Washington v. Lambert , 98 F.3d 1181 ( 1996 )

REINHARDT, Circuit Judge:

“The security of one’s privacy against arbitrary intrusion by the police — which is at the core of the Fourth Amendment — is basic to a free society.”

Wolf v. Colorado, 338 U.S. 25, 27, 69 S.Ct. 1359 [1361], 93 L.Ed. 1782 (1949).

In this case we apply the Fourth Amendment’s promise of security in one’s being to what is, unfortunately, an all too familiar set of circumstances- — an intrusive law enforcement stop and seizure of innocent persons on the basis of suspicions rooted principally in the race of “the suspects.”¹ On June 25, 1991, several Santa Monica police officers in police cruisers, including appellant, Skystone Lambert, followed George Washington and Darryl Hicks, two African-American men, as they drove into the parking garage of their hotel. The police shone searchlights on their car, ordered them out at gunpoint, handcuffed their hands behind their backs and placed them in separate police cars for a period of from five to 25 minutes.² Only after the officers frisked the two men, searched their car and checked their identification did they release “the suspects.”³

Asserting that their Fourth Amendment rights had been violated, Washington and Hicks filed suit under, inter alia, 42 U.S.C. § 1983. The district court granted them judgment as a matter of law after denying Lambert’s pre-trial and trial motions for judgment on the basis of qualified immunity.

The defendant justifies his actions against the plaintiffs primarily on the ground that the men bore a resemblance to a general description of two African-American suspects — namely, one fairly tall, one fairly short. The description was contained in a police bulletin. We find this an insufficient basis for such an intrusive stop. Because Lambert clearly violated Washington and Hicks’ Fourth Amendment rights, we affirm both the denial of qualified immunity and the grant of judgment as a matter of law.

FACTS AND PROCEDURAL HISTORY

Around midnight on June 25, 1991, Washington, a picture editor with Sports Illustrated, and Hicks, a senior program analyst at the Bank of New York, who were visiting the Los Angeles area from New York, were returning from a baseball game at Dodger Stadium. Perhaps not reflecting the best gustatory judgment, they decided to stop at a Carl’s Jr. restaurant in Santa Monica to get some food to take back to their hotel. Their decision proved to be an unfortunate one. Skystone Lambert, a uniformed Santa Monica police officer, had also chosen to visit Carl’s Jr. that evening. He observed Hicks and Washington and thought they resembled the description of two suspects being sought for 19 armed robberies, most of which had taken place in the western part of the vast Los Angeles metropolitan area. Lambert also thought that Washington appeared nervous.⁴ None of the robberies had occurred in the City of Santa Monica, and the most recent had occurred six days earlier.

Police knowledge of the suspects in the robberies consisted of the following. They were described as two African-American males, aged 20-30, one tall (6’ to 6’2”) and 150-170 pounds, and the other short (5’5” to 5’7”) and 170-190 pounds. They were known to have driven a variety of get-away ears— including a Porsche 911, a BMW and a stolen, white Oldsmobile Cutlass. The police bulletin also stated that they were considered armed and dangerous.

Neither Washington nor Hicks fit the specifics of the descriptions of the suspects. Washington was 6’4” and weighed 235 pounds. He was taller and far heavier than the “tall suspect.” Hicks was 5’7]é” and weighed 135-140 pounds. He was much thinner than the “short suspect.”

Based principally on what appeared to him to be physical similarities between Washington and Hicks and the two suspects, Lambert called for back-up and followed Washington and Hicks out of the fast-food restaurant. Hicks noticed they were being followed and told Washington. Washington and Hicks entered a white Plymouth Dynasty, which bore a rental car company sticker on the back bumper, and drove off. Lambert followed in his squad car. A second police car soon joined Lambert in following Hicks and Washington. Washington looked back several times, which Lambert found suspicious. While following the car, Lambert requested a check on the license plate, which revealed that it had not been reported stolen.

Washington and Hicks reached their hotel and entered the underground parking garage. Lambert did not immediately follow them into the garage because he did not observe them make the turn into the garage entrance. Thus, the police cars did not arrive until Washington and Hicks were preparing to get out of their car. The officers shone spotlights on the two men and pointed their guns at them.⁵ Using the police vehicle’s speaker system, Lambert ordered Hicks to open the car door and get out, raise his hands and interlock his fingers behind his head, face the wall, and close the car door with his feet. Lambert repeated the instructions for Washington. He ordered Washington and Hicks one by one, to walk backward toward him. He then handcuffed their hands behind their backs, patted them down, and placed them in separate police cars. Washington and Hicks complied with all orders and offered no resistance.

The officers searched the rental ear and opened up Hicks’ fanny-pack/pouch where he found identification. Lambert then reached into Washington’s pants and retrieved his wallet. The officers looked at the men’s identification and may have run a computer check. If so, it failed to reveal any outstanding warrants or other problems.⁶ In any event, shortly after the officers concluded their investigation, they released the two men.

In total, three or four police cars gathered in the hotel garage in order that the officers assigned to them could help detain Washington and Hicks. Washington estimated that there were about seven officers at the scene. Sergeant Grant, a supervisor who arrived at the end of the incident, testified that he believed that four officers were present when he arrived. No one disputes that one of the policemen was a K-9 officer with a police dog in tow.

