dissenting:

I dissent. This is a particularly intriguing case because an anonymous tip that was false — and patently manufactured out of whole cloth in at least one crucial respect — resulted in the seizure of evidence that led to the defendant’s conviction for an entirely unrelated offense. Because the tip was false, the police could not obtain corroboration of any of its details, other than of wholly neutral facts that failed to suggest the existence of any criminal activity, planned or otherwise. 1 Yet, the majority holds that the police had founded suspicion sufficient to justify a number of police officers’ drawing their weapons, stopping an individual at gunpoint, immobilizing him, and then frisking and questioning him. By finding that the wholly innocuous facts observed by the police provided sufficient corroboration for an anonymous tip, and justified the use of force that ensued, the majority lowers even further the constitutional barriers that separate a free society from a police state.

The district court framed the central question in this case as “whether the police can lawfully stop a vehicle or person with their guns drawn, acting solely on the basis of an anonymous phone call properly identifying where the person could be found, without any supporting evidence that a crime is being, has been, or is about to be committed.” In actuality, this is not one question but several, with which this court has struggled for years. Among them are two that we must address today: Under what circumstances may police rely on an anonymous tip to form reasonable suspicion for an investigatory stop? 2 What degree of corroboration will establish the reliability of an anonymous tip?

Our struggle with these issues has been complicated by the fact that in many respects the answers are fact-specific and, thus, resistant to the development or application of general principles. Still, our prior decisions and those of the Supreme Court have frequently noted the intrinsic unreliability of anonymous tips. It follows that law enforcement officials must corroborate such tips in significant detail before their reliability can be properly established. Because I do not believe that the Santa Ana police corroborated the anonymous tip at issue here in any significant detail, I conclude that they lacked reasonable suspicion to stop Alvarez.

Under the principles of Terry v. Ohio, 3 “ ‘[a]n investigatory stop must be justified by some objective manifestation that the person stopped is, or is about to be, engaged in criminal activity.’ ” United States v. Thomas, 863 F.2d 622, 625 (9th Cir.1988) (quoting United States v. Cortez, 449 U.S. 411, 417 , 101 S.Ct. 690, 694-95 , 66 L.Ed.2d 621 (1981)). The lawfulness of the stop depends upon whether the totality of the circumstances, as they existed “at the time the officer initiate[d] the stop,” Thomas, 863 F.2d at 625 , constitutes “a particularized and objective basis for suspecting the particular person stopped of criminal activity.” Cortez, 449 U.S. at 417-18 , 101 S.Ct. at 694-95 . Under this totality-of-the-cireumstances test, the anonymous tip upon which the police relied was insufficient to provide reasonable suspicion to stop Alvarez.

*841 Both the Supreme Court and this court have indicated that anonymous telephone tips are the least reliable source of information. See Illinois v. Gates, 462 U.S. 213, 233-34, 237-38 , 103 S.Ct. 2317, 2331-32 , 76 L.Ed.2d 527 (1983); Adams v. Williams, 407 U.S. 143, 146-47 , 92 S.Ct. 1921, 1923-24 , 32 L.Ed.2d 612 (1972); United States v. Sierra-Hernandez, 581 F.2d 760, 763 (9th Cir.) (per Kennedy, J.), cert. denied, 439 U.S. 936 , 99 S.Ct. 333 , 58 L.Ed.2d 333 (1978). However, the Supreme Court has held that an anonymous tip may serve as the basis for probable cause where, in light of the totality of the circumstances — including the extent of detail in the tip and police success in corroborating that detail — the police can establish its reliability and the veracity of the anonymous informant. In Gates , the informant accurately predicted the conduct of the defendants, describing in detail their unusual travel plans and the modus operandi of the criminal activity they would be engaged in. Through surveillance, the police confirmed that the defendants travelled to a location where the drug trade was particularly active. They confirmed that the defendants followed a travel route frequently employed by drug couriers. And they confirmed the suspicious travel arrangements predicted by the anonymous informant. The confirmation of those details established the reliability and personal knowledge of the informant and provided probable cause for the police to search the defendants’ property for narcotics.

