(with whom
WALTERS, Chief Judge, joins), specially concurring.We agree with our colleague that the judgment of conviction should be affirmed. We concur fully in Parts II and III of his lead opinion; but we concur only in the result as to Part I, which discusses the search resulting in seizure of money from Cook’s pocket. In our view this search was incident to a valid arrest. Moreover, we resist the suggestion that the doctrine of “inevitable discovery” should be applied in this case. Each point is examined in turn below.
Search Incident to Arrest
The lead opinion suggests that the search was not incident to a valid arrest because reasonable cause for such an ar*220rest did not exist when the search occurred. We disagree.
Idaho Code § 19-603 provides that a police officer may arrest a person without a warrant “[w]hen a felony has in fact been committed and he has reasonable cause for believing the person arrested to have committed it.” Construing this statute and its predecessors, the Idaho Supreme Court has said that reasonable cause exists when the arresting officer possesses information which would lead an ordinarily prudent and cautious officer to believe, or to entertain an honest and strong suspicion, that the person arrested has committed a felony. E.g., State v. PoLson, 81 Idaho 147, 339 P.2d 510 (1959); State v. Autheman, 47 Idaho 328, 274 P. 805 (1929). In reviewing a police officer’s determination of reasonable cause in the field, we have been admonished to take into account the factual and practical considerations of everyday life on which reasonable and prudent persons, not legal technicians, act. Draper v. United States, 358 U.S. 307, 79 S.Ct. 329, 3 L.Ed.2d 327 (1959); Brinegar v. United States, 338 U.S. 160, 69 S.Ct. 1302, 93 L.Ed. 1879 (1949).
Thus, our standard of appellate review should be whether, at the time of the arrest, the police had possession of information which, viewed in light of practical considerations of everyday life, would lead an ordinarily prudent and cautious officer to believe — or to entertain an honest and strong suspicion — that appellant Cook had committed a felony. The critical variables in applying this test are the point in time when the arrest occurred and the quantum of information possessed by the police at that point. Although the state’s brief suggests that information acquired after the arrest may be used to support a finding of probable cause, the overwhelming weight of authority holds that the sufficiency of information must be tested at the time of the arrest itself. E.g., People v. Talley, 65 Cal.2d 830, 56 Cal.Rptr. 492, 423 P.2d 564 (1967); People v. Snelling, 174 Colo. 397, 484 P.2d 784 (Colo.1971); State v. Isham, 1 Wash.App. 415, 461 P.2d 569 (1969).
The lead opinion does not specify when an arrest occurred in this case. It says only that Cook actually may have been under arrest before he was “formally” arrested, handcuffed and led away to a police car. The issue focuses upon the relationship between an investigative stop and an arrest. An investigative stop will ripen into an arrest, which must then be supported by reasonable cause, if the detention becomes overly intrusive. People v. Tooker, 198 Colo. 496, 601 P.2d 1388 (1979); State v. Tourtillott, 43 Or.App. 5, 602 P.2d 659 (1979), affd 289 Or. 845, 618 P.2d 423 (1979), cert. denied, 451 U.S. 972, 101 S.Ct. 2051, 68 L.Ed.2d 352 (1980). Accordingly, an arrest is deemed to have occurred, and must be tested against the requirement of reasonable cause, when the actions of the police go beyond those authorized for an investigative stop.
In this case, appellant Cook concedes that the police had adequate grounds to conduct an investigative stop. See, e.g., State v. Clovis, 127 Ariz. 75, 618 P.2d 245 (App.1980). When the stop occurred, Cook attempted to walk away from the scene. The police, with weapons drawn, ordered him’to halt and to return. Such detention by the police officer was within the permissible scope of the investigative stop. E.g., State v. Taras, 19 Ariz. App. 7, 504 P.2d 548 (1972); State v. Gardner, 28 Wash.App. 721, 626 P.2d 56 (1981). Any investigative stop necessarily involves a brief period of detention. A suspect cannot defeat the purpose of a stop simply by walking away from it. Neither is an investigative stop necessarily converted to an arrest if the police draw weapons. The police are entitled to take reasonable precautions for their own safety and to make a reasonable show of force necessary to effectuate the stop. See generally 1 W. RINGEL, SEARCHES & SEIZURES, ARRESTS AND CONFESSIONS § 13.5(c) (2d ed. 1983). Here, there is no claim that the police acted unreasonably by drawing weapons to prevent an armed robbery suspect from leaving the scene.