Washington and Hicks filed suit under 42 U.S.C. § 1983 alleging a violation of their Fourth Amendment rights. Defendant Lambert moved for summary judgment on the basis of qualified immunity. The district judge, the Honorable Harry L. Hupp, denied the motion. The case went to trial, and on the third day of trial, Judge Hupp again denied defendant’s motion for a judgment of qualified immunity, granted a directed verdict for the plaintiffs, and left only the issue of damages for the jury. The district judge believed he was bound under this circuit’s precedent to grant the directed verdict because, on the undisputed facts, the detention constituted an arrest, the officers lacked probable cause, and the law was clearly established.⁷ The jury deadlocked and was unable to reach a verdict on damages. The defendants subsequently filed a motion to reconsider the previous order denying dismissal on the basis of qualified immunity. The district judge denied the motion as to Officer Lambert, again rejecting his argument that he had only performed a Terry stop and finding instead that he had made what a reasonable officer should have known was an arrest without probable cause. The judge granted qualified immunity as to the other officers on the ground that Lambert was the lead officer and the others had reasonably relied on his orders in carrying out the investigatory stop.

In October 1994, a second trial commenced solely on the question of damages. The jury returned a verdict for Washington and Hicks, awarding them $10,000 each plus costs and attorney fees. Lambert appeals both the directed verdict and the district court’s refusals to grant qualified immunity.

I. Was the police detention of Washington and Hicks a valid investigatory stop or an arrest that violated defendants’ Fourth Amendment rights?

We review the propriety of a directed verdict de novo, Redman v. County of San Diego, 942 F.2d 1435, 1439 (9th Cir.1991) (en banc), cert. denied, 502 U.S. 1074, 112 S.Ct. 972, 117 L.Ed.2d 137 (1992), and view the evidence in the light most favorable to the non-moving party, drawing all inferences in favor of that party. The ease must go to the jury if conflicting inferences can be drawn from the facts. Id.

The principal question at issue in the directed verdict (aside from qualified immunity, which we discuss infra) was whether the police action constituted a Terry stop or an arrest. Lambert argues that the stop of Washington and Hicks was a valid investigatory stop under Terry v. Ohio, 392 U.S. 1, 88 S.Ct. 1868, 20 L.Ed.2d 889 (1968). Washington and Hicks contend that the detention was not an investigatory stop but an arrest for which the police had no probable cause. We agree with the district court that under established law it clearly was an arrest.

There is no bright-line rule to determine when an investigatory stop becomes an arrest. United States v. Parr, 843 F.2d 1228, 1231 (9th Cir.1988). Rather, in determining whether stops have turned into arrests, courts consider the “totality of the circumstances.” United States v. Del Vizo, 918 F.2d 821, 824 (9th Cir.1990) (quoting United States v. Baron, 860 F.2d 911, 914 (9th Cir.1988), cert. denied, 490 U.S. 1040, 109 S.Ct. 1944, 104 L.Ed.2d 414 (1989)). As might be expected, the ultimate decision in such cases is fact-specific.

In looking at the totality of the circumstances, we consider both the intrusiveness of the stop, i.e., the aggressiveness of the police methods and how much the plaintiffs liberty was restricted, United States v. Robertson, 833 F.2d 777, 780 (9th Cir.1987) (‘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.”), and the justification for the use of such tactics, i.e., whether the officer had sufficient basis to fear for his safety to warrant the intrusiveness of the action taken. E.g., United States v. Jacobs, 715 F.2d 1343, 1345-46 (9th Cir.1983) (per curiam). In short, we decide whether the police action constitutes a Terry stop or an arrest by evaluating not only how intrusive the stop was, but also whether the methods used were reasonable given the specific circumstances. See Del Vizo, 918 F.2d at 824-25. As a result, we have held that while certain police actions constitute an arrest in certain circumstances, e.g., where the “suspects” are cooperative, those same actions may not constitute an arrest where the suspect is uncooperative or the police have specific reasons to believe that a serious threat to the safety of the officers exists. “The relevant inquiry is always one of reasonableness under the circumstances.” Allen v. City of Los Angeles, 66 F.3d 1052, 1057 (9th Cir.1995) (quoting United States v. Sanders, 994 F.2d 200, 206 (5th Cir.), cert. denied, 510 U.S. 1014, 114 S.Ct. 608, 126 L.Ed.2d 572 (1993)).

Whether we deem a particular detention a Terry stop or an arrest is ■ of great importance because the decision we make will frequently determine whether the police conduct was lawful or not. If we conclude that the detention was only a “stop” it will be lawful even though there was no probable cause. If, under the same circumstances, we term it an arrest, it will not be lawful in the absence of such cause. In the latter case, the evidence seized will be excludable and the arresting officer may be liable for damages. It would be far simpler, and on the surface far more logical, to base the decision as to whether an arrest occurred solely on the determination of how intrusive the stop and eventual detention was, on how severely the police action infringed on the suspect’s liberty. If we were to do so, we could set forth simple rules as to how intrusive police action must be before a suspect is deemed to have been arrested. There are, however, valid reasons why the law has developed differently.⁸

Although our doctrine lacks the simplicity and logic of a bright-line rule, the analytical course we have chosen proved necessary to accommodate our compelling and legitimate concerns for the safety of law enforcement personnel. In proclaiming the law of unlawful arrests, we must always keep in mind two important concerns: the safety of those who serve the public by enforcing the law and the constitutionally guaranteed “right of the people to be secure in their persons ... against unreasonable searches and seizures.” U.S. Const, amend. IV. Accordingly, when we evaluate whether the police conduct was lawful or unlawful, we must do so in light of the dangerousness of the particular situation that confronted the police. Sometimes, an investigatory stop may involve more than the ordinary risks inherent in any contact between police officers and suspects. Even though the officers may not have sufficient cause to make an arrest, they may have to take particular measures to protect themselves during the course of the stop. As a result, we allow intrusive and aggressive police conduct without deeming it an arrest in those circumstances when it is a reasonable response to legitimate safety concerns on the part of the investigating officers.