While there are many distinctions between Gates and the present case, 4 the key distinction for our purposes is that here the police could not verify any significant detail of the anonymous tip, and certainly not the allegation that Alvarez planned to rob a bank or that he was carrying explosives. While under police surveillance, Alvarez engaged in no behavior that would suggest that he planned a bank robbery and made no moves that would indicate he was armed. He was observed parked in a bank parking lot for five minutes, as if waiting for someone — possibly the anonymous tipster. That is the only conduct the police observed. 5 He made no move to enter the bank, and while parked in the lot he did nothing to indicate that a bank robbery was about to commence or that he was awaiting further events before proceeding with criminal conduct. After several minutes, he drove out of the lot without engaging in any suspicious activity whatsoever, 6 *842 and when police stopped him shortly thereafter, they had received no information or reports indicating that a robbery had been attempted. In short, everything about Alvarez’s conduct suggested that he had not robbed a bank, was not engaged in a bank robbery, and did not intend to rob a bank. The allegation that Alvarez was carrying explosives also proved ultimately to be untrue. 7

Thus, the police could not corroborate any significant detail about Alvarez’s conduct before stopping him; their anonymous telephone informant’s reliability extended only to his ability to offer a description of an individual sitting in his car in a public place. This falls far short of the corroboration deemed sufficient in Gates , which the Supreme Court described as a “range of details relating not just to easily obtained facts and conditions existing at the time of the tip, but to future actions of third parties ordinarily not easily predicted.” 462 U.S. at 246 , 103 S.Ct. at 2336 . Surely, even under the lesser Terry standards, reasonable suspicion must be founded upon more than an accurate description of an individual’s present location and appearance, especially where the allegations of his future criminal conduct are called into question by surveillance. In this case, the anonymity of the informant and the inability of the police to corroborate any significant details of his tip preclude any finding that the police had any particularized or objective basis for suspecting Alvarez of criminal activity. Cortez, 449 U.S. at 417-18 , 101 S.Ct. at 694-95 . Because the police had no reasonable suspicion that Alvarez was engaged in or planning to engage in criminal activity, the stop and subsequent searches of Alvarez and his car were illegal under Terry.

The majority’s discussion of the tip’s corroboration — without which the tip is admittedly unreliable — is deficient in three respects. First, the majority asserts that “the officers verified each of the details of the tip.” In fact, the officers verified only the fact that a man meeting the anonymous caller’s description was indeed parked in, of all places, a parking lot; the significant details of the tip were actually called into question by police observation.

Second, the majority asserts that the facts observed by the police “were consistent with the actions of a would-be bank robber who decided to take a break until the coast was clear.” Yet even if the facts were as the majority claims — and they are not 8 — Terry demands a higher standard. Reasonable suspicion does not exist whenever a citizen’s actions are merely consistent with the actions of a would-be criminal. The majority correctly points out that police need not directly observe criminal activity, but it ignores an important corollary emphasized by the Supreme Court in Gates : “In making a determination of probable cause the relevant inquiry is not whether particular conduct is ‘innocent’ or ‘guilty,’ but the degree of suspicion that attaches to particular types of noncriminal acts.” 462 U.S. at 243 -44 n. 13, 103 S.Ct. at 2335 n. 13 (emphasis added). Clearly, behavior does not become suspicious simply because the police are told about it. However, in deciding that the conduct observed by the Santa Ana police adequately corroborated the anonymous tip, the majority fails to indicate a single detail or combination of details that would be suspicious in the absence of a tip. This circular manner of analyzing the tip makes it essentially self-corroborating, even though the information supplied by the caller here could *843 have been supplied by anyone within sight of Alvarez at the time, or anyone who had told Alvarez to meet him at the lot. 9

Third, the majority sums all of this up by stating, “The anonymous tip, corroboration of the details of the tip, and the pre-stop observations of the police officers gave rise to reasonable articulable suspicion to make an investigatory stop.” Yet the anonymous tip, which contained no indicia of inherent reliability, was entitled to no weight unless corroborated in significant detail, and police surveillance tended to contradict more significant details than it corroborated.

