John McCottrell v. Marcus White

In the United States Court of Appeals For the Seventh Circuit No. 17-2295 JOHN MCCOTTRELL and DUSTIN S. CLAY, Plaintiffs-Appellants, v. MARCUS WHITE and LABARIN WILLIAMS, Defendants-Appellees. Appeal from the United States District Court for the Northern District of Illinois, Eastern Division. No. 1:15-cv-03208 — Amy J. St. Eve, Judge. ARGUED OCTOBER 25, 2018 — DECIDED JULY 29, 2019 Before ROVNER, HAMILTON, and BARRETT, Circuit Judges. ROVNER, Circuit Judge. The plaintiffs were inmates at Stateville Correctional Center when they were struck by buckshot fired by the defendant prison guards. The plaintiffs sued under 42 U.S.C. § 1983, asserting that the guards violated 2 No. 17-2295 their rights under the Eighth Amendment when they discharged their shotguns over a crowded prison dining hall. The guards countered that they fired the shots as a necessary warning to two other inmates who were fighting with each other and resisting the efforts of other guards who were trying to break up the conflict. The district court granted summary judgment in favor of the defendants. We vacate and remand. I. On summary judgment, we must construe the facts in favor of the nonmovant, and may not make credibility determinations or weigh the evidence. Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 255 (1986); Payne v. Pauley, 337 F.3d 767, 770 (7th Cir. 2003). At the time of the November 6, 2013 incident, inmates John McCottrell and Dustin Clay were eating lunch in the Stateville Correctional Center dining hall. Guards Marcus White and Labarin Williams were stationed in a tower some fifteen feet above the dining hall and overlooking the area where inmates wait in line to enter the hall (a fenced space which the parties call “the chute”) and the seating area. White and Williams were armed with loaded shotguns. In the dining hall, security staff are vastly outnumbered by inmates, who enter the hall without hand or foot restraints. Guards on the floor of the dining hall are armed only with pepper spray, and as many as four hundred inmates may be in the dining hall at a meal. Fights occur more frequently in the dining hall than in any other part of the facility because mealtimes are one of the few occasions where large groups of unrestrained inmates interact. Fights may escalate quickly, and inmates are sometimes armed with improvised weapons. Fights therefore pose a serious security threat to staff and other inmates. No. 17-2295 3 On this occasion, a scuffle broke out between two inmates (for security reasons, we will not name them) who were entering the hall via the chute, approximately forty to fifty feet from where the plaintiffs were seated. The brawling inmates were not armed and were wrestling or tussling with one another. Several correctional officers quickly intervened in the confrontation. Both inmates initially resisted the staff but short blasts of pepper spray from the officers on the floor brought compliance from each. According to the plaintiffs, after the inmates were separated and subdued, they were in the process of being handcuffed when White and Williams simultaneously discharged their shotguns over the dining hall.1 The ceiling of the dining hall is equipped with a “shot box,” a device intended to reduce the ricochet from warning shots. Neither White nor Williams hit the shot box, instead discharging their weapons either in the direction of the plaintiffs or into the ceiling, as we will clarify in a moment. Along with two other inmate bystanders, McCottrell and Clay were struck by buckshot from the blasts. The entire incident lasted less than a minute. Clay was wounded in his right arm above the elbow 1 At his deposition, Clay testified that the guards “separated the individu- als that was tussling, put cuffs on them, and then a shot rang out.” R. 75-3, Tr. at 11. He also testified that, “Both inmates were separated. Both inmates was in handcuffs, so there was no need to shoot.” R. 75-3, Tr. at 65. But he also testified that the inmates were “being cuffed” at the moment the shot rang out. R. 75-3, Tr. at 25–26. In response to the clarifying question, “They were both handcuffed?” Clay responded, “They was being cuffed at that time.” R. 75-3, Tr. at 25. That specific clarification governs our understand- ing of the timing, although it is not necessary to the outcome to determine whether the cuffs were in place or in the process of being applied when the shots were fired. 