This case presents the question whether police officers violated the Fourth Amendment in "seizing" the appellee, Mark A. Maragh, without reasonable suspicion. A squad of drug interdiction police officers was on duty in Union Station, Washington, D.C., at mid-afternoon. The officers noticed Maragh, who had just disembarked from a train from New York City. After watching Maragh talk to two friends, who left Maragh briefly and then returned to him after making a telephone call, the officers decided to try to interview him. Two officers approached Maragh in such a way as not to block Maragh's exit, exercised no force, displayed no weapons, and were in plainclothes; in addition, there is nothing in the record to indicate that the officer who actually spoke to Maragh was anything less than courteous and polite. One of the officers identified himself as a police officer and asked Maragh whether he would answer some questions. Maragh agreed. The appellee also agreed to allow the officer to search his shoulder bag. The search revealed drugs, whereupon another officer arrested Maragh. The entire process of interview, search and arrest lasted approximately five minutes.

The District Court granted Maragh's motion to suppress evidence discovered pursuant to the consensual search. The trial court relied in part on two cases from other circuits in concluding that the act of approaching a citizen and identifying oneself as a narcotics officer itself effects a "seizure" for Fourth Amendment purposes; that the police did not have reasonable suspicion to seize Maragh; and that therefore the stop violated the Fourth Amendment. See United States v. Maragh, 695 F. Supp. 1223 (D.D.C.1988). The District Court also concluded that "the same threatening environment that converted the conversations into a 'stop' vitiates any claim that a reasonable person in the circumstances of defendant would feel free to refuse the request to search his bag or feel free to leave." Id. at 1225.

We do not doubt, and the Government does not deny, that the officers in this case lacked the requisite reasonable suspicion to justify even a "Terry stop" of Maragh. See Terry v. Ohio, 392 U.S. 1, 88 S. Ct. 1868, 20 L. Ed. 2d 889 (1968). However, several decisions from this court, rendered after the District Court's judgment, make it clear that, under the circumstances presented in this case, the police did not "stop" or "seize" Maragh. We also find that, in concluding that Maragh behaved involuntarily in allowing the officers to search his bag, the District Court relied on incorrect legal premises. We therefore reverse and remand for further proceedings consistent with this opinion.

At approximately 4:00 p.m. three members of a District of Columbia Metropolitan Police Department detective squad observed Maragh disembark from a train from New York City accompanied by two others. Maragh's two companions went to make a phone call and then returned to Maragh who, after making eye contact with one of the officers, Detective Beard, started walking toward the exit.

At this point Detective Beard decided to interview Maragh. According to Beard's testimony, which the District Court apparently credited, Beard approached Maragh from his left rear, "stopped at an oblique angle to [Maragh] (so that [Maragh] could continue walking if he chose to do so)," displayed an identification folder, told Maragh that he was a police officer, and asked Maragh if he could speak to him. See Maragh, 695 F. Supp. at 1224. Meanwhile another officer, Detective Hanson, "positioned himself as a back-up," while a third, Detective Cassidy "left the immediate area to position himself at the exit," and was presumably not visible to Maragh. See id.

Upon Beard's request, Maragh agreed to talk to him. Beard asked whence Maragh had travelled, and received the answer "New York." He asked to see Maragh's ticket, which Maragh produced. He also ascertained that Maragh had lived in Washington for four months and had just been in New York for about four days. Beard then explained that he was a narcotics officer (he had earlier identified himself only as an officer) seeking to intercept drugs coming into Washington. Beard asked whether Maragh had drugs in his shoulder bag; Maragh said that he did not. Beard asked whether he might look into Maragh's bag, and Maragh agreed. The search revealed drugs, and, at Beard's signal, Detective Hanson approached and arrested Maragh.1

Maragh moved to suppress the evidence that Beard discovered in Maragh's bag. After a hearing on the motion, the District Court filed an opinion that began by focusing on the question "[w]hether an officer's commencement of a conversation with a person in a public place by identifying himself or herself as a narcotics officer constitutes a 'stop' implicating the Fourth Amendment." Maragh, 695 F. Supp. at 1224. The court asserted that this question had "not apparently been addressed by our Court of Appeals" (as of September 26, 1988, the date the District Court filed its opinion). See id. Relying on cases from the Fifth and Seventh Circuits, the court concluded that:

