Rosanna MORELLI, Plaintiff, Appellant, v. Steven WEBSTER, Defendant, Appellee, 552 F.3d 12

SELYA, Circuit Judge. In this case, a prostitution sting backfired, spawning a series of events that culminated in a civil suit. That suit, brought pursuant to 42 U.S.C. § 1983, included claims of unlawful detention and use of excessive force. The district court granted

summary judgment in favor of the defendant (a police officer). See Morelli v. Webster, 554 F.Supp.2d 46 (D.Me.2008). We affirm as to the unlawful detention claims, but reverse as to the excessive force claims. I. BACKGROUND Because this is an appeal from the entry of

summary judgment, we take the facts in the light most flattering to the nonmovant (here, the plaintiff). See Cordi-Allen v. Cordon, 494 F.3d 245, 248 (1st Cir.2007). This means that where, as here, there is a wide divergence between the parties’ first-hand accounts of the relevant events, we must adopt the nonmov-ant’s version. Scott v. Harris, 550 U.S. 372, 127 S.Ct. 1769, 1774-75, 167 L.Ed.2d 686 (2007). Our narrative draws heavily upon the plaintiffs filings under D. Me. R. 7. Plaintiff-appellant Rosanna Morelli, an exotic dancer, hired out for private parties.

Though self-employed, she occasionally accepted engagements through Serena’s Heaven on Earth, a purveyor of adult entertainment services in Portland, Maine. On March 3, 2006, the plaintiff, using the nom de guerre “Vanessa,” responded to a call placed to a number advertised by Serena’s. She and the caller agreed on a price for an exotic dance to be performed in room 203 of the Best Western hotel in South Portland. Unbeknownst to the plaintiff, the caller was an undercover police officer named McVane, and the call was part of a prostitution sting operation orchestrated by the local constabulary.

The site of the proposed performance— room 203 — had been outfitted with audio and video surveillance equipment. The output was to be transmitted surreptitiously to a so-called “observation room” next door. A local prosecutor and several police officers had congregated in the observation room (among them, defendant-ap-pellee Steven Webster). McVane met the plaintiff at the door to room 203. Upon entering the room, the plaintiff hugged McVane. At that point, one of the policemen in the observation room (Officer Farris) recognized the plaintiff and exclaimed that she was a known prostitute.

Because McVane was nervous and acting strangely, the plaintiff soon began to suspect that something was amiss. She nonetheless placed her heavy coat on the bed and requested the agreed fee. McVane placed several bills on a counter but the plaintiff did not touch them. In response to McVane’s repeated urgings that she disrobe, the plaintiff informed him that she was only there to dance. When McVane persisted, she sarcastically suggested that he remove his clothes. Finally she started to shed her jeans.

At about that time, however,, she became convinced that McVane was not a bona fide customer, was probably associated with law enforcement, and was wasting her time. She pulled up her jeans, told McVane that she was leaving, and informed him that she was taking $20 to cover transportation and wasted time. She grabbed a bill in that denomination from the pile of currency on the counter and took her leave. In the corridor the plaintiff encountered several officers, including Webster. When one of them accused her of having stolen money; she surrendered the $20 bill.

She then moved toward the exit, looking straight ahead and proceeding slowly. Her coat was draped over her right arm, and she kept that arm stiff. Webster stands about a foot taller than the plaintiff — he is 6'2" and in good physical shape — and substantially outweighs her. Although clad in civilian clothing, he prominently displayed a badge on his belt. Among the cohort of policemen, only Webster attempted to stop the plaintiff from leaving the hotel. He deliberately positioned himself to block her path. As the plaintiff tells it, she tried to walk around him and brushed against him as she passed.

Webster reacted angrily, grabbing her wrist, yanking her around, slamming her against the hallway wall, pinning her there by her forearms, and saying “look missy, you’re not going anywhere, you need to go back into the room.” Webster held the plaintiff against the wall for a full three or four minutes before guiding her back to room 203. Once there, Webster told the plaintiff that he could not arrest her. Nevertheless, he forced her to remain in the room while.he questioned her. Two other officers were present during this interrogation.

