition as a purported "waiver” or demonstration of "consent” will allow California to accomplish indirectly what it may not accomplish directly.

Rather, we address (and dispose of) the argument that the "Fourth Waiver” condition functions as a blanket consent to search because that argument has been pressed upon us by the federal government in an effort to preserve the California practice. It would in our view be inappropriate simply to ignore the argument or to dismiss it summarily. Nevertheless, it should be plain that the government’s consent theory hardly lies at the heart of our opinion. The theory is merely an additional argument that we must confront once we decide the principal question and hold that reasonable suspicion is required to search the home of a parolee subject to a parole condition. It is only because of the government’s additional assertion that the condition constitutes valid consent eliminating all Fourth Amendment protection, that we are required to hold not only that the Fourth Amendment prevents the state from imposing a suspicionless search upon parolees in the home, but that a state is also precluded from conducting such a search on the ground that the parolee’s signature on a mandatory parole condition waives his Fourth Amendment rights.

. Our decision does not render the "Fourth Waiver” parole condition a nullity. The condition is still a part of the totality of the circumstances that can lead to the conclusion that a reduced privacy interest is warranted in the case of parolees. See Knights, 534 U.S. at 118-19, 122 S.Ct. 587; see also supra sections II.A. 1 and II.A.2.

Nor, contrary to the dissent's fears, does our decision render law enforcement powerless to prevent recidivism. See post at 730-732, 736-737. We hold only that the Fourth Amendment prevents law enforcement officials from invading the home of a parolee subject to a parole condition when they have no reasonable suspicion that a search will uncover evidence of crime.

. Although the government relies heavily on Wellins in its brief, we note that the Wellins *721defendant was permitted to consult with his attorney between the illegal search and his later confession. See Wellins, 654 F.2d at 555. Indeed, his ability to consult with an attorney was the “crucial factor in th[e] case,” and the principal reason no taint was found. Id.

. In 2000, a reasonable officer should have known that searches conducted for ordinary law enforcement purposes rather than purposes related to parole or probation were illegal under more than three decades of consistent Ninth Circuit precedent. See cases cited in Ooley, 116 F.3d at 372. The fact that this distinction was subsequently discarded, see Knights, 534 U.S. at 121, 122 S.Ct. 587, does not render the officer's disregard of existing law any less flagrant.

. Several of our attenuation cases have suggested that "flagrancy” refers to the aggressive and frightening manner in which a search is conducted. See, e.g., George, 883 F.2d at 1416 (holding that "rushing into [the defendant’s] apartment, with guns drawn” weighs in favor of suppression); Delgadillo-Velasquez, 856 F.2d at 1300 (holding that an arrest calculated to cause surprise, fright, and confusion weighs in favor of suppressing a later confession). However, the finding that officers engaged in a "flagrant” illegal search can also refer to the degree to which searching officials intentionally overstepped the law. In both cases, looking to the flagrancy of the search suits the purposes of the taint analysis. The former bears on the likelihood that the evidence in question was flushed out by the illegal search, and therefore the degree to which the evidence would impugn judicial integrity if admitted; the latter bears on the need for increased deterrence. In either circumstance, a finding of flagrancy weighs in favor of suppressing the evidence in question.

. To persuade its readers that the preservation of individual rights is too costly, and that liberty must generally yield to security, the dissent deploys a bogeyman wholly unconnected to the case at hand. This is a familiar tactic. Prosecutors have long used such pariahs, both contemporary, see United States v. Ortiz, 315 F.3d 873, 903 (8th Cir.2002) (Jeffrey Dahmer et al); Culkin v. Purkett, 45 F.3d 1229, 1235 (8th Cir.1995) (John Wayne Gacy); Bittaker v. Enomoto, 587 F.2d 400, 402 n. 2 (9th Cir.1978) (Charles Manson); United States v. Phillips, 476 F.2d 538, 538 (D.C.Cir.1973) (Sirhan Sirhan, James Earl Ray, Richard Speck, Jack Ruby); United States v. Gross, 451 F.2d 1355, 1359 (7th Cir.1971) (Lee Harvey Oswald); and timeless, see Johnston v. Luebbers, 288 F.3d 1048, 1061 (8th Cir.2002) (Heaney, J., dissenting) (Satan); Martin v. Parker, 11 F.3d 613, 615-16 (6th Cir.1993) (Adolf Hitler); United States v. Hitow, 889 F.2d 1573, 1579 (6th Cir.1989) (Al Capone); United States v. Steinkoetter, 633 F.2d 719, 720 (6th Cir.1980) (Pontius Pilate and Judas Iscariot). With no fewer than five references to Richard Allen Davis — who is, of course, neither a party nor an unindicted co-conspirator — the dissent joins those seeking to capitalize on the notoriety of evildoers.

We have confronted such tactics before. As we stated in Bittaker:
The state mentions several times that one of its prisoners who may benefit from the [] decision is Charles Manson. We do not encourage this type of advocacy. A federal court must make its decisions in accord with the Constitution and the laws, without regard to the notoriety of parties or nonparties.

Bittaker, 587 F.2d at 402 n. 2. We agree with the Bittaker panel's admonition, and accordingly consider only constitutional precedent, legal rules, and the circumstances of the parties actually involved. We therefore render our decision here on the facts of the case before us and the principles embodied in the Bill of Rights.