I respectfully dissent because I disagree with the majority on two points: first, its treatment of the factual findings made by the district court and, second, its treatment of the circuit split regarding whether a suspect must do more than merely pause briefly in order to be seized within the meaning of the Fourth Amendment.
I
First, I do not agree that the facts of the case before us today are comparable to the situation in United States v. Baldwin, 496 F.3d 215 (2d. Cir. 2007).
Baldwin stands clearly for one legal proposition and arguably for a second legal proposition. First, the opinion states that “a suspect must do more than halt temporarily” in order “to comply with an order to stop[ ]and thus to become seized.” Id. at 218. Second, Baldwin suggests, in a single sentence, that “evasion of police authority,” which I take to mean conduct that is part of a suspect’s plan to flee from custody, will not constitute submission. Id. at 219.
The facts found by the district court do not support a decision under either of these rules. Huertas did not simply “halt momentarily” in this case, but instead stopped and answered some of the officer’s questions. See United States v. Huertas, No. 3:14cr141, 2015 WL 1517403, at *1. The majority pins a great deal on the district court’s statement that Huertas “remained still” when the officer approached, United States v. Huertas, No. 3:14-cr-141, 2015 WL 1517403, at *1, and that Huer-tas’s brief states that he remained in a “‘fixed’ position” through the encounter. Appellant’s Opening Br. at 17. The majority takes this to mean Huertas did not halt after having been walking previously. I doubt it matters, for purposes of the Fourth Amendment, whether, someone had previously been walking and then stopped, or instead had previously been standing still and continued to do so while engaging with an officer. I doubt even more, however, that the district court crisply distinguished between these two possibilities, given that the court elsewhere referred to the incident as a “stop,” 2015 WL 1517403, at *3 n.l, applied a rule dealing with cases where suspects “halt temporarily,” id. at *2, and, as the majority notes, accepted Huertas’s argument that he “briefly stopped walking before he ran,” id. In view of both Huertas’s having stopped, and the fact that he answered the officer’s questions, I would hold that the limited “momentary halt” rule in Baldwin simply does not cover the situation here.
Moreover, to the extent that Baldwin contains an additional “anti-evasion” principle, the principle seems to be limited to situations where, as in Baldwin, an entire course of conduct is undertaken in an effort to flee from the police. In Baldwin, the suspect stopped his car in order to lure officers out of their own vehicles, thus giving himself an advantage in the ensuing car chase. Baldwin, 496 F.3d at 217. No other explanation existed for the suspect’s conduct, since he never spoke to the officers or otherwise engaged with them. In this case, on the other hand, the district court made no factual finding that Huertas stopped and answered questions as part of a plan to flee from the police. Huertas, 2015 WL 1517403, at *1-*2. To the contrary, the district court accepted Huertas’s version of events for purposes of resolving the “submission” dispute, and Huertas’s version of events involved a genuine submission to the officer’s authority, followed by a change of heart and a decision to flee. Id. at *2 (“According to Mr. Huertas ... he submitted to Officer Lattanzio’s authority by pausing before running and by beginning to answer the officer’s questions. *220However, even if Mr. Huertas briefly stopped walking before he ran as he contends, given the totality of circumstances here, that brief stop and verbal exchange did not constitute ‘submission^]’ ”).
An important distinction exists between initial, earnest submission followed by later flight, as opposed to an entire course of conduct undertaken to ensure a getaway. See, e.g., United States v. Brodie, 742 F.3d 1058, 1061 (D.C. Cir. 2014) (“Later acts of noncompliance do not negate a defendant’s initial submission, so long as it was authentic.”); United States v. Valentine, 232 F.3d 350, 359 (3d Cir. 2000) (“Under some circumstances[,] we have held that a defendant was seized despite his subsequent flight.”).
The majority attempts to establish Huertas’s “evasion” in two ways. First, it suggests that Huertas did indeed stop in order to improve his chances of getting away. Op. at 216. Nothing in the district court’s factual findings, however, supports this view. Any statement that Huertas engaged with the officer'in order to improve his chance' of escape would require us to find new facts about his mental state—a type of fact-finding that appellate courts such as this one are ill-situated to conduct.
Second, the majority extends the definition of “evasion” well beyond activity intended to slpw down pursuing officers. The majority states that Huertas’s' conduct was undertaken to “quiet suspicion and hope that Officer Lattanzio would drive away after being satisfied with answers to his questions,” and that Huertas thereby intended to “evade” the police. Op. at 217. The majority thus adopts the view that answering questions to clear one’s name counts, as “evasion” just as much as does pretending to submit so that officers put themselves in a worse position for an impending chase.