The plaintiff complained that she was in pain and, after several minutes, Webster told the plaintiff that she could leave as long as she did so “respectfully.” The plaintiff found this admonition “condescendingly odd.” The plaintiff returned home. She immediately called the police department to report the incident. Later, she visited the emergency room of a local, hospital. She was told that she had sustained contusions and a first-degree shoulder separation. An orthopedic surgeon treated her for some time and, eventually, more sophisticated testing revealed evidence of a rota-tor cuff tear.

She continues to experience pain in her shoulder, arm, and lower back. Invoking 42 U.S.C. § 1983, the plaintiff sued Webster in Maine’s federal district court. She claimed that Webster’s actions constituted an unreasonable seizure of her person because they amounted to a detention without adequate justification, see, e.g., United States v. Romain, 393 F.3d 63, 70-71 (1st Cir.2004), and because he effected that detention through the use of excessive force, see, e.g., Calvi v. Knox County, 470 F.3d 422, 428 (1st Cir.2006).

The plaintiffs complaint also included supplemental claims under Maine law, which mirrored her federal claims. Upon the completion of discovery, Webster moved for

summary judgment, arguing that the undisputed facts justified the detention and made pellucid that the force he had applied was not excessive. In the alternative, he argued that he was entitled to qualified immunity. The district court granted this motion, holding that the undisputed facts established that Webster had adequate justification to seize the plaintiff regardless of whether that seizure amounted to an investigatory stop or a de facto arrest. Moreili 554 F.Supp.2d at 53.

In explaining this holding, the court employed the pooled knowledge doctrine to credit Webster with knowledge possessed by his compatriots and determined that the seizure amounted to a temporary detention, prompted by reasonable suspicion. Id. at 52-53. The court added that even if the seizure amounted to a de facto arrest, it was justified by probable cause. See id. (citing Me.Rev.Stat. Ann. tit. 17-A § 353(1)(A)).

In all events, the court posited that Webster would be entitled to qualified immunity because “a reasonable official would have believed that criminal activity, prostitution, was afoot, that Maine law permitted him to arrest Plaintiff and that he had probable cause to believe that Plaintiff had committed or was in the process of committing a crime.” Id. at 53 n. 9. With respect to the excessive force claim, the lower court relied entirely on qualified immunity.

It concluded that a reasonable officer in Webster’s situation could well have believed that the plaintiff had committed theft, was connected to prostitution, posed a threat to those around her, and was actively evading detention. Id. at 56-57. Thus, the force used by Webster was of a degree that a reasonable officer could have believed was appropriate under the circumstances. Id. Along the way, the district court noted that the resolution of the federal claims would dictate the outcome of the mirror-image supplemental claims. Id. at 51. The latter claims had been asserted under Me.Rev.Stat.

Ann. tit. 5 § 4682, a statute patterned on 42 U.S.C. § 1983. See Jenness v. Nickerson, 637 A.2d 1152, 1158 (Me.1994). Thus, there was no need for a separate analysis. With this reasoning in mind, the district court entered an adverse judgment on all of the plaintiffs claims. This timely appeal ensued. II. THE LEGAL LANDSCAPE Before addressing the assignments of error, we pause to discuss three sets of legal principles, each of which to some extent informs our analysis. A. Pooled Knowledge.

The pooled knowledge doctrine, sometimes known as the collective knowledge doctrine, is a mechanism that in some circumstances allows a court to “impute” facts known by one police officer to another police officer engaged in a joint mission. See, e.g., United States v. Meade, 110 F.3d 190, 193 (1st Cir.1997). The doctrine derives from a felt sense that officers acting in concert actually do, and are entitled to, assume that fellow officers are acting in a manner consistent with their legal responsibilities. See Whiteley v. Warden, Wyo.

State Pen., 401 U.S. 560, 568, 91 S.Ct. 1031, 28 L.Ed.2d 306 (1971); see generally Wayne R. LaFave, Search and Seizure § 3.5 (2004). Thus, for example, when an officer who has probable cause “directs an officer who lacks that knowledge to make the arrest, we ‘impute’ to the arresting officer the directing officer’s knowledge.” Meade, 110 F.3d at 193; see United States v. Cook, 277 F.3d 82, 86 (1st Cir.2002); Burns v. Loranger, 907 F.2d 233, 236 n. 7 (1st Cir.1990). In the case at hand, the district court invoked the pooled knowledge doctrine on two different points.