The complexity of this doctrinal scheme — i.e., that the identical police action can be an arrest under some circumstances and not in others — originated with Terry v. Ohio and subsequent decisions allowing the police to stop suspects for “investigatory detentions” with less than probable cause. When the investigatory stop became an accepted part of police procedure, courts began allowing police, in certain circumstances, to take intrusive steps to protect themselves as part of a Terry stop, while recognizing that in other circumstances the use of those same methods in connection with such a stop might turn it into an arrest. This doctrinal flexibility allows officers to take the steps necessary to protect themselves when they have adequate reason to believe that stopping and questioning the suspect will pose particular risks to their safety. See generally Terry, 392 U.S. at 23, 88 S.Ct. at 1881 (“Certainly it would be unreasonable to require that police officers take unnecessary risks in the performance of their duties.”); see also Jacobs, 715 F.2d at 1346. It is because we consider both the inherent danger of the situation and the intrusiveness of the police action, that pointing a weapon at a suspect and handcuffing him, or ordering him to lie on the ground, or placing him in a police car will not automatically convert an investigatory stop into an arrest that requires probable cause. Del Vizo, 918 F.2d at 825; see also Allen, 66 F.3d at 1056.

Although we are mindful that those who serve the public by taking on the dangerous job of enforcing the criminal laws are not required by the Fourth Amendment to take unreasonable risks, we also recognize that, by their choice of a profession, they have knowingly agreed to subject themselves to some physical jeopardy. That is inherent in the job of a law enforcement officer.⁹ In fact, it is the nature of a democratic society that all of us, especially the police, take some risks in the interest of preserving freedom. While we must not compel police officers to take unnecessary risks, total security is possible, if at all, only in a society that puts a much lesser premium on freedom than does ours. Therefore in determining whether a stop was lawful or unlawful, we must consider the risk to the police officers inherent in the situation, but we must also consider the liberty interests all Americans cherish — specifically the freedom from unreasonable searches and seizures guaranteed by the Fourth Amendment to our Constitution.

In this nation, all people have a right to be free from the terrifying and humiliating experience of being pulled from their ears at gunpoint, handcuffed, or made to lie face down on the pavement when insufficient reason for such intrusive police conduct exists. The police may not employ such tactics every time they have an “articulable basis” for thinking that someone may be a suspect in a crime. The infringement on personal liberty resulting from so intrusive a type of investigatory stop is simply too great. Under ordinary circumstances, when the police have only reasonable suspicion to make an investigatory stop, drawing weapons and using handcuffs and other restraints will violate the Fourth Amendment. Del Vizo, 918 F.2d at 825; United States v. Delgadillo-Velasquez, 856 F.2d 1292, 1295 (9th Cir.1988). In fact, even markedly less intrusive police action has been held to constitute an arrest when the inherent danger of the situation does not justify the intrusive police action. E.g., United States v. Ricardo D., 912 F.2d 337, 340-42 (9th Cir.1990) (finding arrest occurred where police held suspect in patrol car for twenty minutes without drawing weapons or handcuffing him); Robertson, 833 F.2d at 781 (numerous police officers drew guns and detained suspect, neither handcuffed nor in a police car, for 5-15 minutes); id. at 787 (Noonan, J., dissenting) (acknowledging that if the police had ordered suspect to “prone out” it probably would have constituted an arrest even if the police did not handcuff or touch the suspect); Kraus v. County of Pierce, 793 F.2d 1105, 1109 (9th Cir.1986) (holding that arrest occurred where police used guns and searchlights but neither handcuffed suspects nor restrained them in a police car), cert. denied, 480 U.S. 932, 107 S.Ct. 1571, 94 L.Ed.2d 763 (1987).

In balancing the interests in freedom from arbitrary government intrusion and the legitimate needs of law enforcement officers, we cannot help but be aware that the burden of aggressive and intrusive police action falls disproportionately on African-American, and sometimes Latino, males. Notwithstanding the views of some legal theoreticians, as a practical matter neither society nor our enforcement of the laws is yet color-blind. Cases, newspaper reports, books, and scholarly writings all make clear that the experience of being stopped by the police is a much more common one for black men than it is for white men. See, e.g., Kolender v. Lawson, 461 U.S. 352, 354, 103 S.Ct. 1855, 1856, 75 L.Ed.2d 903 (1983) (Lawson, a law abiding African-American man, was stopped or arrested fifteen times in primarily white neighborhoods in a 22-month period); Jeff Brazil and Steve Berry, Color of Driver is Key to Stops in 1-95 Videos, Orl. Sent., Aug. 23, 1992, at Al and Henry Curtis, Statistics Show Pattern of Discrimination, Orl. Sent. Aug. 23, 1992 at All (videotapes show that 70% of stops made by drug interdiction unit on portion of 1-95 in Florida are of African-Americans or Hispanics, although they made up only 5% of the drivers on that stretch of the interstate. Only about 5% of these stops lead to arrests); Michael Schneider, State Police 1-95 Drug Unit Found to Search Black Motorists f Times More Often Than White, Balt. Sun, May 23, 1996, at B2 (reporting similar statistics in Maryland four years later); David A. Harris, Factors for Reasonable Suspicion: When Black and Poor Means Stopped and Frisked, 69 Ind. L.J. 659, 679-80 (1994); Elizabeth A. Gaynes, The Urban Criminal Justice System: Where Young + Black + Male = Probable Cause, 20 Ford. Urb. L.J. 621, 628-25 (1993); Tracey Maclin, Black and Blue Encounters— Some Preliminary Thoughts About Fourth Amendment Seizures: Should Race Matter?, 26 Val. U.L.Rev. 243, 250-57 (documenting incidents); Developments in the Law — Race and the Criminal Process, 101 Harv. L.Rev. 1472, 1505 (1988) (“[P]olice often lower their standards of investigation when a suspect has been described as a minority, thus intruding upon a greater number of individuals who meet the racial description than if the suspect had been described as white.”).