Beard's questioning of defendant turned into an investigative stop when the Detective identified himself as member of the Narcotics Branch whose purpose it was to stop drugs from coming into Washington, stood obliquely in front of him, with Detective Hanson behind him and Detective Cassidy "downfield", and asked to search defendant's bag. While there is no testimony as to whether defendant knew of the back up and blocking positions of the other officers, a reasonable person of defendant's years could have sensed the maneuvers and the presences, as well as the in terrorem effect of Detective Beard's identification of himself as a narcotics officer after defendant had denied possession of drugs.

Maragh, 695 F. Supp. at 1225; see also id. at 1224-25 (quoting United States v. Gonzales, 842 F.2d 748, 752 (5th Cir.1988) and United States v. Palen, 793 F.2d 853, 857 (7th Cir.1986)).

After concluding that the officers had stopped Maragh without the constitutionally required reasonable suspicion, the District Court considered "whether, even though the stop was unlawful, the subsequent search was voluntary because defendant was free to refuse to agree to permit the search of his bag, and was indeed free to leave." Maragh, 695 F. Supp. at 1225 (citation omitted). The judge's answer was that "the same threatening environment that converted the conversations into a 'stop' vitiates any claim that a reasonable person in the circumstances of defendant would feel free to refuse the request to search his bag or feel free to leave." Id. Based on these two conclusions--that the officers stopped Maragh unlawfully and that his "consent" to the search was not voluntary--the District Court granted Maragh's motion to suppress. The Government's appeal followed.

A. The Standard of Review on the "Seizure" Question

We examine the District Court's judgment that the police seized Maragh under a de novo standard of review, as firmly entrenched doctrine in this court and the Supreme Court requires. See, e.g., United States v. Mendenhall, 446 U.S. 544, 554-55, 100 S. Ct. 1870, 1877-78, 64 L. Ed. 2d 497 (1980) (Opinion of Stewart, J.) (describing circumstances in absence of which "otherwise inoffensive contact between a member of the public and the police cannot, as a matter of law, amount to a seizure of that person " (emphasis added)).

Contrary to the suggestion of our dissenting colleague, we can find no Supreme Court case that has ever held that the "seizure" inquiry is other than a question of law. Furthermore, the Court has never deferred to the trier of fact regarding the question of seizure. In Florida v. Royer, 460 U.S. 491, 103 S. Ct. 1319, 75 L. Ed. 2d 229 (1983), by way of one plurality opinion, two concurrences, and two dissents, eight members of the Court adopted Mendenhall 's legal test for "seizure." Justice White's plurality opinion applied the test without relying on or deferring to the district court regarding the question of seizure. See id. at 501-02, 103 S.Ct. at 1326-27. Similarly, in Michigan v. Chesternut, 486 U.S. 567, 108 S. Ct. 1975, 100 L. Ed. 2d 565 (1988), the Court iterated its support for Mendenhall 's objective legal test for seizure. See id. 108 S.Ct. at 1979 (discussing evolution of seizure test from Terry to Mendenhall, and noting that "[t]he Court has since embraced [Mendenhall 's] test"). As in Royer, the Chesternut Court applied the test without any deference to the district court on the question of seizure. See id. 108 S.Ct. at 1980 ("Applying the Court's test to the facts of this case, we conclude that respondent was not seized...." (emphasis added)).

Additionally, the soundest of jurisprudential considerations compel appellate courts not to shirk their responsibility independently to apply important constitutional standards. In the Fourth Amendment context, as in the First Amendment setting, appellate judges have "a constitutional responsibility that cannot be delegated to the trier of fact." See Bose Corp. v. Consumers Union, Inc., 466 U.S. 485, 501, 104 S. Ct. 1949, 1959, 80 L. Ed. 2d 502 (1984). It does not matter that Mendenhall requires the courts to consider whether a reasonable person would have felt free to leave "in view of all the circumstances surrounding the incident." Mendenhall, 446 U.S. at 554, 100 S.Ct. at 1877. Although Mendenhall 's seizure test "is necessarily imprecise" and "flexible enough to be applied to the whole range of police conduct in an equally broad range of settings, it calls for consistent application from one police encounter to the next, regardless of the particular individual's response to the actions of the police." See Chesternut, 108 S. Ct. at 1979-80 (emphasis added). De novo review helps to ensure "consistent application."