First, the court used it to credit Webster with knowledge possessed by a fellow officer, Farris, to the effect that the plaintiff was a known prostitute. Moreili 554 F.Supp.2d at 48 n. 2. Second, the court used it to credit Webster with knowledge possessed by McVane to the effect that the plaintiff had taken a $20 bill from the counter in room 203. Id. at 52-53. The plaintiff argues with considerable force that these are improper applications of the pooled knowledge doctrine. For the reasons explained below, we need not enter this thicket.

The plaintiffs version of events places Webster in the observation room when she arrived in room 203. Under that version, Webster must have heard Farris announce that he believed the plaintiff to be a known prostitute. Thus, the pooled knowledge doctrine added nothing to Webster’s store of reputational information. As to the theft, the plaintiff herself admits that she took the $20 bill. And on her account, Webster watched her actions from the observation room and did not enter the hallway until after she left room 203.

It follows that Webster must have seen her take the money and, accordingly, the pooled knowledge doctrine is of no consequence on this point. B. Standard of Review. We afford de novo review to a district court’s grant of

summary judgment. See Dávila v. Corporación de P.R. Para La Diofusión Pública, 498 F.3d 9,12 (1st Cir.2007).

Summary judgment is appropriate only when the record reflects no genuine issue as to any material fact and indicates that the moving party is entitled to judgment as a matter of law. See id.; see also Fed.R.Civ.P. 56(c). In a case in which the parties offer diametrically opposite versions of the facts, each founded on first-hand knowledge, we must ask whether the account propounded by the nonmov-ant suffices to thwart the swing of the

summary judgment ax. See, e.g., Scott, 127 S.Ct. at 1775. C. Qualified Immunity. This case differs from a garden-variety

summary judgment case because it involves the doctrine of qualified immunity. Qualified immunity is a judicial gloss designed to allow public officials to perform discretionary tasks without the constant threat of legal liability. See Pagán v. Calderón, 448 F.3d 16, 31 (1st Cir.2006). As the Supreme Court has explained, the doctrine is intended to protect “all but the plainly incompetent [and] those who knowingly violate the law.” Malley v. Briggs, 475 U.S. 335, 341, 106 S.Ct. 1092, 89 L.Ed.2d 271 (1986). Determining the applicability of the qualified immunity doctrine in a given case requires a three-step inquiry. Operating within that framework, an inquiring court must consider (

i) whether the plaintiffs proffered version of the facts, if true, makes out a violation of a constitutionally protected right; (ii) if so, whether that right was clearly established at the time of the putative violation; and (iii) if the answers to the preceding two queries are affirmative, whether a reasonable public official, situated similarly to the defendant, should have understood the challenged act or omission to violate the discerned right. See Pagán, 448 F.3d at 31; Limone v. Condon, 372 F.3d 39, 44 (1st Cir.2004). The doctrinal intersection of qualified immunity principles and

summary judgment principles is not well mapped. Plotting that intersection can present thorny analytic problems — problems that are magnified because of the desire to resolve claims of qualified immunity at the earliest practicable stage of litigation. See Cox v. Hainey, 391 F.3d 25, 29 (1st Cir.2004). The difficulty arises because the

summary judgment standard requires absolute deference to the nonmovant’s factual assertions (as long as those assertions are put forward on personal knowledge or otherwise documented by materials of evi-dentiary quality, see, e.g., Greenburg v. P.R. Marit. Shipping Auth., 835 F.2d 932, 936 (1st Cir.1987)), whereas qualified immunity, when raised on

summary judgment, demands deference to the reasonable, if mistaken, actions of the movant, see, e.g., Cox, 391 F.3d at 31. In order to ease this inherent tension, we think it wise for courts to cabin these standards and keep them logically distinct, first identifying the version of events that best comports with the

summary judgment standard and then asking whether, given that set of facts, a reasonable officer should have known that his actions were unlawful. See Wilson v. City of Boston, 421 F.3d 45, 53 n. 10 (1st Cir.2005) (noting that genuine disputes anent material facts must be resolved at trial even though qualified immunity is a question of law for the judge); Kelley v. LaForce, 288 F.3d 1, 7 (1st Cir.2002) (explaining that disputes as to material facts sometimes will preclude

summary judgment based on qualified immunity). We proceed in that vein. III. UNLAWFUL DETENTION We begin our journey through the plaintiffs asseverational array with her claim that Webster’s actions constituted an unlawful detention. The background principles are familiar. A detention at the hands of a police officer constitutes a seizure of the detainee’s person and, thus, must be adequately justified under the Fourth Amendment. Romain, 393 F.3d at 70-71.