Although much of the evidence concerns the disproportionate burden police action imposes on African-American males who are young and poor, there is substantial evidence that the experience of being stopped by police is also common both for older African-Americans and for those who are professionals — lawyers, doctors, businessmen, and academics.¹⁰ E.g., Schneider, 1-95 Drug Unit, at B2 (reporting complaints of an African-American couple in their mid-sixties whose minivan was pulled over and searched for drugs on their 40th wedding anniversary); Gaynes, Probable Cause, 20 Ford. Urb. L.J. at 625 (“Most black professionals can recount at least one incident of being stopped, roughed up, questioned, or' degraded by white police officers.”). For example, Deval Patrick, formerly a partner in a prestigious Boston law firm and now an Assistant Attor-. ney General of the United States and head of the Civil Rights Division at the Department of Justice, recently reported that “I still get stopped if I’m driving a nice car in the ‘wrong’ neighborhood.” Deval Patrick, Have Americans Forgotten Who They Are?, L.A. Times, Sept. 2, 1996, at B5. Christopher Darden, a suddenly well-known prosecutor, recently wrote that he is stopped by police five times a year because “I always seem to get pulled over by some cop who is suspicious of a black man driving a Mercedes.” Christopher Darden, In Contempt 110 (1996). Henry L. Gates, Jr. has written, poignantly, “[n]or does [University of Chicago Professor] William Julius Wilson ... wonder why he was stopped near a small New England town by a policeman who wanted to know what he was doing in those parts. There’s a moving violation that many African-Americans know as D.W.B.: Driving While Black.” Thirteen Ways of Looking at a Black Man, New Yorker, Oct. 23, 1995 at 59; see also Michael A. Fletcher, Driven to Extremes; Black Men Take Steps to Avoid Police Stops, Wash. Post, March 29, 1996, at A1 (reporting frequent stops by police of black professionals). These encounters are humiliating, damaging to the detainees’ self-esteem, and reinforce the reality that racism and intolerance are for many African-Americans a regular part of their daily fives. See, e.g., Charles N. Jamison, Jr., Racism: The Hurt That Men Won’t Name, Essence, Nov. 1992 at 64; Patrick, Have Americans Forgotten?.

Did the Police Arrest Washington and Hicks?

In determining the severity of the intrusion and the aggressiveness of the police action, we have stated, “handcuffing substantially aggravates the intrusiveness of an otherwise routine investigatory detention and is not part of a typical Terry stop.” United States v. Bautista, 684 F.2d 1286, 1289 (9th Cir.1982) cert. denied 459 U.S. 1211, 103 S.Ct. 1206, 75 L.Ed.2d 446, 447 (1983); see also Del Vizo, 918 F.2d at 825 (use of handcuffs important factor in determining that an arrest had occurred).

Similarly, if the police draw their guns it greatly increases the seriousness of the stop. See United States v. Alvarez, 899 F.2d 833, 838 (9th Cir.1990), cert. denied, 498 U.S. 1024, 111 S.Ct. 671, 112 L.Ed.2d 663 (1991); United States v. Strickler, 490 F.2d 378, 380 (9th Cir.1974) (armed approach to a vehicle surrounded by police constituted an arrest). The Seventh Circuit has cogently explained how a drawn gun affects the nature of an encounter with the police:

The significance of the pointed gun is that it makes the encounter far more frightening than if the officer’s gun remains holstered, or even drawn but pointed down at his side; and certainly where the danger of the encounter to the officer, though potentially serious, is not clear and present, the deliberate pointing of a gun at the suspect is problematic. It would be a sad day for the people of the United States if police had carte blanche to point a gun at each and every person of whom they had an ‘articulable suspicion’ of engaging in criminal activity.

United States v. Serna-Barreto, 842 F.2d 965, 967 (7th Cir.1988) (Posner, J.) (citation omitted).

Finally, whether the police physically restrict the suspect’s liberty is an important factor in analyzing the degree of intrusion effected by the stop. Del Vizo, 918 F.2d at 825 (removing suspect from his car and making him lie down in the street was a factor in whether an arrest had occurred); Ricardo D., 912 F.2d at 340-42; United States v. Chamberlin, 644 F.2d 1262, 1267 (9th Cir.1980) (holding suspect in police ear and questioning him for twenty minutes constituted an arrest), cert. denied, 453 U.S. 914, 101 S.Ct. 3148, 69 L.Ed.2d 997 (1981).¹¹

In determining whether the use of intrusive techniques turns a stop into an arrest, we examine the reasonableness of the police conduct in light of a number of factors. Despite the absence of a bright-line rule, our cases make clear that we have only allowed the use of especially intrusive means of effecting a stop in special circumstances, such as 1) where the suspect is uncooperative or takes action at the scene that raises a reasonable possibility of danger or flight;¹² 2) where the police have information that the suspect is currently armed;¹³ 3) where the stop closely follows a violent crime;¹⁴ and 4) where the police have information that a crime that may involve violence is about to occur.¹⁵ Clearly, some combination of these factors may also justify the use of aggressive police action without causing an investigatory stop to turn into an arrest.¹⁶

Further, in a case like the one before us, we consider the specificity of the information that leads the officers to suspect that the individuals they intend to question are the actual suspects being sought, see Alexander, 64 F.3d at 1322, as well as the specificity of the information that the persons actually being sought are likely to forcibly resist police interrogation. The more specific the information in both these regards, the more reasonable the decision to take extraordinary measures to ensure the officers’ safety. In other circumstances, similar concerns regarding specificity will also come into play.