*221However, when one of the police officers subsequently reached into Cook’s pocket, rather than conducting a “pat-down” search for weapons, the scope of a permissible investigative stop was exceeded. State v. Post, 98 Idaho 834, 573 P.2d 153 (1978), overruled on other grounds, State v. Bottelson, 102 Idaho 90, 625 P.2d 1093 (1981). See also State v. Melear, 630 P.2d 619 (Hawaii 1981). Therefore, in this case, the time of the arrest — for the purpose of testing the existence of reasonable cause — was coincident with the search itself.
The next question is what information the police possessed at that point. Before listing each item, it should be noted that the arresting officer’s information came in part from the police radio. In such circumstances, reasonable cause must be tested upon the full information in police possession which caused the radio messages to be sent, not merely upon the messages themselves. E.g., State v. Pokini, 45 Haw. 295, 367 P.2d 499 (1961). See generally W. RINGEL, supra, § 4.3(d).
The police were informed of an armed robbery at the Circle K store. They also were aware that, just before the robbery, a man fitting Cook’s general description had been seen walking past another store, then entering the passenger’s side of a waiting automobile. Despite the warm summer weather, the man had been wearing a heavy coat and what appeared at that time to be a stocking cap. The automobile was “tailed” to a location very near the Circle K store. The same man and automobile were observed minutes later and, when stopped, the man attempted to leave the scene. When ordered to return, the man was seen to have what then appeared to be a ski mask protruding from his back pocket.
We believe these facts, considered together and viewed in light of practical considerations of everyday life, would lead an ordinarily prudent and cautious police officer to believe, or to entertain an honest and strong suspicion, that Cook had participated in the reported robbery. We acknowledge that probable cause must consist of more than mere suspicion, especially generalized suspicion. See cases collected in 5 AM.JUR.2d Arrest, § 45 (1962). However, in this case the police did not have a general suspicion; rather, they had a particularized suspicion based upon Cook’s proximity to the crime, suspicious behavior and inappropriate clothing. This particularized suspicion was coupled with Cook’s attempt to leave the scene of the investigative stop. It is well settled that such an attempt is an important factor, though not in itself dispositive, in determining reasonable cause for arrest. E.g., State v. Melear, supra; State v. Baxter, 68 Wash.2d 416, 413 P.2d 638 (1966). Particularized suspicion, when combined with an attempt to avoid an investigative stop, may constitute adequate grounds for a finding of reasonable cause. See State v. Elliott, 626 P.2d 423 (Utah 1981). Finally, after Cook’s attempted departure was halted, the police observed the ski mask.
We conclude that the arrest of Cook was supported by reasonable cause and that the search of Cook’s pocket was valid. We acknowledge that the question is a close one, and we respect our colleague’s differing view. However, by purporting to invalidate the arrest at the time of the search, the lead opinion departs from practical considerations of everyday life. It plays the role of legal technician condemned by the Supreme Court in Draper and Brinegar.
The Doctrine of Inevitable Discovery
Because the lead opinion treats the search as invalid, it is forced to seek another basis for upholding admission of the money in evidence at trial. The opinion settles upon the doctrine of inevitable discovery — a doctrine new to Idaho law. However, in this case the state at no time urged such a doctrine, nor did the district court make any findings of fact to support application of the doctrine. For these reasons, and those set forth below, we would not apply the doctrine here.
*222A
Inevitable discovery is a device that some courts have employed to narrow the exclusionary rule. In order to appreciate the significance of this device, it is necessary to understand the exclusionary rule itself and its existing exceptions. The exclusionary rule bars the use at trial of evidence illegally obtained. It applies to indirect as well as direct products of unlawful police conduct. Silverthorne Lumber Co. v. United States, 251 U.S. 385, 40 S.Ct. 182, 64 L.Ed. 319 (1920). To use the “timeworn metaphor” of the “poisonous tree,” Harrison v. United States, 392 U.S. 219, 222, 88 S.Ct. 2008, 2010, 20 L.Ed.2d 1047 (1968), both the evidence unlawfully seized and the “fruit of the poisonous tree” — that is, derivative evidence to which the police are led by the tainted evidence — are barred under the exclusionary rule. However, as Justice Holmes noted in Silverthorne, not all evidence is inadmissible simply because it has some connection to other, tainted evidence:
[Application of the exclusionary rule] does not mean that the facts thus obtained become sacred and inaccessible. If knowledge of them is gained from an independent source they may be proved like any others, but the knowledge gained by the Government’s own wrong cannot be used by it____ [251 U.S. at 392, 40 S.Ct. at 183.]