4 No. 17-2295 and the injury was serious enough to require stitches. McCottrell was struck in the neck and the leg, and was given bandages for his wounds. In addition to their physical injuries, both men suffered mental health issues arising from the incident. Before proceeding, we must clarify the summary judgment record regarding the direction in which the shots were fired. In the district court, the plaintiffs argued that the guards fired toward the inmates in the dining hall, rather than into the ceiling. R. 82, at 1–2. Both plaintiffs testified in their depositions that, although they did not see where the guards were aiming their guns, they both assumed that the guns must have been pointed toward the inmates because of the number of inmates who were hit by buckshot and because of their own wounds. R. 75-3, Tr. at 34; R. 75-2, Tr. at 21. The guards averred that they fired into the ceiling. In response to the defendants’ statement of uncontested facts, the plaintiffs’ lawyer contended that it was unlikely that both guards had fired into the ceiling because it was composed of acoustic tile and yet four inmates were struck by buckshot.2 R. 84, at 10. But there is no evidence in the record regarding the composition of the ceiling. Oral Argument at 14:51–15:15 (defendants’ counsel confirming that the record does not contain evidence regarding composition of ceiling). The district court appropriately rejected as hearsay the 2 The plaintiffs’ counsel argued that ricochet was unlikely in response to the claim by the guards that they fired at the ceiling rather than directly at the crowd. Having not seen the direction of fire, the plaintiffs argued in the alternative that the guns were either pointed directly at the crowd or at the ceiling and away from the shot boxes. We address the plaintiffs’ preserva- tion of this claim infra at note 5 and accompanying text. No. 17-2295 5 plaintiffs’ additional assertions that other inmates told them that the guns were pointed toward the crowd.3 The court then accepted as undisputed fact the defendants’ claims that they aimed at the ceiling. But the direction of fire cannot be conclusively resolved on this thin and disputed record. Circumstantial evidence supports two possible paths for the buckshot to travel to the plaintiffs. The plaintiffs’ assertion that the guns were pointed toward the inmates is a reasonable inference drawn from circumstantial evidence given that the buckshot penetrated the clothing and the skin of multiple bystander inmates (including the plaintiffs). One could argue that when A fires a gun and the bullet ends up in B, the most natural inference is that the gun was pointed at B. That inference in this case is supported by the force with which the buckshot arrived (again, the shot penetrated the clothing and skin of the plaintiffs and buckshot remains embedded in Clay’s arm) and the number of persons injured. The defendants’ alternate assertion that the buckshot arrived indirectly, by ricochet, is also a reasonable inference, given that the shotguns (devices designed to scatter the pellets 3 We agree with the dissent that the district court correctly resolved the hearsay issue but as we explain, hearsay was not the only evidence supporting an inference that the guards fired in the direction of the inmates. The district court did not address the plaintiffs’ circumstantial evidence on this point. On summary judgment, the scope of our review is plenary. Although we review the court’s ruling on the hearsay issue for abuse of discretion, we review the grant of summary judgment de novo, granting no deference to the district court. Haynes v. Indiana Univ., 902 F.3d 724, 730 (7th Cir. 2018) (“We review a summary judgment de novo, … and we set aside the district court’s evidentiary rulings only for an abuse of discretion[.]”). 6 No. 17-2295 loaded in the shell) were fired indoors, in a crowded room, and away from the device installed to reduce ricochet. That inference is supported by the defendants’ affidavits, but those affidavits do not conclusively resolve which competing inference is correct.4 Circumstantial evidence “is proof of a fact, or a series of facts, that tends to show that some other fact is true.”Seventh Circuit Pattern Jury Instructions - Civil § 1.12. For example, the observation of someone entering a room carrying a wet umbrella is circumstantial evidence that it is raining. Id. Similarly, the presence of buckshot in a person’s body gives rise to competing inferences that the shotgun was either pointed in that person’s direction or pointed in a manner that allowed the shot to arrive indirectly, by ricochet. Courts routinely direct juries that the “law makes no distinction between the weight to be given to either direct or circumstantial evidence.” Id. That is so because “[c]ircumstantial evidence is not only sufficient, but may also be more certain, satisfying and persuasive than direct evidence.” Desert Palace, Inc. v. Costa, 539 U.S. 90, 100 (2003) 4 As the dissent notes, an internal investigation into the incident also characterizes the injuries as occurring from the ricochet of warning shots. But the report does not expressly analyze the issue of the direction of fire and reaches no official conclusion on that issue. The part of the report quoted by the dissent is based on “information” that initiated the investiga- tion. The report then addresses various versions of the incident from staff and inmate witnesses. The report assumes that the injuries occurred by ricochet. In any case, the report presents nothing more than a competing view of a contested fact, and it is not conclusive for summary judgment purposes. No. 17-2295 7 (quoting Rogers v. Missouri Pacific R. Co., 352 U.S. 500, 508 n.1 (1957)). Indeed, circumstantial evidence is sufficient to establish guilt beyond a reasonable doubt in criminal cases, and deserves no less respect in civil matters. See Holland v. United States, 348 U.S. 121, 139–40 (1954). The dissent’s assertion that the plaintiffs have “no evidence that the officers shot into the crowd” is incorrect. The plaintiffs possess circumstantial evidence that the gun was pointed in their direction. In the dissent’s view, a court would be required to conclusively credit a witness’s sworn statement that it was not raining even in the face of the proverbial wet umbrella. But in the face of circumstantial evidence, a jury would not be required to credit the defendants’ statements here, and that would leave the two competing inferences that the shot arrived either directly or indirectly. On summary judgment, a “‘court may not assess the credibility of witnesses, choose between competing inferences or balance the relative weight of conflicting evidence; it must view all the evidence in the record in the light most favorable to the non-moving party and resolve all factual disputes in favor of the non-moving party.’” Orton-Bell v. Indiana, 759 F.3d 768, 772–73 (7th Cir. 2014) (quoting Abdullahi v. City of Madison, 423 F.3d 763, 769 (7th Cir. 2005)). And if a jury not only rejected the defendants’ assertion but also concluded that the defendants affirmatively lied about the direction of fire, that lie would be additional circumstantial evidence supporting the inference that the shot was aimed toward the plaintiffs. See Reeves v. Sanderson Plumbing Products, Inc., 530 U.S. 133, 147 (2000) (in the employment discrimination setting, proof that the defendant’s purportedly nondiscriminatory explanation for its actions is “unworthy of 8 No. 17-2295 credence is simply one form of circumstantial evidence that is probative of intentional discrimination, and it may be quite persuasive.”). Contrary to the dissent’s charge that we have given the plaintiffs “an assist,” the plaintiffs’ arguments below and on appeal encompass both the possibility that the shots were fired into the ceiling, as the defendants claim, or toward the inmates, as may be inferred from the injuries sustained. See R. 82, at 1–2 (characterizing the defendants’ actions as “[f]iring shotguns into a crowd of innocent bystanders …” and asserting that defendants “were aiming at them when they fired”).5 Having preserved the issue in the district court, the plaintiffs do not expressly concede on appeal that the shots were directed at the ceiling (as the defendants claim), instead arguing that the 5 Indeed, even the defendants extensively referenced the plaintiffs’ theory that the shots were fired in their direction in their Statement of Uncontested Facts. R. 75, at 5, ¶¶ 24–29. Paragraph 27 specifies, “Plaintiff McCottrell admits that the only reason he believes that the gun was aimed at him when he was shot is the fact that he was hit.” The plaintiffs, in turn, admitted that assertion in their Response. R. 84, at 6. The defendants argued in their Memorandum of Law in Support of Their Motion for Summary Judgment that the plaintiffs’ injuries were not sufficient evidence of the defendants’ intent, and that there was no evidence supporting a claim that they fired directly into the dining hall. R. 76, at 12. The plaintiffs then countered in their response brief below that they both credibly believed that the defendants were aiming at them when the shotguns were discharged. R. 82, at 2. In their reply, the defendants argued that the plaintiffs’ injuries were not evidence of the direction of fire, characterizing this circumstantial evidence as an assertion of the plaintiffs’ “unsupported beliefs.” R. 90, at 3–4. But as we explain, it is reasonable to infer from the presence of buckshot in a person’s body that the shooter was pointing the shotgun in that person’s direction. No. 17-2295 9 district court “improperly credited Defendants’ version of the facts.” Opening Brief at 3. On appeal, the plaintiffs focus primarily on the guards’ avoidance of the shot box, but contrary to the dissent’s assertion that the direction of the shots is undisputed, there is a preserved fact question regarding the direction of fire. See Opening Brief at 2 (“fired gunshots that hit at least four bystander inmates”); at 4–5 (“fired buckshots that hit four inmates”); at 6 (“failed to shoot the ‘shot box’”); at 11 (“neither Defendant shot the shotboxes”); at 20 (“extreme