In Mendenhall, the Supreme Court established a test to determine whether a person has been "seized" for purposes of the Fourth Amendment. "[T]he police can be said to have seized an individual 'only if, in view of all of the circumstances surrounding the incident, a reasonable person would have believed that he was not free to leave.' " Michigan v. Chesternut, 486 U.S. 567, 108 S. Ct. 1975, 1979, 100 L. Ed. 2d 565 (1988) (quoting Mendenhall, 446 U.S. at 554, 100 S.Ct. at 1877 (Opinion of Stewart, J.)). This court has firmly embraced that test. See, e.g., United States v. Savage, 889 F.2d 1113, 1116 (D.C.Cir.1989); United States v. Baskin, 886 F.2d 383, 386 (D.C.Cir.1989); United States v. Carrasquillo, 877 F.2d 73, 76 (D.C.Cir.1989); United States v. Lloyd, 868 F.2d 447, 450 (D.C.Cir.1989). Moreover, police officers "do not violate the Fourth Amendment by merely approaching an individual on the street or in another public place, by asking him if he is willing to answer some question [or] by putting questions to him if he is willing to listen." Royer, 460 U.S. at 497, 103 S.Ct. at 1324 (plurality); see Carrasquillo, 877 F.2d at 76; Lloyd, 868 F.2d at 451. "Nor would the fact that the officer identifies himself as a police officer, without more, convert the encounter into a seizure requiring some kind of objective justification." Royer, 460 U.S. at 497, 103 S.Ct. at 1324 (plurality).

In determining whether police have seized a citizen, this court has consistently looked to such factors as whether the officers displayed weapons, were physically intimidating or threatening, or approached the citizen at an unusual place or time. See, e.g., Baskin, 886 F.2d at 386-87; Lloyd, 868 F.2d at 450; United States v. Brady, 842 F.2d 1313, 1314 (D.C.Cir.1988). According to the District Court's findings, in this case Maragh was aware of one plainclothes police officer, who approached him in mid-afternoon, in a public place, displayed no weapons, and did not block his path.

This case is on all fours with relevant parts of Baskin, a case decided too late for the District Court below to have benefitted from its binding guidance. In Baskin, two police officers approached a passenger who had disembarked from a train, identified themselves as officers, asked whether they could talk to the passenger, to which he replied "sure." After asking a few questions about the passenger's ticket and identification, the officers told the passenger that they were members of a narcotics squad, and asked whether a dog they had with them could sniff the passenger's bag. The passenger agreed to both a dog sniff and a hand search by the officers, each of which revealed drugs, for which the police arrested the passenger. Citing the Chesternut and Mendenhall test for seizure noted above, this court concluded that the police did not seize Baskin, as a matter of law, until they arrested him. See Baskin, 886 F.2d at 386-87.

In light of cases such as Baskin, we must conclude that the District Court's judgment--that there was a stop in the totality of this case's circumstances--was wrong as a matter of law. The crux of that judgment has two identifiable components. One was the District Court's assertion that "a reasonable person of defendant's years could have sensed the maneuvers and the presences" of detectives Hanson (who was several feet behind Maragh, see Tr. 42, 46), and Cassidy (who was at the exit). See Maragh, 695 F. Supp. at 1225. Another was the court's conclusion that "Detective Beard's identification of himself as a narcotics officer after defendant had denied possession of drugs" had an "in terrorem " effect on Maragh. See id.

The first of these factors is clearly not among this circuit's criteria for seizure. As noted above, in Baskin this court found no Fourth Amendment infirmity even where two officers approached the defendant, identified themselves as officers, received permission to talk to the defendant, asked to examine his ticket, did so, and then told the defendant that they were narcotics officers and asked to search the defendant's bag. See Baskin, 886 F.2d at 386-87. If two officers could directly approach the defendant in Baskin, it is hard to see how we could reach a different result here, where only one officer approached Maragh, even while he may have "sensed the maneuvers and presences" of other officers standing several feet away.