An additional factor courts consider in analyzing the reasonableness of the use of aggressive investigatory tactics as part of a Terry stop is the number of police officers present. For example, in Serna-Barreto, the court found that where a police officer approached the suspects with his gun drawn, “it was prudent” for him to do so because he was alone and outnumbered. 842 F.2d 968 (distinguishing United States v. Ceballos, 654 F.2d 177 (2d Cir.1981) where numerous policemen approached and surrounded a single suspect with guns drawn); see also Jacobs, 715 F.2d at 1346 (holding that it was reasonable shortly after a bank robbery for single officer to order two suspects out of the ear at gunpoint and to “prone out”); Thompson, 906 F.2d at 1297 (presence of seven squad cars factor in determining that intrusive actions taken by police against two suspects in car constituted an arrest).

In this ease, Washington and Hicks did nothing immediately prior to or during their confrontation with the police to justify Lambert’s use of a complete battery of intrusive and threatening procedures in the context of a Terry stop. Thus, this case resembles Del Vizo, where we held that in light of the defendant’s complete cooperation at the scene an arrest had occurred where the police drew and pointed guns, handcuffed the suspect, and placed him in a police car. 918 F.2d at 825 (distinguishing Taylor, supra, where suspect was not cooperative). We also said in Del Vizo that, under the totality of the circumstances test, the aggressive police actions constituted an arrest even though the officers believed that the person to be detained was a drug dealer. See also Delgadillo-Velasquez, 856 F.2d at 1294—96 (finding an arrest without probable cause had occurred). This was true notwithstanding our earlier recognition of the dangers inherent in the drug trade and our previous statements that drug dealers often carry weapons. See, e.g., United States v. McConney, 728 F.2d 1195, 1206 (9th Cir.) (en banc), cert. denied, 469 U.S. 824, 105 S.Ct. 101, 83 L.Ed.2d 46 (1984).¹⁷

Moreover, none of the other factors that we have held may justify aggressive police action in the absence of probable cause existed here. Unlike in Greene, there was no specific information indicating that either Hicks or Washington was armed. Unlike in Alexander and Jacobs, there were neither specific similarities between Washington and Hicks and the suspects sought, nor had there been a violent crime in the vicinity shortly before the stop. Unlike in Buffington, there was no reason to believe that Washington and Hicks were about to commit any crime. Finally, because there were two suspects and at least four officers and a police dog present, the ratio of officers to suspects present in the garage weighs against Lambert’s using such intrusive action.

The only basis for linking Washington and Hicks to the supermarket robberies was the purported general similarity of their physical characteristics to those of the actual suspects: two African-American males, one reasonably short and one reasonably tall. Not only were the descriptions exceedingly vague and general — there were, for example, no specific descriptive features such as facial hair or scars — but Washington and Hicks did not even match the few physical details that did accompany the general physical descriptions. The differences in weight were more than significant and exceedingly difficult to ignore. For example, the difference in appearance between a short man of 170 to 190 pounds and a short man of 135 to 140 pounds is substantial. If the general descriptions relied on here can be stretched to cover Washington and Hicks, then a significant percentage of African-American males walking, eating, going to work or to a movie, ball game or concert, with a friend or relative, might well find themselves subjected to similar treatment, at least if they are in a predominantly white neighborhood. Moreover, other equally general descriptions could serve as the basis for similar demeaning treatment of many other African-Americans. We note, incidentally, that it is extremely questionable whether the tenuous general physical similarities between Washington and Hicks and the supermarket robbers give rise to even the reasonable suspicion necessary to make a Terry stop. Because the parties did not raise that issue, we will not pursue it farther. Nevertheless, it is clear that Officer Lambert had an insufficient basis on which to justify conducting an investigatory stop in so aggressive and intrusive a manner.

Lambert contends that other factors supported his actions. He argues that because Washington and Hicks were in a white rental car, his reaction was justified.¹⁸ This contention is just plain absurd. That a stolen white car of a different make and model was used in one of the numerous robberies does not lend any credence to the argument that the police reasonably suspected that they had found the serial robbers. The culprits had also used a number of other vehicles, none of which was reported to be white. There is no reason in this case therefore (or undoubtedly in any other) to believe that the wrong-doers had a particular affinity for vehicles that were colored white.¹⁹ Few if any criminals have heretofore been reported to have any such predilection or fetish.