response of firing gunshots into the ceiling of a crowded area”); at 25 (“district court gave Defendants credit for purportedly choosing to aim ‘at the ceiling and at the shot board on the ceiling’”); at 26 (“A reasonable jury could instead determine that Defendants’ decision to shoot in the first place and their failure to hit the shot box intended to catch buckshot was because they did not want the buckshot caught.”). In addition to these references in the Opening Brief, the plaintiffs expressly characterize the shots as being fired into a crowded area or near a crowd in their Reply Brief. Reply Brief at 11–15. As we explain above, that is a reasonable inference drawn from circumstantial evidence. That there is a competing inference and testimony to the contrary creates an issue for the jury. We will also focus primarily on the claim that the shots were aimed at the ceiling and away from the shot box, but the plaintiffs are free to dispute the direction of fire on remand, having never conceded the point. As we conclude below, summary judgment is not appropriate in either scenario. White and Williams filed written Incident Reports with prison authorities and submitted to interviews with an internal affairs investigator from the Illinois Department of Corrections 10 No. 17-2295 regarding the event. They also submitted affidavits with their motion for summary judgment. White and Williams wrote the Incident Reports within hours of the event, and the interviews took place in the early afternoon of the same day. In his brief Incident Report, Williams wrote that when staff rushed in to separate the inmates, the inmates began to struggle with the staff. He asserted that he believed the lives of the staff were in danger and so he “fired a warning round into the ceiling.” He wrote that staff members were then able to secure the inmates. Williams told the investigator that he saw the inmates begin to fight but did not see any weapons. The inmates were wrestling up against a wall when staff rushed in to separate them. Williams saw the staff separate the inmates but did not see whether any officers on the ground used pepper spray. After the inmates were pulled apart, he could see only one of the inmates and that man was resisting and non-compliant but did not strike any staff members. Williams claimed that he thought the lives of the guards were in danger and so he fired a warning shot into the ceiling. R. 86, at 52–53. In his Incident Report, White wrote that he saw the two inmates striking each other in the face and head with closed fists. He saw security staff respond to the scene and attempt to break up the fight but the men continued to be combative and would not stop fighting. At that point, White said, he believed that lives were in imminent danger and so he fired one warning shot into the ceiling. After the shot, the guards on the floor were able to separate and restrain the brawling inmates. White told the investigator that Williams directed his attention to the chute where two inmates were “punching each other in the face.” He saw “a bunch of staff” attempt to stop the fight No. 17-2295 11 but the inmates continued to scuffle. White reported that the inmates were “still going at it” when he fired a warning shot into the ceiling. He did not see a weapon and did not see the inmates strike any staff member but would not rule out the possibility that a staff member had been struck. White told the investigator that he fired the warning shot because he felt that the staff members were in imminent danger. Finally, White said that he saw the staff members use pepper spray after he fired the warning shot. After completing interviews and receiving reports from staff and inmate witnesses, the investigator issued a report concluding that both White and Williams “violated Department Rules regarding Use of Excessive Force[.]” R. 86, at 5. The investigator found that the shots were fired immediately after one of the guards administered pepper spray toward one of the brawling inmates. Both defendants failed to note in their Incident Reports that the fighting inmates had been separated before the warning shots were fired. Both White and Williams also failed to give accurate depictions of the incident in chronological order in their official reports. For example, White’s claim that the inmates were “still going at it” when he fired a warning shot conflicted with the video and other evidence.6 Moreover, no staff members were punched or struck during the altercation, and the video showed no signs that anyone within view sustained injuries that could be 6 The district court concluded that the factual findings from the internal affairs report were admissible under Federal Rule of Evidence 803(8), and the defendants have not challenged that conclusion. We discuss below the use of the report in summary judgment proceedings. 