Baskin is similarly dispositive of the second factor central to the District Court's conclusion.2 Baskin and other cases have rejected the proposition that a consensual encounter becomes an investigative stop at the moment an officer identifies herself as a narcotics officer. See Baskin, 886 F.2d at 386-87; Lloyd, 868 F.2d at 449, 451 (police-citizen encounter was not seizure even though officer identified himself as member of drug interdiction unit); Mendenhall, 446 U.S. at 547-48, 555, 100 S. Ct. at 1873, 1877 (Opinion of Stewart, J.) (respondent should have felt free to break off exchange with officers even though they had identified themselves as "federal narcotics agent[s]"). The District Court relied on cases from other circuits which came to a contrary conclusion. See Maragh, 695 F. Supp. at 1224-25. While that analysis of Fourth Amendment seizure may have much to be said for it, it is inconsistent with the law of this circuit.

The District Court also concluded that, in allowing the officers to search his bag, Maragh behaved involuntarily. The District Court reached this conclusion on the basis of a mistaken legal determination that the officers improperly stopped Maragh. See Maragh, 695 F. Supp. at 1225. Because the District Court's mistake of law infected its factual finding regarding consent, we are constrained to reverse and remand for further findings on that factual question. See Pullman-Standard v. Swint, 456 U.S. 273, 292, 102 S. Ct. 1781, 1792, 72 L. Ed. 2d 66 (1982) (stating "elementary" proposition that "where findings are infirm because of an erroneous view of the law, a remand is the proper course unless the record permits only one resolution of the factual issue").

The District Court's analysis of the consent issue appears to be mistaken in two respects. First, the court apparently proceeded under the view that consent could cure an unlawful seizure. See Maragh, 695 F. Supp. at 1225. This is an erroneous statement of law. See Wong Sun v. United States, 371 U.S. 471, 83 S. Ct. 407, 9 L. Ed. 2d 441 (1963). Second, in addition to giving decisive weight to the mistaken conclusion that the officers unlawfully seized Maragh, the District Court also apparently treated the tests for seizure and voluntary consent as identical. Although there is overlap in these tests, they are not identical. See Schneckloth v. Bustamonte, 412 U.S. 218, 248-49, 93 S. Ct. 2041, 2058-59, 36 L. Ed. 2d 854 (1973) ("Voluntariness is a question of fact to be determined from all the circumstances...."). Indeed, it is because we cannot say, on this record, that a finding of involuntary consent was clearly erroneous--even though we have already concluded that a reasonable person in Maragh's position would have felt free to leave--that we are constrained to remand rather than simply reverse.

In assessing voluntariness, the District Court must consider factors such as those articulated by the Supreme Court in Schneckloth v. Bustamonte, 412 U.S. 218, 226, 93 S. Ct. 2041, 2047, 36 L. Ed. 2d 854 (1973), and United States v. Mendenhall, 446 U.S. 544, 558, 100 S. Ct. 1870, 1879, 64 L. Ed. 2d 497 (1980).

The District Court granted Maragh's motion to suppress because the court concluded that the police seized Maragh without adequate justification. Because we hold that the police did not seize Maragh, and because the District Court's findings on consent are infirm due to erroneous views of law, we reverse the suppression order and remand for further proceedings consistent with this opinion.

