(dissenting).
I dissent because I believe that this case goes beyond the outer limits of a permissible investigative stop under Terry v. Ohio, 392 U.S. 1, 88 S.Ct. 1868, 20 L.Ed.2d 889 (1968).
The inquiry into the appropriateness of an investigative stop under Terry and its progeny involves a consideration of two factors: (1) “whether the officer’s action was justified at its inception”; and (2) “whether it was reasonably related in scope to the circumstances which justified the interference in the first place.” Terry 392 U.S. at 20, 88 S.Ct. at 1879. The Terry stop must be conducted in a manner proportionate to the degree in which an officer’s reasonable suspicions have been aroused: “The greater the intrusion, the stronger the basis for the officers’ action must be.” United States v. Vasquez, 638 F.2d 507, 520 (2d Cir.1980), cert. denied, 450 U.S. 970, 101 S.Ct. 1490, 67 L.Ed.2d 620 (1981). I have grave reservations about the adequacy of the basis on which the stop here was initially made. In addition, even assuming that some stop was justified under the circumstances, I believe the court’s opinion misconstrues the relationship between the two prongs of the test. The court does not limit the extent of the stop to the degree of the suspicion aroused. Rather, the court ratifies the police officer’s drawing of his gun by overstating both the strength of those suspicions and the basis for assuming that the officers were in danger. I do not believe these findings are supported by the record.
The Supreme Court, interpreting Terry, has stressed that “ ‘[the] demand for specificity in the information upon which police action is predicated is the central teaching of this Court’s Fourth Amendment jurisprudence.’ ” United States v. Cortez, 449 U.S. 411, 418, 101 S.Ct. 690, 695, 66 L.Ed.2d 621 (1981) (quoting Terry v. Ohio, 392 U.S. at 21 n. 18, 88 S.Ct. at 1880 n. 18 (1968)). The facts in Terry provide an example of reasonable suspicions based on specific observations and a graduated police response.1 General circumstances may be considered as factors in combination with specific facts but they do not suffice when they stand alone.
*115The Court has accordingly held that suspicions based solely on very general characteristics of a “drug courier profile” do not meet the specificity requirement. Reid v. Georgia, 448 U.S. 438, 440-41, 100 S.Ct. 2752, 2753-54, 65 L.Ed.2d 890 (1980). Similarly, the mere presence of an individual in a location known as a high crime area is not a “specific, objective” fact under Terry. Brown v. Texas, 443 U.S. 47, 51, 99 S.Ct. 2637, 2640, 61 L.Ed.2d 357 (1979). Finally, society’s legitimate interest in encouraging trained police officers to use their judgment does not permit them to conduct searches and seizures at their “unfettered discretion.” Id.
The Court has upheld Terry stops where these general factors were combined with specific circumstances. Thus, in United States v. Cortez, 449 U.S. 411, 101 S.Ct. 690, 66 L.Ed.2d 621 (1981), cited by the court here, the Supreme Court viewed the characteristics of the location of the stop to be of “critical importance.” This general circumstance was combined in that case, involving the smuggling of aliens, with a police analysis of remarkable specificity, predicting the day, time, place, nature of the vehicle sought and other circumstances.2
While I would not require that level of specificity in all cases, the specific facts here are sparse. A man let another man into his car after a brief conversation and drove about two and a half blocks, stopping on a deserted street. After they “appeared” to have another brief conversation, with heads inclined, the passenger exited the car and walked back where he came from. During the last conversation, the two men were in full view of the unmarked car containing the police officers and made no attempt to conceal what they were doing. The officers did not see any objects being exchanged.
The district judge apparently felt uneasy about simply deferring to the officers’ judgment on these facts alone. After weighing the pros and cons of such deference, he stated:
In my view the key to resolving the conflict in this case is the location. I agree with defense counsel that the same sequence of events occurring in the center of a suburban town would not have resulted in a stop by the police____ I have seen Hayward Place. It is a depressing and inhospitable location which one would hardly choose for a legitimate transaction.
For the district judge, then, the location was not only an important background element against which to interpret the specific facts, but the central “fact” itself. But this “fact,” i.e., the Combat Zone, does not carry the dispositive weight attributed to it, even aside from the holding in Brown v. Texas.3 As the district judge noted:
The case is complicated by the fact that the stop in this case occurred in the early afternoon, when there are legitimate businesses operating in the Combat Zone and Washington Street serves as a conduit for legitimate traffic to the downtown area. There are also many people in the Combat Zone for personal purposes that, while certainly not admirable, are not yet illegal.
