A. Arrest
12
We review de novo the question whether the seizure of Ricardo prior to his formal arrest1 exceeded the bounds of an investigatory stop. United States v. Parr, 843 F.2d 1228, 1231 (9th Cir.1988). At the hearing on the suppression motions, the district court found that probable cause for the arrest existed, without any explicit finding of when the arrest occurred. Given that the government essentially concedes that there was no probable cause to arrest Ricardo prior to his first admission, the district court implicitly found that the field detention was nothing more than a Terry stop.
13
The Supreme Court itself has recognized that distinguishing a Terry investigative stop from a de facto arrest "may in some instances create difficult line-drawing problems." United States v. Sharpe, 470 U.S. 675, 685, 105 S.Ct. 1568, 1575, 84 L.Ed.2d 605 (1985). As noted by the Court in the seminal case of Terry v. Ohio, 392 U.S. 1, 30, 88 S.Ct. 1868, 1884, 20 L.Ed.2d 889 (1968), each case must be decided on its own facts. "Whether an arrest has occurred depends on all the surrounding circumstances, including the extent to which liberty of movement is curtailed and the type of force or authority employed." United States v. Robertson, 833 F.2d 777, 780 (9th Cir.1987). In reviewing the facts and circumstances of each case, we must be mindful of the narrow scope of the Terry exception--an exception based on a brief, street encounter between police and a suspect. To do otherwise would be to risk allowing the " 'exception' ... to swallow the general rule that Fourth Amendment seizures are 'reasonable' only if based on probable cause." Dunaway v. New York, 442 U.S. 200, 213, 99 S.Ct. 2248, 2257, 60 L.Ed.2d 824 (1979); see Florida v. Royer, 460 U.S. 491, 510, 103 S.Ct. 1319, 1331, 75 L.Ed.2d 229 (1983) (plurality opinion) (Brennan, J., concurring).
14
Under the totality of the circumstances, we conclude that the detention of Ricardo during field questioning by officers Bork and Emery amounted to a de facto arrest. With the high beams of a police car shining in his face, 16-year-old Ricardo was patted down, gripped by the arm, told he was not to run anymore, and directed to the back of one of two patrol cars present at the scene. In United States v. Baron, 860 F.2d 911, 916 (9th Cir.1988), cert. denied, --- U.S. ----, 109 S.Ct. 1944, 104 L.Ed.2d 414 (1989), we found coerciveness to be the critical factor in determining that an arrest had occurred. Similarly, in this case we conclude that taking hold of and isolating an unarmed, compliant juvenile in the back of a police car was unnecessarily coercive, and thus transformed the investigatory stop into an arrest.
15
Our conclusion is consistent with decisional guidelines concerning the scope of a permissible Terry stop. In general, the investigative methods used should be the least intrusive means reasonably available. Florida v. Royer, 460 U.S. at 500, 103 S.Ct. at 1325. Although the use of some force does not automatically transform an investigatory detention into an arrest, any overt show of force or authority should be justified under the circumstances. See, e.g., United States v. Holtzman, 871 F.2d 1496, 1502 (9th Cir.1989) (restraints justified by belief suspect was attempting to flee); United States v. Buffington, 815 F.2d 1292, 1300 (9th Cir.1987) (given officer's knowledge of suspect's history of violence, show of force justified by fear for personal safety). Similarly, as suggested by the Court in Florida v. Royer, 460 U.S. at 504-05, 103 S.Ct. at 1328, the police may move a suspect from the location of the initial stop without converting the stop to an arrest when it is necessary for safety or security reasons. Baron, 860 F.2d at 915.
16
We find no such justifications for taking hold of Ricardo, telling him not to run anymore, and seating him in the back of the patrol car for questioning. In Holtzman, we held that manual restraints and a show of authority were justified by the officer's belief that Holtzman was attempting to flee. 871 F.2d at 1502; see also United States v. Bautista, 684 F.2d 1286, 1289 (9th Cir.1982) (handcuffs permissible when defendant kept pacing and looking as if he was about to run), cert. denied, 459 U.S. 1211, 103 S.Ct. 1206, 75 L.Ed.2d 446 (1983). In this case, however, Officer Bork testified that Ricardo made no attempt to flee when approached. Moreover, we discern no reasons of safety or security for moving Ricardo to the patrol car. The officers had no reason to believe Ricardo was dangerous. Furthermore, the location of the "stop" was rural and there were no other people in the area when the officers pulled up. Compare Eberle v. City of Anaheim, 901 F.2d 814, 819 (9th Cir.1990) (no arrest when officers took suspect to security office at football stadium due to safety concerns caused by large, unruly crowd).