Consequently, the exclusionary rule does not apply to evidence obtained from an “independent source,” even though the evidence also may have been the product or the “poisoned fruit” of an illegal search. In Nardone v. United States, 308 U.S. 338, 60 S.Ct. 266, 84 L.Ed. 307 (1939), the Supreme Court established a second exception to the exclusionary rule. The Court said that the rule should not apply to evidence that has “become so attenuated as to dissipate the taint.” Id. at 341, 60 S.Ct. at 267.
Inevitable discovery occupies an uncertain place in this mosaic. Although called a third exception to the exclusionary rule, in reality it extends the concept of an independent source to cover hypothetical— rather than actual — sources of evidence. Thus, under the inevitable discovery doctrine, evidence illegally obtained may be admitted even though it was not obtained through an actual independent source, so long as the court is satisfied that another, hypothetical source inevitably would have yielded the evidence.
This broadening of the independent source doctrine is said to have begun in Wong Sun v. United States, 371 U.S. 471, 83 S.Ct. 407, 9 L.Ed.2d 441 (1963). In that case, federal narcotic agents entered a building in a manner later held to be unlawful. They arrested an occupant of the building, obtaining from him information which led to discovery of drugs and other evidence incriminating the occupant himself as well as several others, including a person named Wong Sun. The Supreme Court was asked to decide whether this derivative evidence could be used at trial. As to the building occupant, the Supreme Court said:
The prosecutor candidly told the trial court that “we wouldn’t have found those drugs except that [the building occupant] helped us to.” Hence this is not the case envisioned by this Court where the exclusionary rule has no application because the Government learned of the evidence “from an independent source,” [citing Silverthorne ] nor is this a case in which the connection between the lawless conduct of the police and the discovery of the challenged evidence has “become so attenuated as to dissipate the taint.” [Citing Nardone.] We need not hold that all evidence is “fruit of the poisonous tree” simply because it would not have come to light but for the illegal actions of the police. Rather the more apt question in such a case is “whether, granting establishment of the primary illegality, the evidence to which instant objection is made has been come at by exploitation of that illegality or instead by means sufficiently distinguishable to be purged of the primary taint.” [J.] Maguire, Evidence of Guilt, 221 (1959). We think it clear that the narcotics were *223come at by the exploitation of that illegality” and hence that they may not be used against [the building occupant].
Id. at 487-88, 83 S.Ct. at 417.
Turning to the man named Wong Sun, the Court noted that he later had made an oral confession, implicating himself in drug transactions. The Court said:
Wong Sun’s unsigned confession was not the fruit of that [illegal arrest of the building occupant], and was therefore properly admitted at trial. On the evidence that Wong Sun had been released on his own recognizance after a lawful arraignment, and had returned voluntarily several days later to make the statement, we hold that the connection between the arrest and the statement had “become so attenuated as to dissipate the taint.” [Citing Nardone.]
Id. at 491, 83 S.Ct. at 419.
Thus, the Supreme Court actually decided the Wong Sun case upon the two well-established exceptions to the exclusionary rule — the “independent source” and “attenuation.” The Court said nothing about inevitable discovery. However, a number of courts and commentators began to read more into the Wong Sun opinion. One important article suggested that the “fruit of the poisonous tree” could be unpoisoned by a government showing that the police “would have acquired the evidence in any event.” R. Maguire, How to Unpoison the Fruit — the Fourth Amendment and the Exclusionary Rule, 55 J.CRIM.LAW, CRIMINOLOGY, & POLICE SCI. 307, 315 (1964) (emphasis original). That same year, a federal appellate court held that evidence obtained during an unlawful search of a criminal defendant’s apartment, which led to information concerning the location of a homicide victim’s body, was not barred under the exclusionary rule because the body would have been discovered in any event. Killough v. United States, 336 F.2d 929 (D.C.Cir.1964).
However, another federal appellate court, in a decision immediately preceding Wong Sun, expressed a different view:
[A] showing that the government had sufficient independent information available so that in the normal course of events it might have discovered the questioned evidence without an illegal search cannot excuse the illegality or cure tainted matter____ The test must
be one of actualities, not possibilities. United States v. Paroutian, 299 F.2d 486, 489 (2d Cir.1962) (emphasis added).