The case law recognizes two classes of seizures falling along this continuum: arrests (whether actual or de facto) and temporary detentions (such as investigatory stops). The justification needed for these two types of seizures is qualitatively different: an arrest must be grounded on a showing of probable cause, see, e.g., Carroll v. United States, 267 U.S. 132, 156, 45 S.Ct. 280, 69 L.Ed. 543 (1925), whereas a temporary detention may be grounded on a lesser showing equivalent to reasonable suspicion, see, e.g., United States v. Zapata, 18 F.3d 971, 975 (1st Cir.1994).

In this case, the district court declined to locate Webster’s actions along this continuum. Rather, it declared those actions justified under either scenario. See Morelli, 554 F.Supp.2d at 53. We examine this conclusion. The line between temporary detentions and de facto arrests is often blurred. To complicate matters, that line can shift in the course of a single encounter so that what starts out as an investigatory stop may morph into a de facto arrest. See, e.g., United States v. Lee, 317 F.3d 26, 31 (1st Cir.2003).

It follows that an inquiring court must determine whether a police officer’s initial action was justified and, if so, whether subsequent (more coercive) actions undertaken by the officer were justified by developing circumstances. See United States v. Sowers, 136 F.3d 24, 27 (1st Cir.1998). In this instance, we can say with assurance that the facts known to Webster warranted an investigatory stop. Such a temporary detention, known colloquially as a Terry stop, see Terry v.

Ohio, 392 U.S. 1, 21, 88 S.Ct. 1868, 20 L.Ed.2d 889 (1968), requires only articulable facts giving rise to a reasonable suspicion that a suspect may be involved in criminal activity. United States v. Ruidiaz, 529 F.3d 25, 28 (1st Cir.2008). It is undisputed that the facts known to Webster at the inception of the encounter included, among other things, that the plaintiff had taken money from McVane before leaving room 203. That fact gave rise to a reasonable suspicion that the plaintiff had committed

an act of theft. See Me.Rev.Stat. Ann. tit. 17-A § 353(1)(A). In turn, that suspicion justified a temporary detention in order to investigate the possible commission of that crime. See Ruidiaz, 529 F.3d at 28; Zapata, 18 F.3d at 975. But giving credence to the plaintiffs version of events (as the

summary judgment standard requires), it is at least arguable that the character of the stop changed in mid-stream. Although there are no scientifically precise benchmarks for distinguishing between temporary detentions and de facto arrests, the standard mode of inquiry is to assess the totality of the circumstances and ask whether a reasonable person in the suspect’s shoes would understand herself to be subject to restraints comparable to those associated with an arrest. See Berkemer v. McCarty, 468 U.S. 420, 442, 104 S.Ct. 3138, 82 L.Ed.2d 317 (1984); Zapata, 18 F.3d at 975.

The “ultimate inquiry” is whether there was a “restraint on freedom of movement of the degree associated with a formal arrest.” United States v. Trueber, 238 F.3d 79, 93 (1st Cir.2001) (quoting Thompson v. Keohane, 516 U.S. 99, 112, 116 S.Ct. 457, 133 L.Ed.2d 383 (1995)). Hallmark factors such as physical restraint or a show of force may be suggestive, but in certain circumstances such factors may be perfectly consistent with an investigatory stop. See, e.g., United States v. Taylor, 162 F.3d 12, 21 (1st Cir.1998); United States v. Quinn, 815 F.2d 153, 156 (1st Cir.1987); see also United States v.

Kapperman, 764 F.2d 786, 790 n. 4 (11th Cir.1985) (stating that “neither handcuffing nor other restraints will automatically convert a Terry stop into a de facto arrest”). There is no per-se rule. For present purposes, we must take the plaintiffs account as true. On that version, Webster restrained her by the use of main force (forcibly halting her progress and pinning her against a wall for a period of three to four minutes). He told her that she was not going anywhere. He then escorted her from the public hallway into a hotel room where other officers were assembled. He proceeded to interrogate her.