17
Although, as stated by the Fourth Circuit, the back of a patrol car "is not an ideal location for the purposes of an investigatory detention," United States v. Manbeck, 744 F.2d 360, 377 (4th Cir.1984), cert. denied, 469 U.S. 1217, 105 S.Ct. 1197, 84 L.Ed.2d 342 (1985), we recognize that the element of isolation from the public view (and concomitant coerciveness) in such a case is somewhat weaker than when a suspect is taken to a private room for interrogation. See Baron, 860 F.2d at 916 (noting that the suspect would probably have felt less threatened sitting in a patrol car in public than shut in a darkened bedroom). We have stated that there is no per se rule that detention in a patrol car constitutes an arrest, Parr, 843 F.2d at 1230, yet, "[t]here is clearly no mechanical checklist to distinguish between Terry stops and formal arrest or the equivalent of arrest," id. at 1231. Thus, we held that when a defendant was placed in the back of a patrol car for twenty minutes, questioned, and never informed that he was free to leave, the detention was equivalent to an arrest. United States v. Chamberlin, 644 F.2d 1262, 1267 (9th Cir.1980), cert. denied, 453 U.S. 914, 101 S.Ct. 3148, 69 L.Ed.2d 997 (1981).
18
Decisions in this and other circuits support the proposition that detention in a patrol car exceeds permissible Terry limits absent some reasonable justification or when the detention is something other than a brief procedure employed in a routine traffic stop. In Parr, we held that no arrest occurred when, after a routine vehicle stop, the officer placed Parr in the patrol car for a few minutes while the officer spoke to Parr's companion and conducted a search of Parr's vehicle. 843 F.2d at 1231. In that case, however, as the officer was stopping Parr, he observed Parr and his companion "bend toward the floorboard and 'make furtive movements.' " Id. at 1229. Although we did not explicitly consider this circumstance, it would appear that reasons of safety and security thus justified moving Parr to the police car. Furthermore, we noted the reasoning of the Seventh Circuit that " 'sitting in a patrol car for several minutes was merely a normal part of traffic police procedure for identifying delinquent drivers' and did not constitute custodial arrest." Id. at 1230 (quoting United States v. Rodriquez, 831 F.2d 162, 166 (7th Cir.1987), cert. denied, 485 U.S. 965, 108 S.Ct. 1234, 99 L.Ed.2d 433 (1988)). Our decision thus rested in part on a distinction between "routine traffic arrests" and "custodial arrests" for the purpose of limiting the scope of an incidental search. See id.
19
Other circuits addressing the patrol car issue have relied on similar justifications for upholding the detention as a mere investigatory stop. See, e.g., Manbeck, 744 F.2d 377-78 (defendant would have had possible access to weapons and a means of escape if allowed to remain in his own truck, and due to inclement weather, placement in patrol car was only feasible alternative); United States v. Lego, 855 F.2d 542, 545 (8th Cir.1988) (because suspect already found to have a weapon, safety concerns made it proper for lone officer to have suspect sit in patrol car while officer ran a check for outstanding warrants).
20
No comparable circumstances serve to explain or justify Ricardo's patrol car detention. As discussed previously, the officers had no reason to fear for their own safety or the safety of others. Nor was this a traffic stop with any attendant need to remove the suspect from his own vehicle and place him in a safe location. Finally, Officer Bork did not, and in fact could not, conduct a warrants check on Ricardo. See Transcript of Proceedings at 62, CR 89-184-TUC-WDB (June 14, 1989) (testimony of Officer Bork that "[t]he only time a juvenile can be entered is if he were a runaway"). In sum, there was no reason to move Ricardo from a public place to the patrol car for questioning.
21
Although the movement of Ricardo to the patrol car is a significant factor in our analysis, we reiterate that each case is decided under the totality of the particular circumstances. For example, in conjunction with the place of detention, in Baron we considered the disparity in power stemming from the fact that Baron was a female detained by three male officers. Baron, 60 F.2d at 916. Here, we take into account that a juvenile was confronted by several officers.2 To a limited extent, we also consider Officer Bork's threat of arrest for providing false information as adding to the coercive atmosphere of the detention. Thus, all the circumstances, including the facts that the suspect was sixteen, taken by the arm, told not to run, and placed in the back of the police car, indicate a degree of coercion unacceptable as part of a Terry stop, and unsupportable on anything less than probable cause. We therefore hold that the officers effectively arrested Ricardo when they detained him in the patrol car for questioning.