12 No. 17-2295 considered “great bodily harm.” R. 86, at 5. The report concluded that “White and Williams used an unreasonable amount of force (warning shot) outside the scope and not in accordance with Departmental Rule 501.30 Justifiable Use of Force.” R. 86, at 5.7 McCottrell and Clay sued Williams and White under 42 U.S.C. § 1983, alleging that the guards violated their rights under the Eighth Amendment by using excessive force when they fired shotguns into the crowded dining hall after the conflict was over.8 During the course of the litigation, some three years after the internal affairs report was issued, White and Williams signed affidavits describing the event somewhat differently than they had on the day of the shooting. White’s affidavit states, in relevant part: On November 6, 2013, I witnessed two offenders fighting within the dining hall. Correctional staff acted quickly to separate the offenders. However, one of the offenders continued to struggle with correctional staff and refused to comply. Correctional staff members were having difficulty restraining this offender. I discharged a warning shot into the ceiling. 7 There is no “Rule 501.30 Justifiable Use of Force,” and so the reference to Rule 501.30 may have been in error. Rule 501.30 covers “Resort to Force,” and Rule 501.40 is titled “Justifiable Use of Force.” Both rules would arguably apply to the incident at issue here. 8 The plaintiffs initially raised a number of additional claims against the guards and the prison’s health care providers but the only claim that is on appeal is the one against White and Williams for excessive force. No. 17-2295 13 R. 75-5, at 2. Williams’ affidavit repeats this language word for word, adding the phrase, “toward the nearest shot-box” to the end of the final sentence. R. 75-4, at 2. White added to his affidavit that he shot into the ceiling as opposed to the shot box because he had been advised that it would decrease the risk of ricochet and he wished to minimize that risk. Thus, contrary to their earlier reports, they acknowledged during the litigation that the inmates had been separated and that only a single inmate purportedly continued to struggle at the time the shots were fired. Williams had not previously reported that his firearm was directed toward the shot box, and White had not previously claimed that he believed the ceiling to present less of a risk of ricochet than the shot box. Both guards averred that they believed that warning shots were necessary to restore safety and order in the dining hall, that the struggling offender posed a serious threat to the correctional staff attempting to subdue him, that they did not know whether the offender possessed a weapon, and that they did not fire their weapons with the intent to injure anyone. The district court granted summary judgment in favor of the defendants after concluding that the plaintiffs lacked evidence that the officers’ use of force was wanton or unnecessary. Noting the statements of White and Williams that they fired the shots in order to restore order, the court found that the defendants’ belief that the shots were necessary was reasonable given that prison fights often escalate quickly and inmates sometimes use makeshift weapons. Crediting the defendants’ statements that the shots were fired toward the ceiling rather than in the direction of the plaintiffs, the district court found that the direction of the shots indicated an attempt 14 No. 17-2295 to temper the severity of the response. The court also noted that neither officer knew the plaintiffs or harbored any personal ill will toward them. After reviewing a blurry surveillance video of the dining hall, the court concluded that the shots were fired while one of the inmates continued to struggle with the officers, and the court found that it was therefore reasonable to believe that the warning shot was necessary to restore order.9 Because the plaintiffs lacked 9 Like the district court, we too have reviewed the video. The quality of the blurred black and white video is extremely poor and there is no audio track. It is difficult to distinguish one individual from another in the area where the fight took place, and it is only from the reactions of the seated bystand- ers that one could hazard a guess as to when the shots were fired. Although there is a counter/time index included with the video, the parties could not agree during Clay’s deposition whether the index represented seconds elapsed. R. 75-3, Tr. at 31. The jerky motion of the persons moving in the video suggests that more than one second has elapsed between frames or that perhaps only one image was taken each second; it is simply impossible to tell. It is also impossible to determine when the pepper spray was applied. We cannot discern from the video if the fighting inmates were continuing to struggle with officers at the time the shots were fired. But the