None of these actions, in isolation, necessarily would have converted the initial Terry stop into a de facto arrest. See, e.g., United States v. Campa, 234 F.3d 733, 738-39 (1st Cir.2000) (directing suspects to move from hallway to kitchen did not convert a Terry stop into a de facto arrest); Zapata, 18 F.3d at 976-77 (holding that physical touching did not work such a conversion); United States v. Perdue, 8 F.3d 1455, 1463 (10th Cir.1993) (holding that immobilization does not automatically work such a conversion).

Moreover, there are other indicators that counsel against a finding that a de facto arrest occurred. For example, the plaintiff admits that, once they had repaired to room 203, Webster stated that he was not arresting her. Yet, at this stage of the litigation, we must take the evidence and all reasonable inferences therefrom in the light most favorable to the plaintiff. Cordi-Allen, 494 F.3d at 248.

From that coign of vantage, we think that a rational jury could conclude that the combination of events that transpired — the aggressive physical touching, the relatively lengthy immobilization, the removal of the plaintiff from a public hallway to a private room, the command not to leave, and the ensuing interrogation in a small room populated by multiple police officers — would have sufficed to lead a reasonable person in the plaintiffs place and stead to believe that Webster had imposed restraints on her freedom of movement tantamount to those implicated in a formal arrest.

See Zapata, 18 F.3d at 975; see also United States v. Acosta-Colon, 157 F.3d 9, 21 (1st Cir.1998) (stopping, handcuffing, and moving suspect to detention room constitute a de facto arrest). This brings us to the issue of justification. Because a reasonable jury could find a de facto arrest, Webster’s actions would be justified only if he acted on probable cause. Proof of probable cause is not to be confused with the more onerous standard of proof of guilt beyond a reasonable doubt. See United States v. Winchenbach, 197 F.3d 548, 555-56 (1st Cir.1999).

Rather, probable cause exists when an officer, acting upon apparently trustworthy information, reasonably can conclude that a crime has been or is about to be committed and that the suspect is implicated in its commission. See United States v. Brown, 500 F.3d 48, 56 (1st Cir.2007); United States v. Figueroa, 818 F.2d 1020, 1023 (1st Cir.1987). The district court discerned probable cause based on the plaintiffs “connections to prostitution, theft of the police department’s money, and [disobedience of] a police command to stop in the public corridor of a hotel.” Morelli, 554 F.Supp.2d at 53.

In this regard, the court seems to have been impressed that the plaintiff “admitted to making physical contact” with Webster as she tried to bypass him in the hallway. Id. We are hesitant to endorse the district court’s assessment as a whole. For one thing, while a belief that the plaintiff was connected to prostitution might have added marginally to the justification for a Terry stop, that belief had no bearing on any crime that conceivably could have been committed on the day in question. It cannot, therefore, undergird a probable cause determination. See Brown, 500 F.3d at 56.

By the same token, the plaintiffs attempt to walk past Webster might have given Webster the right to use a modicum of force to effectuate a stop. See, e.g., Isom v. Town of Warren, 360 F.3d 7, 10-11 (1st Cir.2004). But the plaintiff consistently has described the initial contact between her and Webster as amounting to “brushing against him” as they passed in the hallway. As described, that conduct was neither a crime nor evidence of a crime. See, e.g., State v. Worrey, 322 A.2d 73, 80 (Me.1974) (defining criminal assault). We nonetheless agree with the district court’s bottom-line conclusion.

On the undisputed facts, Webster did have knowledge of a likely crime — the apparent theft of $20 in police money. Webster had been in the observation room during the plaintiffs interaction with McVane and knew that she had taken a $20 bill from the pile of police money. As the district court correctly explained, theft is a crime under Maine law, for which an observing officer is empowered to make a warrant-less arrest. See Morelli, 554 F.Supp.2d at 53 (citing Me.Rev.Stat. Ann. tit. 17-A §§ 15(2), 353(1)(A)). The plaintiff strives to blunt the force of this conclusion in two ways.