No prior decision of this court has ever credited an anonymous tip with so little corroboration. The majority cites United States v. Rodriguez, 869 F.2d 479, 483 (9th Cir.1989), for the proposition that “seemingly innocent conduct” can provide reasonable suspicion. However, the “seemingly innocent conduct” observed by police in Rodriguez involved a number of suspicious details, including large stacks of currency, frequent trips to the airport, and “evasive driving.” 869 F.2d at 483 . Moreover, there were actually two anonymous tips in Rodriguez , one of which had already been corroborated when a consensual search of an apartment turned up $140,000 to which a narcotics dog later alerted; and these two tips were supplemented by a third tip from a known informant, as well as information obtained from other narcotics officers. Id. at 482-83 . The majority’s citation of United States v. Thomas, 863 F.2d 622 (9th Cir.1988), and United States v. Sutton, 794 F.2d 1415 (9th Cir.1986), is even more difficult to understand: Neither Thomas nor Sutton involved an anonymous tip.

Nor does the Fifth Circuit’s decision in another case, also entitled United States v. Rodriguez, 835 F.2d 1090 (5th Cir.1988), support the result the majority reaches here. In that case, the investigating, customs agent was able to verify that the anonymous informant had accurately stated the license number of a truck that would be lightly loaded with watermelons at a particular fruit company, with room left in the truck for contraband. The agent followed the truck from the fruit company to a small fruit stand, where he observed boxes sometimes used to transport marijuana being loaded onto the truck. Because this degree of corroboration far surpasses anything offered by the Government here, the majority’s citation of the Fifth Circuit’s Rodriguez opinion does not provide support for its decision.

The majority suggests that the law of the District of Columbia Circuit would justify finding reasonable suspicion based solely on police confirmation that the anonymous tip correctly described Alvarez and his car and accurately identified their location. United States v. McClinnhan, 660 F.2d 500 (1981); United States v. White, 648 F.2d 29 , cert. denied, 454 U.S. 924 , 102 S.Ct. 424 , 70 L.Ed.2d 233, 235 (1981). We are of course entitled to consider the D.C. Circuit’s decisions as persuasive authority, and to follow them if we deem them correct; but here the majority provides no justification for the adoption of that circuit’s law over any other principle. Cf. White, 648 F.2d at 43 (question whether an anonymous tip corroborated only by observations of innocent details justifies a Terry stop is a live and disputed one in circuit and state courts); Jernigan v. Louisiana, 446 U.S. 958 , 100 S.Ct. 2930 , 64 L.Ed.2d 816 (1980) (White, J., joined by Brennan and Marshall, JJ., dissenting from a denial of certiorari and calling for resolution of whether corroboration of innocent details is sufficient for a Terry stop); Alabama v. White, — U.S. -, -, 110 S.Ct. 834, 861 , 107 L.Ed.2d 830, 946 (1990) (granting certiorari on the issue raised ten years *844 earlier in Jernigan). Once White and McClinnhan are examined with care, the majority's endorsement of them becomes extremely troublesome.

First, those decisions appear to be in some conflict with the way this court has handled the corroboration of tips in general and anonymous tips in particular. We have emphasized — as did the Supreme Court in Gates — “that when examining information provided by an informant, the two-prong Aguilar-Spinelli test, Spinelli v. United States, [ 393 U.S. 410 , 89 S.Ct. 584 , 21 L.Ed.2d 637 (1969)]; Aguilar v. Texas, [ 378 U.S. 108 , 84 S.Ct. 1509 , 12 L.Ed.2d 723 (1964) ], is ‘highly relevant’ in determining the value of the information.” United States v. Miller, 753 F.2d 1475, 1479 (9th Cir.1985) (citing Gates ; parallel citations omitted); see also United States v. Angulo-Lopez, 791 F.2d 1394, 1396-97 (9th Cir.1986). While Gates holds that “a weakness in either the ‘veracity’ or ‘basis of knowledge’ prong is not fatal,” Angulo-Lopez, 791 F.2d at 1396 , the tips held reliable in White and McClinnhan — and in the majority’s decision today — suffer from more than mere “weakness.” All three tips were in the form of anonymous telephone calls to the police which disclosed absolutely nothing to show either the informant’s veracity or the basis of the informant’s knowledge. 10 Our prior decisions deem this “highly relevant,” yet the D.C. Circuit is apparently willing to credit anonymous tips fully, “[e]ven without these in-dices of reliability,” White, 648 F.2d at 44 . 11