District Court Opinion at 7-8. There is, moreover, a certain circularity to the reasoning concerning Hayward Place: that *116street is said to be an unlikely choice for a “legitimate transaction” but the main basis for concluding that there had been any transaction was that the car had stopped on Hayward Place.
The court cites appellant’s conduct, as interpreted by trained police officers, as a second articulable basis for police suspicions. The paucity of the objectively suspicious elements of that conduct, however, means that the court is in fact according dispositive deference to the officers’ discretion. To be sure, as the court correctly notes, the issue here is not whether appellant’s conduct might be “consistent only with guilt.” Yet the court would have us find reasonable suspicion on the basis of an unwarranted reliance on police judgment in the absence of persuasive objective facts. In nearly every reported case in which location played a critical role, significantly greater specific characteristics of the individuals detained or of the date, the time, or the location, supported the officers’ judgment. See, e.g., United States v. Rickus, 737 F.2d 360 (3d Cir.1984) (car travelled abnormally slowly through closed business district at 3:30 a.m.; area had experienced a recent spate of robberies; car drove back and forth in front of stores, then stopped in front of residence); United States v. Harley, 682 F.2d 398 (2d Cir.1982) (specific building staked out over long period); United States v. Gomez, 633 F.2d 999 (2d Cir.1980) (detailed background information on suspects; car paused repeatedly for apparently prearranged meetings; agents stopped suspects only after determining destination of brown paper bag), cert. denied, 450 U.S. 994, 101 S.Ct. 1695, 68 L.Ed.2d 194 (1981); United States v. Constantine, 567 F.2d 266 (4th Cir.1977) (1:40 a.m.; recent rash of vandalism), cert. denied, 435 U.S. 926, 98 S.Ct. 1492, 55 L.Ed.2d 520 (1978); United States v. Magda, 547 F.2d 756 (2d Cir.1976) (officer saw two men exchange something; one of the men walked away rapidly after looking in officer’s direction; incident occurred near a park under 24-hour surveillance), cert. denied, 434 U.S. 878, 98 S.Ct. 230, 54 L.Ed.2d 157 (1977); United States 485 F.2d 365 (2d Cir.1973) (known prominent drug trafficker with brown bag exited restaurant known for narcotics transactions), cert. denied, 415 U.S. 931, 94 S.Ct. 1444, 39 L.Ed.2d 490 (1974).
Here, by contrast, we are asked to find reasonable suspicion on the basis of quite general characteristics of a sizeable area of the city, when the suspicion was not grounded in any specific information about date, time, or the particular individuals. Nor was there any finding that Hayward Place was a street known specifically for drug deals or other illegalities. The suspects did indeed commit the act of appearing to converse briefly on that street. Yet, although they are supposed to have gone to Hayward Place because it was deserted, they made no attempt to conceal themselves from a car that that had followed them into the deserted street and that had parked behind them with an unobstructed view of their car. Their failure to conceal themselves is particularly striking because the second car, which we now know was an unmarked police car, contained three men who were observing them and discussing their observations. It would seem that, for the court, the Combat Zone is a per se region of lessened expectation of privacy, at all times of the day and at all periods of the year, where practically unlimited deference is granted to police officers’ discretion. More particularly, the court, following the district judge’s views about where one would “choose” to conduct one’s affairs, has effectively eliminated any fourth amendment scrutiny of police suspicions concerning activity on Hayward Place.
The test for a Terry stop cannot end at the evaluation of the suspicion but must balance that suspicion against the nature of the stop. As the Supreme Court has noted, “[sjtreet encounters between citizens and police officers ... range from wholly friendly exchanges ... to hostile confrontations of armed men.” Terry, 392 U.S. at 13, 88 S.Ct. at 1875. The factors justifying the particular kind of stop at issue, no less than those supporting the making of any stop at all, must be evaluat*117ed in light of Terry’s specificity requirement. The court here seeks to justify the armed character of the stop due to the officers’ reasonable fears for their safety. Two factors are cited to support this view: (1) “the officer suspected appellant of dealing in narcotics ... conduct rife with deadly weapons”; and (2) the nature of the location. I have already discussed the factor of the location but wish to make an additional observation. It seems strange that the record does not provide any explanation for the fact that the officers did not stop the suspect on Hayward Place but rather some distance away. Does this have some bearing on either the strength of the suspicions or on when those suspicions jelled in the officers’ minds?