We examine each of these initiatives. First, the plaintiff posits that Webster knew she had returned the money before he detained her. Building on this foundation, she contends that, by returning the money, she never possessed unauthorized control of it and, in the bargain, manifested that she lacked the requisite intent for the crime of theft. This gambit is unavailing. Even if the plaintiff returned the funds, she did not wipe the slate clean. Theft under Maine law comprises two elements: (

i) obtaining unauthorized control over another’s property (ii) with intent to deprive that person of it. See Me.Rev.Stat. Ann, tit. 17-A § 353(1)(A). Here, the putative crime was completed (or so the officer reasonably could have thought) as soon as the plaintiff took the money. It could not be erased by its perpetrator’s subsequent return of the funds. It is, after all, hornbook law that unauthorized control can be evinced by moving property even a slight distance in the presence of the owner. See Wayne R. LaFave, Substantive Criminal Law § 19.3(b) (2003).

By like token, an intent to deprive only need exist at the moment of taking. See id. § 19.5(f). Whatever a jury ultimately might find under a reasonable doubt instruction, Webster had probable cause to think that the requisite intent existed. See supra note 3. The plaintiffs second argument is no more robust. She maintains that after returning her to room 203 Webster stated: “I’m not arresting you, I can’t arrest you.” Even were we to read this comment favorably to the plaintiff, as an admission by Webster that he felt that he lacked probable cause, our inquiry would not end.

The test for the existence vel non of probable cause “is objective in nature.” Cox, 391 F.3d at 31; see Whren v. United States, 517 U.S. 806, 812, 116 S.Ct. 1769, 135 L.Ed.2d 89 (1996). Webster’s subjective belief is, therefore, not determinative of the probable cause issue. That ends this aspect of the matter. For the reasons discussed, we conclude that the district court did not err in granting

summary judgment on the unlawful detention claim (and, thus, on the mirror-image supplemental claim as well). IY. EXCESSIVE FORCE We turn next to the excessive force claim. The district court analyzed this claim under the rubric of qualified immunity. It concluded that the plaintiff had made out a colorable claim of a violation of a clearly defined constitutional right, but that a reasonable public official in Webster’s position reasonably could have believed that the force used to detain the plaintiff was appropriate. Morelli, 554 F.Supp.2d at 56-57. We disagree with the last step in this progression. We begin with bedrock. As we already have said, see supra

Part 11(C), the qualified immunity inquiry is threefold. An inquiring court must determine (

i) whether the plaintiff has asserted a cognizable violation of a constitutional right; (ii) whether that right was clearly established at the relevant time; and (iii) whether a reasonable public official in the defendant’s position should have understood that his actions infringed that right. See Pagán, 448 F.3d at 31; Limone, 372 F.3d at 44. We apply this template to the facts at hand and, because the case was terminated at the

summary judgment stage, we take as true the plaintiffs account of the relevant events. The first branch of the qualified immunity test is satisfied here. To establish a Fourth Amendment excessive force claim, a plaintiff must show that the defendant employed force that was unreasonable under all the circumstances. See Graham v. Connor, 490 U.S. 386, 396, 109 S.Ct. 1865, 104 L.Ed.2d 443 (1989); Asociación de Periodistas de P.R. v. Mueller, 529 F.3d 52, 59 (1st Cir.2008). In addressing this question, we do not write on a pristine page.

The Supreme Court has furnished a non-exclusive list of criteria for determining the objective reasonableness of a police officer’s use of force. These criteria include “the severity of the crime at issue,” the extent (if any) to which “the suspect poses an immediate threat to the safety of the officers or others;” and whether the suspect “is actively resisting arrest or attempting to evade arrest by flight.” Graham, 490 U.S. at 396, 109 S.Ct. 1865. When the facts in this case are marshaled in the light most hospitable to the plaintiff, these criteria cut sharply in her direction.

If a crime was committed at all, it was a Class E crime (the lowest level of criminality recognized under Maine law). See Me.Rev.Stat. Ann. tit. 17-A § 353(1)(A); see also id. §§ 4, 4-A. Despite the district court’s contrary assertion, Morelli, 554 F.Supp.2d at 56-57, there is no evidence cognizable under the

summary judgment standard indicating that the plaintiff posed a threat to the safety of the officers or others. And, finally, on the plaintiffs account, there is no evidence of any meaningful degree of resistance. A jury might also choose to infer that the defendant, frustrated at the looming failure of the sting operation and the prospect of the plaintiff avoiding arrest, lost his temper and intentionally used more force than the situation warranted. To say more about this element of the test would be to paint the lily.