Lambert also relies on the fact that the stop took place at night, the time when the robberies had occurred. The logic of this argument eludes us. If Washington and Hicks had been lingering near a supermarket at night, the fact that the supermarket robberies had all occurred at night-time might make the time of their lingering relevant. But we see no connection whatsoever between the fact that the robberies had taken place at night, and the fact that Washington and Hicks stopped for dinner at night, or turned into the parking garage of a hotel where guests sleep at night. In fact, the only significance we can attach to the fact that the two black men were observed out together at night is that the officers may have thought that they were in the wrong place at the wrong time, that there was no legitimate reason for them to be at a restaurant in that neighborhood at night-time. See Kolender, 461 U.S. at 354, 103 S.Ct. at 1856 (African-American man stopped or arrested fifteen times in 22-month period in white neighborhoods, primarily at night). Any such assumption by law enforcement officers or others must be vigorously rejected. We would be less than candid were we not to note that we very much doubt that, under identical circumstances, two white men ordering a take-out dinner at a Carl’s Jr. in Santa Monica would have been subjected to such highly intrusive and degrading treatment simply because of a police bulletin that contained a description of two robbers, one of whom was reported to be fairly tall and one fairly short.

Finally, Lambert relies on the fact that he believed that one or both of the men appeared nervous. This fact is wholly unpersuasive given Lambert’s other testimony that the two men resembled the supermarket robbers in that they appeared to be “casual and not ... too nervous.” In any event, Lambert’s testimony that Washington looked at him and looked away a few times inside a restaurant did not make it more likely that Washington was a criminal. People in restaurants often look at other patrons, and when observed, look away. It is also a fact that many innocent black men, and even many innocent white men, will appear nervous when they notice that they are being followed by the police, and even more so when their automobile is being followed by two police cruisers.

Viewing the facts in the light most favorable to defendant, we find Judge Hupp’s decision to be correct. At most there were two African-American men, one short and one tall, in the city of Santa Monica, at night, who appeared to a police officer to be both nervous and casual. This case is not a close one; any reasonable juror would be compelled to find on these facts that the stop was an arrest. Because appellant conceded that there was no probable cause, Judge Hupp rightly directed a verdict and left the issue of damages for the jury.

II. Was Lambert entitled to qualified immunity as a defense in this case?

Lambert also appeals the district court’s denial of his request for qualified immunity. We review that issue de novo. Act Up!/Portland v. Bagley, 988 F.2d 868, 871 (9th Cir.1993).

When a police officer asserts qualified immunity for Fourth Amendment violations, the district court must apply a two-part analysis. Id. Both determinations are questions of law for the court to decide, at least in cases such as this, in which there is no genuine issue of disputed material facts. Id. at 873.

The first question is whether the right at issue is clearly established and stated with particularity. Alexander, 64 F.3d at 1319. “The contours of the right must be sufficiently clear that a reasonable official would understand that what he is doing violates that right.” Id. (quoting Anderson v. Creighton, 483 U.S. 635, 640, 107 S.Ct. 3034, 3039, 97 L.Ed.2d 523 (1987)). The second question is whether the official could have had a reasonable belief that his conduct was lawful. If not, he is not entitled to qualified immunity. Act Up!/Portland, 988 F.2d at 873. We answer the second question by determining whether “a reasonable officer” could have had the belief in question, not whether the individual officer involved actually thought his conduct was lawful. Id.

Here, defendant concedes that probable cause to arrest Washington and Hicks did not exist. As is plain from reading the previous section, at the time of Washington and Hicks’ detentions the law was clearly established that, when making a Terry stop, officers may not use highly intrusive measures such as the ones used here, unless the circumstances reasonably justify such extraordinary procedures in order to ensure the officers’ safety. See, e.g., Jacobs, 715 F.2d at 1346.²⁰ The law was also clearly established that if the Terry-stop suspects are cooperative and the officers do not have specific information that they are armed or specific information linking them to a recent or inchoate dangerous crime, the use of such aggressive and highly intrusive tactics is not warranted, at least when, as here, there are no other extraordinary circumstances involved. See, e.g., Del Vizo, 918 F.2d at 825; Ricardo D., 912 F.2d at 340; Delgadillo-Velasquez, 856 F.2d at 1295; Robertson, 833 F.2d at 782; Kraus, 793 F.2d at 1109 (affirming district court’s denial of qualified immunity). This is particularly true where the similarity of description is tenuous and there is no other information suggesting that the person to be questioned is the person thought to have committed an offense. Thus, under Act Up!/Portland, the first requirement for holding qualified immunity inapplicable is met.²¹

The second question is whether, given the meager amount of information Lambert had about Washington and Hicks, and the tenuous nature of the “articulable suspicion,” a reasonable officer could have believed that it was reasonable to employ so highly intrusive a means of making a stop. Appellant’s reasons for taking such intrusive actions were wholly insubstantial. Having found the facts that it was night and that Washington and Hicks were in a white rental car to be of no significance whatsoever, and the fact that a person appears both “nervous” and “casual” to a police officer to be equally devoid of significance, our inquiry is reduced to the following: should a reasonable police officer have known that a highly intrusive investigatory stop was illegal when it was based on the facts that Washington and Hicks were both relatively young African-American men, that they were out together, and that their heights were in the same general range as the suspects’? The answer is obvious.

Lambert relies primarily on Allen and Alexander to support his claim that he is entitled to qualified immunity. Neither case is of help to him.²² In Allen, there were numerous factors that could have caused a reasonable police officer to believe that the police action was lawful. The suspects led the police on a high-speed chase and refused to pull over. When they finally stopped, the driver was very uncooperative, and Allen was drunk and had no identification. In addition, there were indications that the car had been stolen. Allen, 66 F.3d at 1054 — 57. None of the factors present in Allen was present in the instant case.