Recently this court has been deluged with a spate of criminal appeals concerning whether police encounters constitute fourth amendment seizures. See, e.g., United States v. Tavolacci, 895 F.2d 1423- (D.C.Cir. January p, 1990);United States v. Winston, 892 F.2d 112 (D.C.Cir.1989); United States v. Savage, 889 F.2d 1113 (D.C.Cir.1989); United States v. Baskin, 886 F.2d 383 (D.C.Cir.1989); United States v. Carrasquillo, 877 F.2d 73 (D.C.Cir.1989); United States v. Battista, 876 F.2d 201 (D.C.Cir.1989); United States v. Lloyd, 868 F.2d 447 (D.C.Cir.1989). These cases present varying permutations of the now familiar initial interview between the police detective, frequently a drug enforcement officer, and the potential suspect, often a traveller encountered in or near a transportation terminal. As these cases proliferate, it has become increasingly difficult for this court to apply the "reasonable person" standard articulated in United States v. Mendenhall, 446 U.S. 544, 100 S. Ct. 1870, 64 L. Ed. 2d 497 (1980), to evaluate whether there has been a show of authority sufficient to constitute a seizure. In Mendenhall, Justice Stewart stated that a person is seized within the meaning of the fourth amendment "only if, in view of all of the circumstances surrounding the incident, a reasonable person would have believed that he was not free to leave." Id. at 554, 100 S. Ct. at 1877; see also Michigan v. Chesternut, 486 U.S. 567, 108 S. Ct. 1975, 1979, 100 L. Ed. 2d 565 (1988) (unanimously embracing the Mendenhall test). The increasingly unwieldy body of precedent on this matter tempts us to develop new legal rules with each novel factual circumstance to bring those factual circumstances within the established line of precedents sanctioning consensual police interviews. Alternatively, the precedents encourage us to attach talismanic legal significance to those factual details which have recurred in cases deemed non-coercive--for example, a conversational tone by the inquiring officer, the wearing of plainclothes, or the absence of exposed weapons. I think this approach to the objectives informing the Mendenhall test gets us into very troubled waters.

In Mendenhall, Justice Stewart made it clear that the question of whether a seizure has occurred was one to be evaluated "in view of all the circumstances surrounding the incident" in each individual case. 446 U.S. 544, 554, 100 S. Ct. 1870, 1877. While he proceeded to catalogue "[e]xamples of circumstances that might indicate a seizure,"," his analysis reflects the wholistic, fact-based nature of the inquiry:

On the facts of this case, no "seizure" of the respondent occurred. The events took place in a public concourse. The agents wore no uniforms and displayed no weapons. They did not summon the respondent to their presence, but instead approached her and identified themselves as federal agents. They requested but did not demand to see the respondent's identification and ticket.... In short, nothing in the record suggests that the respondent had any objective reason to believe that she was not free to end the conversation in the concourse and proceed on her way.

Id. at 555, 100 S.Ct. at 1877 (emphasis added). The Supreme Court recently stressed that the Mendenhall test is meant to allow courts to evaluate the particularized context of each case. "The test is necessarily imprecise, because it is designed to assess the coercive effect of police conduct, taken as a whole, rather than to focus on particular details of that conduct in isolation. Moreover, what constitutes a restraint on liberty prompting a person to conclude that he is not free to 'leave' will vary, not only with the particular police conduct at issue, but also with the setting in which the conduct occurs." Michigan v. Chesternut, 108 S. Ct. at 1979 (emphasis added).

Applying those teachings to this case, I cannot understand why we disregard the trial judge's premier role in this difficult evaluation of the constitutional claims. There is no dispute that if the officers' initial encounter with Maragh amounted to a seizure, it would not be supported by the reasonable suspicion necessary to justify a Terry stop. Notwithstanding the trial judge's findings, the majority concludes that no seizure occurred.

After a suppression hearing at which the arresting officer and the defendant both testified, the trial judge concluded that the officer's "questioning of [the] defendant turned into an investigative stop when [Detective Beard] identified himself as a member of the Narcotics Branch whose purpose it was to stop drugs from coming into Washington, stood obliquely in front of him, with Detective Hanson behind him and Detective Cassidy 'downfield,' and asked to search the defendant's bag." 695 F. Supp. 1223, 1225 (D.D.C.1988). In reversing the finding of the trial court, this court substitutes its judgment for that of the trial judge who was in the best position to assess witness credibility and to evaluate whether the totality of the circumstances indicated that the defendant had been seized. Because a determination of whether a seizure has occurred is essentially a fact-based endeavor, appellate courts should not reverse the trial court's conclusion unless it is clearly erroneous. The record contains more than sufficient facts to support a finding that a seizure occurred in this case. I dissent from the majority's reversal of this finding.