internal affairs investigator who cited the video had the benefit of inter- viewing the persons depicted in it and appears to have drawn conclusions about the video based on those interviews as well as on the video itself. A person who participated in the event might be able to identify the blurry figures and the actions taken but we are not able to do so. The district court made no express ruling on the admissibility of the video. The defendants made no attempt in the record to lay a foundation for the video, and so we have no idea if the view presented in the video is the same view that the guards had of the situation. Nor do we know whether the video has been altered in any way. We will not draw independent factual conclusions from it. To the extent that the district court’s interpretation of the poor quality (continued...) No. 17-2295 15 evidence that the defendants acted “maliciously and sadistically,”10 the court granted judgment in favor of the defendants. McCottrell and Clay appeal. II. On appeal, McCottrell and Clay assert that there are genuine issues of material fact regarding the intent of the defendant officers when they fired the shots over the dining hall. They argue that the court should have applied the five- factor test set forth by the Supreme Court in Hudson v. McMillian, 503 U.S. 1 (1992), and Whitley v. Albers, 475 U.S. 312 (1986), in determining the officers’ intent, and that application of those factors would have made clear that the key question of intent was disputed. Instead, they maintain, the court inappropriately resolved disputed fact questions in favor of the defendants. We review the district court’s grant of summary judgment de novo, examining the record in the light most favorable to the plaintiffs and construing all reasonable 9 (...continued) video conflicts with the testimony of the plaintiffs or the conclusions of the investigator who interviewed the persons involved in the incident, we must credit the plaintiffs’ version of events on summary judgment. Cf Scott v. Harris, 550 U.S. 372, 380–81 (2007) (where a reliable videotape clearly captures an event in dispute and blatantly contradicts one party’s version of the event so that no reasonable jury could credit that party’s story, a court should not adopt that party’s version of the facts for the purpose of ruling on a motion for summary judgment). 10 As we will discuss below, in Whitley v. Albers, 475 U.S. 312, 320–21 (1986), the Supreme Court set this as the standard for liability in Eighth Amend- ment excessive force claims related to prison security measures. 16 No. 17-2295 inferences from the evidence in their favor. Anderson, 477 U.S. at 255; Lapre v. City of Chicago, 911 F.3d 424, 430 (7th Cir. 2018). Summary judgment is appropriate when there are no genuine disputes of material fact and the movant is entitled to judgment as a matter of law. Fed.R.Civ.P. 56(a); Anderson, 477 U.S. at 247–48; Lapre, 911 F.3d at 430. A. The Eighth Amendment forbids the imposition of “cruel and unusual punishments” against persons convicted of crimes. Whitley, 475 U.S. at 318. In Whitley, prison personnel were faced with a riot instigated by drunken inmates who took a guard hostage. One inmate was armed with a homemade knife and had threatened to kill the hostage if prison officials sent in an assault squad to end the conflict. After assessing the situation, officials decided to send in the armed assault team with directions to fire a warning shot and then aim low at any prisoners taking a staircase that led to the area where the hostage was being held. The assault squad was successful in retrieving the hostage and capturing the inmate who was armed with a knife. But in the course of the rescue, an inmate uninvolved in the fracas was shot when he attempted to take the staircase back to his cell. 475 U.S. at 314–16. The bystander inmate sustained severe injuries to his leg as well as mental and emotional distress. Like McCottrell and Clay, he brought suit under 42 U.S.C. § 1983, claiming violation of his rights under the Eighth Amendment. Although the district court granted summary judgment to the defendants, the court of appeals reversed. The court of appeals relied on evidence that the disturbance was subsiding at the time that No. 17-2295 17 prison officials sent in the assault squad. Moreover, the plaintiff had presented expert evidence concluding that the amount of force employed was excessive under the circumstances. But the Supreme Court determined that the court of appeals had “effectively collapsed the distinction between mere negligence and wanton conduct … implicit in the Eighth Amendment.” Whitley, 475 U.S. at 322. Although the general disturbance had died down, the situation remained dangerous and volatile; the safety of the guard being held hostage was still in question and an inmate was armed. As for the expert opinion: At most, this