The facts here are also significantly different from the dispositive facts in Alexander. There, a reasonable police officer could have thought that he had justifiable reasons to employ tactics similar to those used here because there were specific reasons to suspect that the detainees were armed and dangerous. First, less than an hour earlier a robbery in which the suspects fired shots had occurred in the area in which the stop was made. There is a significant difference between using aggressive tactics to make a stop because a specific violent crime has occurred in the vicinity within the past hour or less and using such tactics because crimes were committed in a neighboring community a week or so earlier. Second, in Alexander, “there were several specific identifying features ... that justified the officers’ suspicions.” Alexander, 64 F.3d at 1322 (emphasis added). A witness had identified one of the two suspects as having a moustache and a blue and white striped shirt. One of the men the officers stopped had both a mous-tache and a blue and white striped shirt. Third, when the officers stopped the ear, they discovered the vehicle was not leased in the names of either the driver or the passenger, and although they were both men, a woman’s purse and checkbook were in the car.

The court in Alexander decided the qualified immunity question in the officers’ favor, but stated that it was .“undeniably a close question.” Id. at 1321. If Alexander was a close ease, then this one certainly is not. In fact, it is well on the other side of the line from Alexander. Here, there is neither proximity in time to an offense nor specificity in the similarity of description and other information suggesting that the individuals to be questioned were the suspects actually being sought.

The case before us presents, quite simply, an egregious violation of the Fourth Amendment. To build on a theme from Judge Posner — it would be a sad day for the United States if two African-American men who only somewhat resemble a very general description of men suspected of robberies in the same general area of a major metropolis can for that reason alone be subjected to the terrifying and humiliating experience that Washington and Hicks endured in the City of Santa Monica on June 25, 1991. Cf. Serna-Barreto, 842 F.2d at 967.

CONCLUSION

The district court properly denied defendant’s motions for qualified immunity and properly granted a judgment as a matter of law for Washington and Hicks on liability. The evidence supports only such a verdict.

AFFIRMED.

1 . In recent years, police in the Los Angeles area have unlawfully detained Hall of Fame baseball player Joe Morgan at the Los Angeles airport. Morgan v. Woessner, 997 F.2d 1244, 1254 (9th Cir.1993) (affirming district court holding that Morgan’s seizure on basis of tip that “made all black men suspect” and Morgan’s walking away from police was illegal). The police have also erroneously stopped businessman and former Los Angeles Laker star Jamaal Wilkes in his car and handcuffed him, and stopped 1984 Olympic medalist Al Joyner twice in the space of twenty minutes, and once forcing him out of his car, handcuffing him and making him to lie spreadeagled on the ground at gunpoint. A Foul on 'Silk', L.A. Times, March 2, 1991; John Schwa-da, Track Olympian Settles False-Arrest Lawsuit, L.A. Times, Februaiy 8, 1995. Similarly, actor Wesley Snipes was taken from his car at gunpoint, handcuffed, and forced to lie on the ground while a policeman kneeled on his neck and held a gun to his head. John L. Mitchell, Four Actors Allege Abuse by Police, L.A. Times, April 13, 1991. Actor Blair Underwood was also stopped in his car and detained at gunpoint. Id. We do not know exactly how often this happens to African-American men and women who are not celebrities and whose brushes with the police are not deemed newsworthy. It is clear, however, that African-Americans are stopped by the police in disproportionate numbers. See also discussion infra, at 1187-88. We believe that there is much truth to Justice Jackson's statement, made almost fifty years ago in Brinegar v. United States, 338 U.S. 160, 181, 69 S.Ct. 1302, 1313-14, 93 L.Ed. 1879 (1949), that "there are[] many unlawful searches of homes and automobiles of innocent people which turn up nothing incriminating, in which no arrest is made, about which courts do nothing, and about which we never hear.” (Jackson, J., dissenting).

2 . The period from the initial handcuffing to the release lasted 15 to 25 minutes according to Washington, and about five to ten minutes according to the defense. For purposes of our decision, we assume Lambert and Marroquin's version to be correct.

3 . Lambert initially testified that he had radioed back to the station requesting a computer check for outstanding warrants on the two men and that the check had revealed no outstanding warrants. However, the radio log did not reflect such a request, and Lambert subsequently testified that he was not certain whether he had actually requested a computer check on the two men.

4 . Lambert's testimony on this issue is somewhat puzzling. He testified that Washington was nervous and "kept looking at me, staring at me,” but he later testified that "he did not really observe any suspicious behavior of either Mr. Washington or Mr. Hicks,” who appeared "kind of casual,” and thus resembled the supermarket robbers who "were described as being casual and not being too nervous.”

5 .Both Officer Lambert and Officer Rutan testified that they drew their guns and pointed them. Officer Marroquin testified that he did not draw his gun.

6 . See supra note 2. The confusion in Lambert’s mind on this point is not material to any of the issues before us.

7 . Lambert conceded below that if the stop was an arrest, there was no evidence that would establish probable cause. He does not argue there was probable cause on this appeal.

8 . The district judge, an experienced and able jurist, expressed the view that the law, in its current state, lacks logic and expressed hope that this court would read his comments. We have not only read them, we have considered them carefully, and it is for this reason in part that we provide the explanation that follows in the text.

9 . As Officer Lambert stated in response to a question from plaintiffs’ counsel, “I’m always in danger when I’m working, ma’am.”

10 . Cf. supra note 2 (concerning actors and athletes).