Several of our sister circuits have recognized explicitly that the question of whether the police "seized" a person is essentially factual. In this vein they have espoused the "clearly erroneous" standard of review, which attends all district court findings of historical fact, when reviewing fourth amendment seizure determinations. See United States v. Rose, 889 F.2d 1490 (6th Cir.1989); United States v. Gray, 883 F.2d 320, 322 (4th Cir.1989); United States v. Teslim, 869 F.2d 316, 321 (7th Cir.1989); United States v. Archer, 840 F.2d 567, 571 (8th Cir.1988), cert. denied, --- U.S. ----, 109 S. Ct. 364, 102 L. Ed. 2d 354 (1988); United States v. Erwin, 803 F.2d 1505, 1508 (9th Cir.1982); United States v. Gooding, 695 F.2d 78, 82 (4th Cir.1982); United States v. Patino, 649 F.2d 724, 728 (9th Cir.1981). The factual question of whether a seizure has occurred should not be confused with the legal conclusion of whether the seizure was lawful. See, e.g., United States v. Mines, 883 F.2d 801, 803 (9th Cir.1989) ("findings of fact" reviewed under clearly erroneous standard, while "[t]he ultimate conclusion of the lawfulness of a seizure" reviewed "de novo"); United States v. Campbell, 843 F.2d 1089, 1092 (8th Cir.1988) (same).

Contrary to the suggestion of the majority, the Supreme Court has not explicitly identified the trial court's determination of whether a fourth amendment seizure has occurred as a finding of fact, law, or a mixed determination of law and fact. As noted, in Mendenhall Justice Stewart acknowledged that the relevant findings are contextual, citing for example "the use of language or tone of voice indicating that compliance with the officer's request might be compelled." 446 U.S. 544, 554, 100 S. Ct. 1870, 1877. In this same plurality opinion, joined only by Justice Rehnquist, Justice Stewart also stated that "the correctness of the legal characterization of the facts appearing in the record is a matter for this Court to determine." Id. at 553 n. 5, 100 S. Ct. at 1876 n. 5. Elsewhere, Justice Powell characterized the seizure analysis under Mendenhall as "a difficult characterization of fact and law." I.N.S. v. Delgado, 466 U.S. 210, 221, 104 S. Ct. 1758, 1765, 80 L. Ed. 2d 247 (1984) (Powell, J., concurring). In embracing Justice Stewart's Mendenhall test, the Supreme Court has not made any attempt to state its collective position on this issue. Perhaps as a result of this ambiguity, this circuit has never expressly articulated its position on the standard of review applicable to a district court's characterization of the facts under the Mendenhall test. Other circuits bound by the same Supreme Court precedents which bind this court have not felt restrained from characterizing the Mendenhall test as a factual inquiry.

Based on a review of the recent seizure decisions of this court, I believe that it would be appropriate to join those of our sister circuits that have adopted a clearly erroneous standard. A simple comparison of two recent decisions amply demonstrates the virtue of this course. In United States v. Battista, this court found that a seizure occurred where an initial interview took place in a train roomette and "the convergence of several subtle factors" suggested that a similarly situated reasonable person would not have felt free to leave. See Battista, 876 F.2d at 204-05 (officers knocked on door to awaken suspect at 6:30 a.m., questioned suspect while he was partially clothed, and retained his identification). In contrast, in United States v. Savage, 889 F.2d 1113, a police encounter in a train roomette initially did not constitute a seizure although it later was transformed into one by the police detective's direct and forceful questioning. In reaching such a result, both courts necessarily relied on the demeanor interpretations of the trial judge. As to the first stage of the interview, the Savage court opined: "[N]othing in the record indicates that Savage was peculiarly vulnerable as would be a man awakened from bed in the early morning. [The detective] testified, and the district court credited, that his voice and demeanor while talking to Savage were 'low key and informal.' " 889 F.2d at 1117. The trial court's finding that the detective's questioning later became " 'direct and probably forceful,' " however, led the Savage court to conclude that a seizure occurred at that point. Id. at 1115, 1117.

The demeanor findings which proved critical in Savage are the kinds of inferences which only a trial judge can draw. Similarly, in Battista it was the total effect of certain "subtle factors" that weighed in favor of finding a seizure. Such factual inferences typically are gleaned by the trial judge from the testimony of police detectives at the suppression hearing. See, e.g., United States v. Gray, 883 F.2d at 321 ("The evidence at the suppression hearing consisted almost entirely of the testimony of Special Agent Johnson."); United States v. Erwin, 803 F.2d at 1508 ("Only troopers Bowman and McMillon testified at the hearing, and their testimony suggests that they requested Erwin's cooperation and that he freely complied.").