We conclude, without serious question, that a rational jury could find that the force used by Webster to detain an unresisting woman who, at worst, was suspected of being a petty thief, was so disproportionate as to offend the Fourth Amendment. See Alexis v. McDonald’s Rests, of Mass., Inc., 67 F.3d 341, 353 (1st Cir.1995) (collecting cases in which force used to arrest was unreasonable in light of minor nature of crime). The second branch of the test is also satisfied.

A clearly established right is one sufficiently defined at a level of specificity that would put a state actor (such as a police officer) on fair notice that his specific actions offended the Constitution. See Limone, 372 F.3d at 46. Our case law supplies a crystal clear articulation of the right, grounded in the Fourth Amendment, to be free from the use of excessive force by an arresting officer. See, e.g., Alexis, 67 F.3d at 353-54 (concluding excessive force claim triable when officer seized and dragged plaintiff to effectuate arrest for crime of trespassing in a public restaurant).

Given this well-settled jurisprudence, there is no legitimate doubt that the right asserted here was clearly established. Thus, Webster was on notice that a police officer’s use of excessive force would be offensive to the Constitution. The question, then, reduces to whether Webster’s use of excessive force constituted the type and kind of erroneous judgment that a reasonable police officer under the same or similar circumstances might have made. See Camilo-Robles v. Hoyos, 151 F.3d 1, 14-15 (1st Cir.1998). We think not. This inquiry is a complicated one. By definition, excessive force is unreasonable force.

See Graham, 490 U.S. at 394, 109 S.Ct. 1865. But reasonable people sometimes make mistaken judgments, and a reasonable officer sometimes may use unreasonable force. In that event, qualified immunity gives an officer the benefit of a margin of error. See Saucier v. Katz, 533 U.S. 194, 205-06, 121 S.Ct. 2151, 150 L.Ed.2d 272 (2001) (explaining that qualified immunity operates in excessive force cases to “protect officers from the sometimes hazy border between excessive and acceptable force” (internal quotation marks omitted)); Jennings v.

Jones, 499 F.3d 2,18 (1st Cir.2007) (observing that, in effect, “officers receive protection if they acted reasonably in exercising unreasonable force.”). Thus, defeating a qualified immunity defense requires a showing of an incremental degree of error — an incommensurate use of force beyond that needed to establish a garden-variety excessive force claim and, further, beyond the “hazy border” noted by the Saucier Court, 533 U.S. at 206, 121 S.Ct. 2151.

Looked at another way, qualified immunity is appropriate in an excessive force case when an officer “correctly perceive^] all of the relevant facts but [has] a mistaken understanding” as to the legality of his chosen level of force. Id. at 205,121 S.Ct. 2151. Conversely, qualified immunity protection would not be available when the level of force chosen by the officer cannot in any way, shape, or form be justified under those facts. These principles necessitate a case-specific approach. Many cases will defy easy classification.

This case, however, does not present a borderline situation; and we think that the plaintiff has made a sufficient showing to avert the entry of

summary judgment predicated on qualified immunity. As will almost always be true, this determination is based on a matter of degree. Here, the facts, seen through the prism of the plaintiffs account, simply do not justify yanking the arm of an unarmed and non-violent person, suspected only of the theft of $20, and pinning her against a wall for three to four minutes with sufficient force to tear her rotator cuff.

That is particularly so in view of the marked disparity in height and weight between the officer and the suspect, the absence of any evidence of either dangerousness or attempted flight, and the presence of a cadre of other officers at the scene. In short, the plaintiffs version of the relevant facts places Webster’s actions outside the universe of protected mistakes. See, e.g., Vondrak v. City of Las Cruces, 535 F.3d 1198, 1209-10 (10th Cir.2008); Jones v. Parmley, 465 F.3d 46, 63 (2d Cir.2006); Smoak v. Hall, 460 F.3d 768, 784 (6th Cir.2006); Kukla v. Hulm, 310 F.3d 1046, 1050 (8th Cir.2002); DeGraff v.

Dist. of Columbia, 120 F.3d 298, 302 (D.C.Cir.1997). The district court stressed that a reasonable officer, believing that the plaintiff committed a theft, would have thought it lawful to use some force to detain her. Morelli, 554 F.Supp.2d at 57. That is true as far as it goes. Cf. Graham, 490 U.S. at 396, 109 S.Ct. 1865 (noting that the right to make an arrest carries with it the right to use some force).