The consequences of eliding this distinction are far-reaching. Suppose, for example, that a suspect speaks with the police not for one or two minutes, but for an hour or two, because he thinks he can talk his way out of going to jail. Would we say he had not submitted, since his only hope was that the interview would “quiet suspicion” and that the officer would let him go “after being satisfied with answers to his questions”? I am comfortable asserting that the vast majority of criminal suspects engage with the police only when they think they will avoid incarceration by doing so. Under that assumption, the majority’s position suggests that stopping to speak with the police, even at length, is unlikely to constitute a seizure because it instead will constitute evasion..
Accordingly, I cannot agree that Huer-tas’s conduct falls within Baldwin’s “momentary. halting” rule. Moreover, the district court did, not find that Huertas’s entire course of conduct was part of a plan to flee the police, and thus this case does not fit within Baldwin’s anti-evasion rule. To the extent that the majority’s position would extend Baldwin’s anti-evasion rule to cover any action taken by a suspect to “quiet suspicion” in the hope that an officer “would drive away after being satisfied,” it sweeps far too broadly-
II
Second, the majority embraces the wrong side of a deepening split between the circuits regarding whether a suspect must do more than merely pause briefly in order to be sfeized within the meaning of the Fourth Amendment.
The emergent view in the Courts of Appeals, although admittedly uneven within the circuits, is that when a suspect does nearly anything more than pausing briefly, *221including any significant verbal .engagement with the officer, that action is strong evidence of submission.1 See United States v. Camacho, 661 F.3d 718, 726 (1st Cir. 2011) (holding that suspect may “submit[ ] to [the officer’s] show of authority by responding to his questions”); compare Brodie, 742 F.3d at 1061 (holding that suspect’s “action—putting his hands on the car when told to do so by the police— [constituted] full compliance with the officer’s request,” and that the suspect was thus seized by police despite his later attempt to flee), and United States v. Hernandez, 27 F.3d 1403, 1407 (9th Cir. 1994) (holding that hesitation and eye contact, without more, did not constitute submission), and United States v. Morgan, 936 F.2d 1561, 1565, 1566-67 (10th Cir. 1991) (holding that suspect submitted wh¿n officer told him to “hold up” and the suspect replied “What do you want?” before fleeing), with Valentine, 232 F.3d at 359 (holding that “[e]ven if [suspect] paused for a few moments and gave his name, he did not submit in any realistic sense” and thus there was no Fourth Amendment seizure).
Actions more substantial than momentary hesitation, including answering questions, should be considered strong signs of submission. First, courts have recognized that almost any affirmative physical actions suggesting engagement with the officer manifest an intention to submit to authority. See Brodie, 742 F.3d at 1061. It is not evident why there should be a distinction between affirmative physical actions and affirmative non-physical ones, such as answering questions, if they manifest the same intention.
Second, the principle that seizure requires submission originated only recently in California v. Hodari D., 499 U.S. 621, 111 S.Ct. 1547, 113 L.Ed.2d 690 (1991), and must be limited in reach so that it does not eviscerate Fourth Amendment jurisprudence. Hodari dealt with a group of men who immediately took flight upon seeing police officers, and the Supreme Court held that they were not “seized” when they ran away. Id. at 622-23, 629, 111 S.Ct. 1547. That reasoning is workable as far as it goes, but it cannot go very far. For as Justice Stevens noted in dissent, the Supreme Court has in the past fifty years taken an expansive view of what constitutes a “seizure,” id. at 632, 111 S.Ct. 1547, and has written, for example, that a suspect “may not be detained even momentarily without reasonable, objective grounds for doing so,” id. at 640, 111 S.Ct. 1547 (emphasis in original). As even “momentary] detention” may constitute a seizure, we must be careful not to remove constitutional protections surrounding brief seizures even where suspects later flee.
The rules surrounding police investigations have become a veritable minefield for the.unwary. The. majority opinion further complicates and impairs the constitutional protections afforded to persons facing police questioning, and will increase uncer*222tainty about protections applicable during the course of investigations. In so doing, the majority joins the wrong side of a deepening split between the circuits over this important issue.
Accordingly, I respectfully dissent.
. Acknowledging that consensus reveals certain guideposts of submission, including verbal engagement with an officer, is consistent with the “totality of the circumstances" standard governing whether a suspect has submitted, Such guideposts, which are sound generalizations applicable in the majority of cases, appear in other areas governed by totality-of-the-circumstances tests. See, e.g., Brower v. Cty. of Inyo, 489 U.S. 593, 598-99, 109 S.Ct. 1378, 103 L.Ed.2d 628 (1989) (noting that the use of a roadblock generally establishes seizure); United States v. Martinez-Fuerte, 428 U.S. 543, 556, 96 S.Ct. 3074, 49 L.Ed.2d 1116 (1976) (noting that vehicle "checkpoint stops are ‘seizures' within the meaning of the Fourth Amendment”). In an unusual case— for example, where a suspect speaks with an officer over a physical barrier ' between them—the generalization that talking shows submission might well be overcome.