More importantly, the suppression hearing is purely evidentiary in nature. In order to determine when and whether a seizure occurred, the trial judge must resolve conflicts in testimony, assess credibility, and draw conclusions as to what impact the total factual circumstances of the encounter would have on a similarly situated "reasonable" person. The Ninth Circuit stated the proposition as follows:

The determination of when [contact between the police and a member of the general public] constitutes a seizure within the meaning of the fourth amendment depends upon the facts and circumstances of each case. Proper deference must be given to the district judge who heard the testimony of the officer, his tone of voice and inflection, and who observed the officer's conduct on the stand, his appearance and mannerisms.

United States v. Patino, 649 F.2d at 728. Similarly, the Seventh Circuit reasoned that the "highly factual" nature of the Mendenhall test " 'requires that particular deference be given to the trial judge who had the opportunity to observe the testimony and demeanor of [the witnesses].' " United States v. Teslim, 869 F.2d at 321 (citation omitted).

In adopting the clearly erroneous standard, the Fourth Circuit recognized that differing factual inferences might be drawn from the surrounding circumstances in assessing whether a citizen has been effectively "restrained" by a "show of authority." United States v. Gooding, 695 F.2d at 82. The Seventh Circuit has also upheld a trial court's finding of a seizure even where the appellate court "might have characterized the moment of seizure differently." United States v. Teslim, 869 F.2d at 321. This seems an eminently sensible approach where subtle factors of demeanor have the potential to be critical to the outcome.

The majority suggests that by adopting a clearly erroneous standard, this court would be abdicating its responsibility independently to apply important fourth amendment standards. Yet, this court has applied the same kind of objective approach to determinations of abandonment of property for fourth amendment purposes:

To determine whether there is an abandonment in the fourth amendment sense, the district court must focus on the intent of the person who is alleged to have abandoned the place or object. The test is an objective one, and intent may be inferred from "words spoken, acts done, and other objective facts." Because the ultimate determination hinges on the outcome of a factual inquiry into intent, a finding of abandonment is reviewed under the clearly erroneous standard.... see also United States v. Kendall, 655 F.2d 199, 203 (9th Cir.1981), cert. denied sub nom. Akers v. United States, 455 U.S. 941, 102 S. Ct. 1434, 71 L. Ed. 2d 652 (1982) ("even where arguably mixed with questions of law, [the trial court's finding of abandonment] is subject to attack only if clearly erroneous.")

United States v. Thomas, 864 F.2d 843, 846 (D.C.Cir.1989) (emphasis added) (other citations omitted). The fact that the abandonment inquiry focuses on intent does not render such an approach any less appropriate to meet the standards of Mendenhall. The abandonment inquiry attempts to discern as an objective matter whether a person has forfeited his or her reasonable expectation of privacy in the object. The seizure inquiry attempts to discern as an objective matter whether a person has been restrained.

The majority also suggests that a de novo review standard would ensure consistent application of the Mendenhall test. Yet the Supreme Court's own application of that test in two similar factual circumstances produced divergent results. Compare Florida v. Royer, 460 U.S. 491, 103 S. Ct. 1319, 75 L. Ed. 2d 229 (1983) with Mendenhall, 446 U.S. 544, 100 S. Ct. 1870. The only material difference in the factors considered by the Supreme Court in those two cases was the retention of Royer's ticket in the former case. Royer, 460 U.S. at 503-04, 103 S.Ct. at 1327-28. If by "consistent application," the majority means that the single factor of the retention of an individual's ticket is to be converted into a rule of law that a seizure has occurred, that is certainly not what the Supreme Court intended in embracing a "totality of the circumstances" test. Yet that is what many lower courts have done. See Butterfoss, Bright Line Seizures: The Need for Clarity in Determining When Fourth Amendment Activity Begins, 79 J.Crim.L. & Criminology 437, 460 & n. 123 (1988).