Second, it is difficult to see how the almost nonexistent level of corroboration apparently sufficient under White and McClinnhan can possibly be consistent with the requirement that an investigative stop be justified by “some objective manifestation that the person stopped is, or is about to be, engaged in criminal activity.” Cortez, 449 U.S. at 417 , 101 S.Ct. at 694-95 ; Thomas, 863 F.2d at 625 . By appealing to the principles announced in White and McClinnhan , the majority today approves a seizure based only on an objective manifestation that Alvarez was tall, dark, Mexican, and parked in a bank parking lot. I believe that our prior decisions clearly require more, and I see no justification for adopting the lower level of scrutiny that apparently prevails in the D.C. Circuit.

Finally, I believe McClinnhan and White would be questionable decisions even if they did not contravene settled principles in this circuit. As the McClinnhan'court conceded, “[I]t is possible for anyone with a grudge to fabricate a tip whose neutral details, such as clothing or location, would provide the corroboration required by the White decision for a stop.” 660 F.2d at 502 . This by itself permits a significant intrusion on the liberty of those who may be the subject of such fabrications. Apparently, the D.C. Circuit is willing to pay this price, or rather to have it paid by those whom the police are most likely to stop. However, the opinions in White and McClinnhan fail to note that if the anonymous tipster also fabricates an allegation that the “suspect” is armed or dangerous, *845 then police officers are entitled to converge upon the “suspect” with guns drawn, issuing orders that would cause any law-abiding citizen considerable distress. Misunderstandings are to be expected in such situations, and the lives of innocent citizens are thereby placed in jeopardy, all without requiring the police to identify anything that is genuinely suspicious about the “suspect’s” conduct.

In summary, I do not find White and McClinnhan persuasive, and I find nothing in the majority’s discussion of them to allay my concerns about the extent to which they conflict with our prior decisions. Nor do I believe that the majority has otherwise justified its reversal of the district court’s suppression order. I would therefore affirm that order on the ground that the Santa Ana police lacked a reasonable suspicion when they stopped Alvarez.

.Another unusual aspect of this case is that the "anonymous” caller eventually became known, and in fact testified at the suppression hearing. The district court characterized the informant’s testimony as follows: "Soler’s testimony that he called the police out of a sense of civic responsibility was not credible. Nothing in his testimony revealed any reason for him to believe that Alvarez was about to rob the bank.” Moreover, Soler's testimony makes it absolutely clear that he had no information whatsoever suggesting that Alvarez had any explosives. Soler's refusal to reveal the basis for his allegations to the police dispatcher prevented the police from learning that the tip was, at least in most significant respects, Soler’s invention — a compelling argument for the importance of ascertaining the basis of an informant’s knowledge before relying on such an intrinsically unreliable source. See page 844 infra.

. The Supreme Court recently granted certiorari in order to address this issue in Alabama v. White, — U.S. -, -, 110 S.Ct. 834, 861 , 107 L.Ed.2d 830, 946 (1990) (No. 89-789).

. 392 U.S. 1 , 88 S.Ct. 1868 , 20 L.Ed.2d 889 (1968).

. Gates is a probable cause case and not a reasonable suspicion case. The degree of corroboration of details required for reasonable suspicion is undoubtedly less than that required in Gates . But as the discussion in the text will make clear, the tip in this case involved almost no details and could be corroborated only in the slightest degree. And the details that could be corroborated did not suggest that the suspect was engaged in any criminal activity.