But the district court’s further conclusion that the plaintiffs actions justified Webster in believing that a significant degree of force could be used, see Morelli, 554 F.Supp.2d at 56-57, cannot be squared with the

summary judgment standard. While the officers’ accounts of the facts may justify the district court’s appraisal, the court was bound to ask not whether those accounts were plausible but, rather, whether under the plaintiffs version of the facts a reasonable officer should have known that the degree of force used was plainly excessive. See Griffith v. Cobum, 473 F.3d 650, 656-57 (6th Cir.2007) (“Because determining reasonableness in [the excessive force] context is such a fact-intensive endeavor

summary judgment is improper if the legal question of immunity turns on which version of the facts is accepted.”); see also Davila, 498 F.3d at 12. To be sure, incidental contact — say, a “gratuitously violent shove” — may be within the realm of conduct protected by qualified immunity. Saucier, 533 U.S. at 208, 121 S.Ct. 2151. Here, however, Webster’s conduct, as described by the plaintiff, eclipsed the bounds of reasonableness.

Given the importance of reasonableness to the qualified immunity calculus in excessive force cases, the existence of such immunity frequently will hinge on the resolution of disputed facts. See, e.g., Jennings, 499 F.3d at 18-20. So it is here. In sum, we conclude that the plaintiff not only has made out a trialworthy issue as to whether Webster’s use of a significant degree of force transgressed her Fourth Amendment right to be free from excessive force but also that she has made a showing adequate to thwart a qualified immunity defense. See Kelley, 288 F.3d at 7.

Because a rational jury could find, on this scumbled record, facts establishing that Webster’s use of force was so objectively unreasonable and so plainly misguided that he should not be protected by the shield of qualified immunity, the district court erred in resolving this claim in advance of trial. See id.; Alexis, 67 F.3d at 352-53. Let us be perfectly clear. We understand that the plaintiffs version of the relevant events is hotly contested, and we make no judgment as to where the truth lies. See, e.g., Greenburg, 835 F.2d at 936 (discussing operation of

summary judgment standard). Y. CONCLUSION We need go no further. For the reasons elucidated above, we affirm the grant of

summary judgment on the unlawful detention claim and the mirror-image supplemental claim, reverse the grant of

summary judgment on the excessive force claim and the mirror-image supplemental claim, and remand to the district court for further proceedings consistent with this opinion. Affirmed in part, reversed part, and remanded. No costs. . We adhere to this approach. Although we analyze the issues in terms of the federal claims, our reasoning applies with undiminished force to the supplemental claims. .

We previously noted that the Supreme Court has never clearly indicated whether the judge or the jury is the proper factfinder when a case is actually tried and a factual dispute underlies a proffered qualified immunity defense. Kelley, 288 F.3d at 7 n. 2. We have, however, expressed some doubt that "the Supreme Court intended this [factual] dispute to be resolved from the bench by fiat.” Id.; see Jennings v. Jones, 499 F.3d 2, 10 (1st Cir.2007) (requiring, on particular facts, that a judge’s post-verdict qualified immunity ruling be consistent with the jury verdict). .

One might argue that Webster should have understood that the plaintiffs statement that she was entitled to the money, coupled with the absence of an immediate protest by McVane, indicated acquiescence. But apart from the fact that the plaintiff has not pressed this point, the circumstances simply are not clear-cut enough either to compel a conclusion that silence equalled acquiescence or to preclude a reasonable belief that a crime was being committed. See generally New Jersey v.

T.L.O., 469 U.S. 325, 346, 105 S.Ct. 733, 83 L.Ed.2d 720 (1985) (explaining that "the requirement of reasonable suspicion is not a requirement of absolute certainty: sufficient probability, not certainty, is the touchstone of reasonableness under the Fourth Amendment”) (internal quotation marks omitted). . Of course, there was evidence that the plaintiff was holding her right arm stiffly, but she plausibly attributed this gesture to the need to provide support for the heavy coat that was draped over it. We do not think that this is sufficient, standing alone, to justify the use of a significant level of force.

Cf. Parker v. Gerrish, 547 F.3d 1, 9 (1st Cir.2008) (finding de minimis resistance to arrest insufficient to justify force used).