Despite its uncertain ancestry, the nascent doctrine of inevitable discovery continued to attract adherents. In People v. Fitzpatrick, 32 N.Y.2d 499, 346 N.Y.S.2d 793, 300 N.E.2d 139 (N.Y.1973), cert. denied, 414 U.S. 1050, 94 S.Ct. 554, 38 L.Ed.2d 338 (1973), the New York Court of Appeals adopted the emerging doctrine. When the United States Supreme Court denied certiorari in that case, Justice White dissented. He stated that he would have granted certiorari because
it is a significant constitutional question whether the “independent source” exception to inadmissibility of fruits [citing Wong Sun ] encompasses a hypothetical as well as an actual independent source.
Id. at 1050, 94 S.Ct. 554, 38 L.Ed.2d 338. Thus, the Supreme Court neither endorsed nor eschewed the inevitable discovery doctrine. By denying certiorari in Fitzpatrick, the Court simply allowed lower federal appellate courts and state appellate courts to continue their efforts to develop the doctrine.
More recently, although the Supreme Court still has not adopted the doctrine, it has recognized in dictum a form of inevitable discovery relating to dead bodies. In Brewer v. Williams, 430 U.S. 387, 406 n. 12, 97 S.Ct. 1232,1243 n. 12, 51 L.Ed.2d 424 (1977), the Court, citing Killough v. United States, supra, said that even if an incriminating statement improperly obtained from a criminal defendant helped police to locate a homicide victim’s body, evidence of the body itself and the location where it was found “might well be admissible on the theory that the body would have been discovered in any event.” Speaking more generally in United States v. Crews, 445 *224U.S. 463, 470, 100 S.Ct. 1244, 1249, 63 L.Ed.2d 537 (1980), the Supreme Court mentioned “three commonly advanced exceptions to the exclusionary rule” — the independent source, attenuation, and inevitable discovery. However, the Court did not explain or apply the inevitable discovery doctrine in that case.
The Supreme Court’s current view of inevitable discovery soon may be clarified in two pending cases where certiorari has been granted: Williams v. Nix, 700 F.2d 1164 (8th Cir.1983) (a reprise of Brewer v. Williams, supra), and People v. Quarles, 58 N.Y.2d 664, 458 N.Y.S.2d 520, 444 N.E.2d 984 (1982). The cases were scheduled for argument in January, 1984. Until those decisions are issued, all courts — including our own — must speculate as to whether this doctrine of uncertain ancestry has a certain future.
B
Assuming for a moment that inevitable discovery is here to stay, we next consider the scope of this doctrine in cases characterized by unlawful searches. Time constraints do not allow a comprehensive survey. However, several illustrative cases suffice to show why the doctrine should not be applied here. Where there has been no lawful arrest, evidence obtained or identified during an invalid search has been held admissible if the evidence would have been discovered anyway as the result of intervening conduct, such as the voluntary statement of a material witness. United States v. Miller, 666 F.2d 991 (5th Cir. 1982). In another case, involving an assertedly invalid arrest by state officials, it was held that fingerprints taken incident to the arrest need not have been excluded from evidence in a federal prosecution where there was another, entirely separate procedure by which federal authorities would have obtained the same fingerprints. United States v. Rowell, 612 F.2d 1176 (7th Cir.1980).
These cases demonstrate that the doctrine of “inevitable discovery” is still struggling to establish a clear identity separate from the “independent source” doctrine. In the cases cited, the “inevitable discovery” doctrine has focused upon an actual, independent source of evidence — i.e., the voluntary statement in Miller, or the separate fingerprinting procedure available in Rowell. This conceptual overlap recently led the Court of Appeals for the Second Circuit, which earlier had expressed disapproval of the “inevitable discovery” doctrine in Parowtian, to criticize the conceptual foundations of the doctrine. In United States v. Alvarez-Porras, 643 F.2d 54, 64 (2d Cir.1981), the Second Circuit questioned the origin of the “inevitable discovery” doctrine and suggested that cases in which it has been applied “can also be explained on the basis of the more firmly established exceptions to the exclusionary rule.” The court stated that the difference between “inevitable discovery” and other doctrines was so tenuous that all of the doctrines should be viewed “more as helpful guides than as rigid tests.” Id. at 60. The court concluded that it would examine each case on an individual basis, balancing the extent of the illegality against the probative value of the evidence, “without embracing or rejecting” the doctrine of inevitable discovery. Id. Upon this pragmatic, rather than doctrinal basis, the court upheld admission of evidence seized during a search which was illegal only because the officers were mistaken in the belief that a warrant had been issued.