No less problematic, in light of Terry’s specificity requirement, is the court’s justification of the armed stop on the basis of the officers’ suspicions of a narcotics transaction. The finding of a specific suspicion of a narcotics transaction is not wholly supported by the record. The district judge found only that the “officers believed that some sort of illegal transaction had occurred.” District Court Opinion at 3 (emphasis added). The district judge appears to have phrased his finding in this way because of some uncertainty in the testimony about exactly what the officers suspected.4 Moreover, when Officer McReynolds testified that the “nature of the transaction” influenced his use of a weapon, he was not necessarily referring to narcotics dealings and their association with deadly weapons. What he said was that “the transaction, the apparent transaction, was such that any easily concealable item could have been bought or sold or exchanged, including ... a weapon of some sort.” Hearing on Motion to Suppress at 23. Indeed, when discussing the initial justification for the making of a stop, the court repeated the district judge’s finding by stating that what the officers suspected was some “illegal transaction.” The district court’s finding of a reasonable suspicion of “some sort of illegal transaction,” itself based on rather general factors, becomes the justification for an armed stop on the basis of the officers’ speculations about what such a transaction might have included.
The district judge stated his own finding about the weapon thus:
McReynolds had a reasonable suspicion that an illegal transaction had just taken place in an area in which narcotics transactions frequently took place and in which people frequently were found to be armed..,. Given the history of the Combat Zone, [drawing a gun] was elementary prudence.
This formulation expresses the way in which rather general ideas were connected to each other to yield the result in this case without much attempt to relate these ideas to the factual findings. For example, assuming the reasonableness of some suspicion, there were no findings concerning the kind of narcotics transactions that are fraught with danger. Are small-scale street transactions “rife with deadly weapons” or only large dealings involving professional traffickers? Is a meeting between two individuals which might involve “some sort of illegal transaction” fraught with danger for police officers? What basis is there in the record for any of the more generalized conclusions supporting the result here? Does “the history of the Combat Zone” concern both nighttime and daytime activities? I do not wish to engage in armchair second-guessing of police officers but think our conclusions should be based on findings which have direct support in the record.
The court cites United States v. Harley, 682 F.2d 398 (2d Cir.1982), for the elements to be considered in evaluating whether the force police employ converts a Terry stop into an arrest. I believe that test to be correct. Yet we may also look to the facts in Harley for some guidance as to the way *118in which to apply the test. Harley involved a specific building under periodic surveillance by DEA agents for several weeks; informants had told of narcotics transactions; the agents motioned to a suspect to pull over and a high-speed car chase ensued; when agents caught up to the suspect, he reached behind him as though for a gun. At that point, the agents drew their weapons.
I would not, of course, limit the reasonableness of drawing a gun to such circumstances. Police officers must have the right to protect themselves. The drawing of guns by the police does not automatically convert an investigative stop into an arrest, although it may do so in certain circumstances. Compare, e.g., United States v. Ceballos, 654 F.2d 177, 184 (2d Cir.1981) (drawing of guns “has been explicitly described as one of the ‘trappings of a technical formal arrest’ ”) (quoting Dunaway v. New York, 442 U.S. 200, 215 & n. 17, 99 S.Ct. 2248, 2258 & n. 17, 60 L.Ed.2d 824 (1979)), with United States v. Nargi, 732 F.2d 1102, 1106 (2d Cir.1984) (“display of guns does not automatically convert a stop into an arrest”). Yet when guns have been allowed in Terry stop situations, the officers’ suspicions have usually been much more solidly based than here. See, e.g., United States v. Santana, 485 F.2d 365 (not unreasonable to assume that notorious New York narcotics violator might be armed). In circumstances presenting a basis for suspicion closer to the present case, courts have stressed the minimal nature of the intrusion in upholding the Terry stop. See, e.g., United States v. Magda, 547 F.2d 756, 759 (officer did not “physically restrain” nor “humiliate” suspect but “simply asked him to stop”); United States v. Constantine, 567 F.2d 266, 267 (officer motioned to suspect to approach police car).