11 . The duration of the detention may also be a factor to consider. The Seventh and Eighth Circuits have stated that the court should consider both the duration of the stop and the "degree of fear and humiliation that the police conduct engenders.” United States v. Lego, 855 F.2d 542, 544-45 (8th Cir.1988) (citing United States v. Serna-Barreto, 842 F.2d 965, 967 (7th Cir.1988)).

Although the length of detention is relevant in some cases, in others the police tactics may be sufficiently aggressive and intrusive that even a brief stop is sufficient to constitute an arrest. Cf. United States v. Delgadillo-Velasquez, 856 F.2d 1292 (9th Cir.1988) (holding that the stop constituted an arrest because "[t]he show of force and detention techniques used in this context are indistinguishable from police conduct in an arrest.”).

12 . For example, in Allen, we held that no arrest occurred despite the police's drawing their weapons, handcuffing Allen and ordering him to lie down on his stomach, where the suspects had led the police on a protracted high-speed car chase and refused to pull over. When they did finally pull over, one suspect was "non-compliant and combative," and Allen was drunk. Allen, 66 F.3d at 1057; see also United States v. Taylor, 716 F.2d 701, 709 (9th Cir.1983) (holding that no arrest had occurred where the police approached the suspect with guns drawn and handcuffed him after he twice refused to raise his hands and "ma[de] furtive movements inside the truck where his hands could not be seen.”); United States v. Greene, 783 F.2d 1364, 1366 (9th Cir.) (per curiam) (suspect in car failed to comply with the initial police command that he put his hands up), cert. denied, 476 U.S. 1185, 106 S.Ct. 2923, 91 L.Ed.2d 551 (1986). See also United States v. Jones, 973 F.2d 928, 931 (D.C.Cir.1992) (action necessary to restrain fleeing suspect), cert. denied, 510 U.S. 1065, 114 S.Ct. 741, 126 L.Ed.2d 704 (1994); Bautista, 684 F.2d at 1289 (one officer watching two suspects, one of whom appears to be thinking about fleeing).

13 . See, e.g., United States v. Thompson, 906 F.2d 1292, 1294 (8th Cir.) (informant’s tip indicating that men in car were going to rob bank and were carrying “three large caliber handguns”), cert. denied, 498 U.S. 989, 111 S.Ct. 530, 112 L.Ed.2d 540 (1990); Greene, 783 F.2d at 1368 ("The informant told the police that she had seen a pistol in the motel room of the two men she had described.”)

14 . See, e.g., Jacobs, 715 F.2d at 1346 (suspects closely match radio description of “possibly armed” group that had robbed a bank twenty minutes earlier); see also Alexander v. County of Los Angeles, 64 F.3d 1315, 1317 (9th Cir.1995) (officers stop suspects in the vicinity of an armed robbery in which shots had been fired forty-five minutes earlier).

15 . See, e.g., United States v. Buffington, 815 F.2d 1292, 1295 (9th Cir.1987) (police stop suspects near bank after informant's tip that three men are planning to rob a bank, and police recognize one suspect who has a history of violent criminal behavior).

16 . See, e.g., Greene, 783 F.2d at 1367-68 (officers have specific information that the suspects are armed, and one suspect fails to obey officer’s initial command to raise his hands).

17 . In Del Vizo, we ultimately held that the police action, which was overly intrusive under the circumstances to be deemed a Terry stop, was lawful, because probable cause to make an arrest existed.

18 . During his testimony, Lambert stated that "various cars were being used in the robberies ... [and that the robbers] could be using anything.”

19 . Lambert correctly notes that in Alexander, in vaguely similar circumstances, this court gave some weight to a policeman’s claim that one of the factors that led to his making a stop was that “the car was a rental car and in his experience, criminals often use rental cars.” 64 F.3d at 1320-21. That statement is irrelevant to this case. In Alexander, the officers did not know whether the car used by the armed robbery suspects was a rental or not; but they had a description of the color and type of car. Therefore, the fact that criminals often use rental cars would have been a relevant consideration in the officers’ suspicion that the men they saw were in fact the suspects. Here, the police contend that it was relevant that Washington and Hick's car was white and a rental. However, before they stopped them, the officers knew that there were no warrants out on the car. The only white car linked to any of the supermarket robberies had been reported stolen, not rented. Certainly, the other vehicles reported to have been used in the robberies, the Porsche 911 and the BMW, are not the types of vehicles ordinarily rented by supermarket robbers. Unlike in Alexander, the fact that Washington and Hicks were in a rental car was therefore absolutely irrelevant.

20 . We must of course look to the law as it existed at the time of the challenged conduct. The only cases discussed in the previous section that had not been decided at the time Lambert detained and questioned Washington and Hicks are Alexander and Allen. Both are cases that Lambert relies on to support his position.

21 . The recent decision in United States v. Torres-Sanchez, 83 F.3d 1123 (9th Cir.1996), is consistent with this rule. In that case, we determined that an investigatory stop did not become an arrest where the suspect was questioned inside a patrol car and detained for twenty minutes after having been stopped for speeding and having no license plate on the truck. In making this determination, we relied on the officer's merely suggesting to Sanchez that the two of them sit inside the police car because it was cold, the presence of a number of specific factors that indicated that Sanchez and his companions were involved in criminal activity, and the fact that the officer used no intrusive tactics, such as pointing a gun at any of the suspects or placing them in handcuffs. Id. at 1125-26.

22 . We do not consider here the question whether Lambert may properly rely on cases decided after the conduct at issue. If the cases do not set forth new law, such reliance could be proper. If they do establish new rules, such reliance would ordinarily be inappropriate. Here, we may reasonably assume that there is nothing new about either case.