. The Government claims that Alvarez was parked in an unusual manner, as if planning a getaway. However, the record shows only that Alvarez’s car was “backed into a parking place, facing the bank.” The record does not show the distance between the car and the entrance to the bank; it does not show the distance between the car and the exit from the parking lot; it does not show whether the driver’s side or the passenger’s side of the car was closer to the bank; it does not show whether the position of Alvarez’s car in the parking space might as easily have resulted from Alvarez pulling forward into his space from the space behind it, as people often do; it does not show whether some or even all of the other cars in the parking lot were parked in a manner similar to Alvarez's. In short, the record does not support the Government’s claim.

The mere fact that Alvarez was parked with the front end of his car facing out of the parking space and toward the bank is not by itself unusual, let alone suggestive of a planned getaway. In fact, an informal survey of the judges’ parking lot in the federal courthouse in Los Angeles on a typical Friday afternoon revealed more than one car which, on the Government’s theory, was parked in a manner that suggested that the driver was planning a fast getaway after perpetrating some form of mischief.

.The Government vaguely suggests some causal relationship between a police car’s arrival and Alvarez’s departure, and the majority seems uncritically to accept this version of events. However, the record fails to establish that Alvarez’s exit was precipitated by the appearance of the police car, or even that Alvarez saw it drive by. While the record is not clear on this point, the Government’s argument below was merely that such a connection was "quite possible, maybe even probable." This possibility was never supported by anything more than counsel’s suggestion, made only after Judge Letts had tentatively ruled against the Government. Indeed, the *842 written declarations of the officers on the scene all suggest that Alvarez may have waited as long as five minutes after the appearance of the vehicle before leaving the parking lot. The district court found that "[¡Instead of getting out of his car and approaching the bank or otherwise acting in a manner consistent with an intent to rob the bank, Alvarez simply left the bank parking lot" (emphasis added). Under these circumstances, any inference that Alvarez was fleeing from police surveillance is entirely unwarranted, and thus, provides no basis for placing any faith in the anonymous allegation that Alvarez was planning to rob the bank.

. Although the absence of explosives was confirmed only after the stop occurred, police knew before they stopped Alvarez that he had not engaged in a bank robbery.

. See notes 5 and 6 supra.

. We need not decide what the result would have been if Alvarez had actually sought to enter the bank, and I express no opinion on that hypothetical question. However, any such case would at least involve a prediction of future conduct — i.e., some conduct on Alvarez’s part that was predicted by the informant and was not observable by the general public at the time of the informant’s telephone call. In this case, the anonymous tip described only the current state of affairs in a public place. The future conduct, Alvarez’s departure from the parking lot, tended to undermine rather than support the reliability of the tip.

. Some anonymous tips have a higher degree of reliability because the details supplied by the informant are so specific, or are of a nature (e.g., a prediction of future conduct) that they establish the personal knowledge of the informant. Gates, 462 U.S. at 234 , 103 S.Ct. at 2330 . The kinds of tips and corroboration of details upon which the D.C. Circuit would apparently allow reasonable suspicion to be founded are of a far lesser order.

. Arguably, not even the D.C. Circuit goes this far. The phrase quoted in the text is in some conflict with the White panel's narrower statement, concluding "that an anonymous tip about an ongoing transaction, detailed as to time and place, including a specific description of one of the participants and their vehicles as well as their modus operandi, and verified by the officers through surveillance in all details except for the actual possession or exchange of narcotics provides a sufficient basis for a legitimate Terry stop to question the occupants as to their identity and visually check inside the car.” 648 F.2d at 43 (emphasis added). Likewise, the McClinn-han court noted that it was considering "an anonymous tip that, while lacking facial indicia of reliability, was corroborated in every significant detail by [the police officers’] pre-stop surveillance." 660 F.2d at 502 (emphasis added). To the extent that White and McClinnhan require corroboration “in all details” or “in every significant detail,” the majority's more expansive decision today is truly unprecedented.