Beyond the conceptual difficulties attendant to defining inevitable discovery, most courts appear to recognize that the doctrine has a limit in search cases. Where the police knowingly have conducted a warrantless search, and where no exception to the warrant requirement exists, the evidence seized during that search has been suppressed despite the government’s contention that if the search had been conducted lawfully the same evidence would have been found. See United States v. Allard, 634 F.2d 1182 (9th Cir.1980); United States v. Griffin, 502 F.2d 959 (6th Cir.1974), cert. denied, 419 U.S. 1050, 95 S.Ct. 626, 42 L.Ed.2d 645 (1974). But see United States *225v. Fitzharris, 633 F.2d 416 (5th Cir.1980), cert. denied, 451 U.S. 988, 101 S.Ct. 2325, 68 L.Ed.2d 847 (1981). In Griffin, the Court of Appeals for the Sixth Circuit stated:
The assertion by police (after an illegal entry and after finding evidence of crime) that the discovery was “inevitable” because they planned to get a search warrant and had sent an officer on such a mission, would as a practical matter be beyond judicial review. Any other view would tend in actual practice to emasculate the search warrant requirement of the [fjourth [ajmendment. [Id. at 961.]
The case now before us is unlike the cases, such as Miller and Rowell, in which the inevitable discovery doctrine has been applied. It also is dissimilar to AlvarezPorras, where a pragmatic view was taken toward miseommunication about existence of a warrant. Here the police knowingly conducted a warrantless search which the lead opinion treats as invalid. Thus, accepting our colleague’s view of that search as the predicate of this discussion, the case is more akin to Griffin and Allard, where application of the inevitable discovery doctrine has been rejected.
The lead opinion seeks to invoke the doctrine by reasoning that if the police had not made what the opinion treats as an invalid arrest, but instead had extended the investigative stop until reasonable cause for arrest could be established, then a valid arrest would have been made and the money would have been discovered in a search incident to that arrest. Thus, the opinion replaces the purportedly unlawful search incident to an invalid arrest with a hypothetical, proper search incident to a postulated, valid arrest.
We reject this approach upon two grounds. First, hypothesizing an extended investigative stop, rather than accepting the investigative stop which actually ripened into an arrest due to actions of the police, is inconsistent with the Idaho Supreme Court’s decision in State v. Post, supra. In Post, a police officer who suspected the defendant of possessing marijuana searched the defendant’s person and his automobile without a warrant, found incriminating evidence, and arrested the defendant. The defendant successfully challenged the validity of the search, and the state appealed. The state contended, initially, that the search was incident to a valid arrest; but the Supreme Court rejected that contention, noting that a search incident to an arrest could not precede the arrest. The state then asked the court to treat the search as incident to an investigative stop. However, the Court noted that the police officer had not engaged in a pat-down search consistent with an investigative stop, but had reached inside the defendant’s pockets, exceeding the limited scope of a search permitted during an investigative stop. The court focused upon the search that actually occurred, not upon a hypothetical search that might have occurred if the police had acted differently.
Secondly, we reject the lead opinion’s approach because it is inconsistent with the manner in which most other courts have applied the inevitable discovery doctrine to searches. Those courts have not sought to sanitize their cases by substituting correct government action for what they have held to be unlawful government action. Rather, they have said that even if the searches there in question were disregarded, the police would have acquired the evidence by a different means — e.g., an entirely separate law enforcement procedure, a voluntary statement, or the finding of a dead body. These courts have hypothesized, in varying degrees, about the alternate means by which the police could have obtained the evidence; but they have not hypothesized that the police would do legally what the police in those cases knowingly did illegally-
Under the lead opinion here, in any case where an illegal search had been conducted, the state would be invited to show that if the police had not searched illegally, but instead had done whatever was necessary to make the search legal (e.g., obtain a warrant or secure additional information to *226establish reasonable cause), the evidence in question would have been obtained. As noted in Griffin, these types of showings “would as a practical matter be beyond judicial review” and “would tend in actual practice to emasculate the search warrant requirement of the [fjourth [ajmendment.”
Again, we respect our colleague’s opposing view. Despite our doubts concerning the conceptual foundation of inevitable discovery, the doctrine has attracted a following and may continue to develop in some form. However, we believe the doctrine, even if accepted, is not intended to swallow the exclusionary rule whole by substituting what the police should have done for what they really did. We conclude that the doctrine of inevitable discovery need not, and should not, be invoked in this case.