In the case sub judice we are again faced with subtle questions. The trial judge's conclusion that the "threatening environment" converted the conversation into a Terry stop was influenced by several factors, including Detective Beard's "commanding presence," the number and posture of the officers involved, and the manner of questioning (e.g. asking to search the defendant's bag after he had denied possession of drugs). The trial judge concluded that the combination of these several factors would lead a reasonable person to feel as though he were not free to leave. In addition, the trial court heard the conflicting testimony of Officer Beard and appellant Maragh as to whether the officers touched Maragh before arresting him and as to the precise nature of the encounter. Clearly, if we had had the benefit of observing the interview firsthand, our perspectives would be dramatically enhanced. This situation demonstrates the critical role of the trier of fact in ascertaining whether, as an objective matter, the totality of factual circumstances amounts to a coercive show of authority. As the Fourth Circuit has observed, "[i]n reviewing fourth amendment seizure determinations, the appellate courts have recognized that the determination of precisely when an officer's 'polite request for an interview' in fact becomes a fourth amendment seizure is 'not always an easy one,' may be 'extremely close,' and calls for a 'refined judgment' by the trial court." United States v. Gray, 883 F.2d at 322 (citations omitted).

In light of the particularized nature of the Mendenhall inquiry, reference to the outcome of other cases with similar facts can and does lead the appellate court astray. In explaining the operation of another fourth amendment "totality of the circumstances" test, the Supreme Court warned against undue reliance on precedents addressing the question of consent in criminal cases. The body of over 30 different Supreme Court cases on consent "yield[s] no talismanic definition of 'voluntariness,' mechanically applicable to the host of situations where the question has arisen." Schneckloth v. Bustamonte, 412 U.S. 218, 224, 93 S. Ct. 2041, 2046, 36 L. Ed. 2d 854 (1972). This body of Supreme Court precedent has produced a long list of factors which the Court has considered to assess whether, in the totality of all the surrounding circumstances, a particular defendant's will was overborne. Id. at 226, 93 S.Ct. at 2047 (citing, inter alia, youth, lack of education, low intelligence, length of detention, and deprivation of food or sleep). Yet, "[t]he significant fact about all of these decisions is that none of them turned on the presence or absence of a single controlling criterion; each reflected a careful scrutiny of all of the surrounding circumstances." Id.

Unfortunately, by failing to treat the Mendenhall test as a purely factual inquiry, the absence of certain factors in a suppression case has come dangerously close to being a talisman in this circuit for a finding that there was no seizure. In United States v. Brady, 842 F.2d 1313, 1314 (D.C.Cir.1988), we concluded that the defendant "faced none of the factors typically found to intimidate persons into thinking that compliance is obligatory: no visible weapons; no physical intimidation; no threats; no unusual setting or time." These "Brady factors" have become a frequent mantra in our seizure jurisprudence. See, e.g., Tavolacci, --- F.2d at ----; Winston, 892 F.2d at 115; Savage, 889 F.2d at 1116; Baskin, 886 F.2d at 386-87; Carrasquillo, 877 F.2d at 76; Lloyd, 868 F.2d at 450. They illustrate why adopting a clearly erroneous standard would better accord with the spirit of the "totality of the circumstances" analysis required by Mendenhall. In Savage we stated that despite the "totality" approach, "[i]t is nevertheless instructive to note instances of particular police behavior that do not, by themselves constitute seizures." At 1117 (emphasis in original). Yet, focus on singular aspects of police conduct is precisely the kind of analysis that the Mendenhall test is designed to avoid. Rather than relying on legal precedents to reach what should be a factual conclusion, appellate courts should rely on the factual inferences of the trial judge. Our decisions seem to reflect more willingness to support a trial judge's findings against coercion than a judge's findings in favor of the defendant's claims. We ill-serve the standard of Mendenhall by applying a "sometimes" deferential standard of review.

In this case the trial court found that the circumstances surrounding the initial questioning of appellant Maragh constituted a seizure and that the search of his bag was non-consensual. Under the clearly erroneous standard, the district court's decision must be affirmed unless upon considering the entire record we are left with a definite and firm conviction that a mistake has been made. United States v. United States Gypsum Co., 333 U.S. 364, 395, 68 S. Ct. 525, 541, 92 L. Ed. 746 (1948). Because the trial court is best suited to assess the subtle factors at issue in this case, and because nothing in the record suggests that the trial court clearly erred in its conclusions, I would affirm its decision.