The armed stop here seems to me to be quite disproportionate to the circumstances in view of the paucity of specific articulable facts. In United States v. Ceballos, 654 F.2d 177, an armed stop was held to be an arrest because of a similarly disproportionate police response. In that case, as in this, the suspect was
completely unknown to the officers, was not reputed to be a major narcotics violator, [cites omitted], was not known to be armed or reasonably suspected of being armed; did not engage in erratic driving designed to avoid surveillance, and did not otherwise act in a way that would lead the officers reasonably to conclude that the degree of force used here was required to effect a Terry stop.
Id. at 184. Ceballos involved a specific basis for suspicion far higher than in the present case; the amount of force employed by police was, in some respects, lesser than here.5
While other courts have considered general characteristics of certain kinds of lawbreakers in analyzing the limits of a Terry stop, they have done so in cases presenting suspicions based on specific information. See, e.g., United States v. Jones, 759 F.2d 633, 640 (8th Cir.1985) (defendant had run when he saw police and was suspected accomplice of a known burglar; court found support for upholding armed Terry stop on “generalities” about burglars, contra Ceballos), cert. denied, — U.S. -, 106 S.Ct. 113, 88 L.Ed.2d 92 (1985); United States v. White, 648 F.2d 29, 43 (D.C.Cir. 1981) (anonymous tip with highly detailed information verified in large part by police observation), cert. denied, 454 U.S. 924, 102 S.Ct. 424, 70 L.Ed.2d 233, 235 (1981). Cf, United States v. Ramos-Zaragosa, 516 F.2d 141 (9th Cir.1975) (stop held to be an arrest when suspects ordered out of vehicle at gunpoint in narcotics case involving detailed informer’s tip).
*119My review of the cases shows that, while there are no bright-line rules in these matters, suspicions and fears based on specific information are the hallmark of judicial application of the two-part Terry test. Balancing the paucity of such specificity here against the degree of force employed, I cannot agree that a lawful Terry stop was conducted. I have difficulty formulating the court’s holding without seeming to exaggerate my brethren’s intent: if a citizen lets another person into his car on Washington Street, where must he let him off to avoid an armed stop by the police? The court’s opinion provides no limiting criteria. Rather, it seems to allow armed stops of individuals who meet in the Combat Zone on the basis of unlimited deference to police discretion.
I respectfully dissent.
. It is perhaps worth restating that Terry was a case where a police officer observed
two men hover about a street corner for an extended period of time, at the end of which it becomes apparent that they are not waiting for anyone or anything; where these men pace alternately along an identical route, pausing to stare in the same store window roughly 24 times; where each completion of this route is followed immediately by a conference between the two men on the corner; where they are joined in one of these conferences by a third man who leaves swiftly; and where the two men finally follow the third and rejoin him a couple of blocks away.
Terry, 392 U.S. at 23, 88 S.Ct. at 1181. The officer followed the two men and approached the group of three. He identified himself as a police officer and asked for their names. It was only when he received a mumble for a response that he finally grabbed Terry and frisked him. Terry at 7, 88 S.Ct. at 1872.
. If, on the night upon which they believed "Chevron” was likely to travel, sometime between 2 a.m. and 6 a.m., a large enclosed vehicle was seen to make an east-west-east round trip to and from a deserted point (milepost 122) on a deserted road (highway 86), the officers would stop the vehicle on the return trip. In a 4-hour period the officers observed only one vehicle meeting that description. And it is not surprising that when they stopped the vehicle on its return trip it contained "Chevron” and several aliens.
Cortez, 449 U.S. at 420-21, 101 S.Ct. at 696-97 (footnote omitted).
. Brown presented some features similar to the present case. Police stopped one of two men who were walking away from one another in an alley in a “high drug problem area.” The officers did not suspect the individual of any particular crime. Brown 443 U.S. 47, 99 S.Ct. 2637. The extent to which Brown may be distinguished depends on the characterization of the suspicions here. See, e.g., my discussion infra at 117.
. See Hearing on Motion to Suppress, at 16 (Officer McReynolds’ testimony) and 38-40 (Officer Blais’ testimony).
. In Ceballos, the suspect had emerged late at night, with a brown paper bag, from the building of a known narcotics dealer who was under surveillance. The dissenting judge in that case, who would have upheld the stop under Terry, stressed this specific basis for the suspicion. In arguing that the force applied was not equivalent to an arrest, the dissent also noted that "[t]his is not a case in which one or more officers approached a suspect and ordered him out of his car at gunpoint [cites omitted].” Ceballos at 188 (Meskill, J., dissenting).