evidence, which was controverted by petitioners’ experts, establishes that prison officials arguably erred in judgment when they decided on a plan that employed potentially deadly force. It falls far short of a showing that there was no plausible basis for the officials’ belief that this degree of force was necessary. Indeed, any such conclusion would run counter to common sense, in light of the risks to the life of the hostage and the safety of inmates that demonstrably persisted notwithstanding repeated attempts to defuse the situation. An expert's after-the-fact opinion that danger was not “imminent” in no way establishes that there was no danger, or that a conclusion by the officers that it was imminent would have been wholly unreasonable. Whitley, 475 U.S. at 323. Ultimately the Court concluded that the prison officials did not violate the plaintiff’s Eighth 18 No. 17-2295 Amendment right to be free from cruel and unusual punishment. Whitley, 475 U.S. at 326. In reaching this conclusion, the Court reasoned that Eighth Amendment claims must be analyzed “with due regard for differences in the kind of conduct against which an Eighth Amendment objection is lodged.” Whitley, 475 U.S. at 320. For example, claims regarding deliberate indifference to a prisoner’s serious medical needs or harsh conditions of confinement do not present the same institutional concerns as those relating to the amount of force needed to maintain or restore order to the prison. Id. As the Court pointed out, the jailer’s “responsibility to attend to the medical needs of prisoners does not ordinarily clash with other equally important governmental responsibilities.” Id. But, in making and carrying out decisions involving the use of force to restore order in the face of a prison disturbance, prison officials undoubtedly must take into account the very real threats the unrest presents to inmates and prison officials alike, in addition to the possible harms to inmates against whom force might be used. Whitley, 475 U.S. at 320. For that reason: Where a prison security measure is undertaken to resolve a disturbance … that indisputably poses significant risks to the safety of inmates and prison staff, … the question whether the measure taken inflicted unnecessary and wanton pain and suffering ultimately turns on “whether force was applied in a good faith effort to maintain or restore discipline or No. 17-2295 19 maliciously and sadistically for the very purpose of causing harm.” Whitley, 475 U.S. at 320–21 (quoting Johnson v. Glick, 481 F.2d 1028, 1033 (2d Cir. 1973) (Friendly, J.)). The ultimate determination of the intent of the person applying the force in an excessive force claim involving prison security measures depends upon a number of factors, including: (1) the need for the application of force; (2) the relationship between the need and the amount of force that was used; (3) the extent of injury inflicted; (4) the extent of the threat to the safety of staff and inmates, as reasonably perceived by the responsible officials on the basis of the facts known to them; and (5) any efforts made to temper the severity of a forceful response. Whitley, 475 U.S. at 321. See also Hudson, 503 U.S. at 7 (listing the five factors); Lewis v. Downey, 581 F.3d 467, 477 (7th Cir. 2009) (same); Fillmore v. Page, 358 F.3d 496, 504 (7th Cir. 2004) (same). “From such considerations inferences may be drawn as to whether the use of force could plausibly have been thought necessary, or instead evinced such wantonness with respect to the unjustified infliction of harm as is tantamount to a knowing willingness that it occur.” Whitley, 475 U.S. at 321. Deference must be accorded to prison administrators in the adoption and execution of policies and practices that are needed to preserve order and ensure institutional security. Whitley, 475 U.S. at 321–22. The Court intended that this deference be extended to a particular security measure taken in response to an actual prison disturbance as well as to preventative measures: 20 No. 17-2295 [That deference] does not insulate from review actions taken in bad faith and for no legitimate purpose, but it requires that neither judge nor jury freely substitute their judgment for that of officials who have made a considered choice. Accordingly, in ruling on a motion for a directed verdict in a case such as this, courts must determine whether the evidence goes beyond a mere dispute over the reasonableness of a particular use of force or the existence of arguably superior alternatives. Unless it appears that the evidence, viewed in the light most favorable to the plaintiff, will support a reliable inference of wantonness in the infliction of pain under the standard we have described, the case should not go to the jury. Whitley, 475 U.S. at 322. Although Whitley involved a prison riot that included a guard being held hostage, the Court later remarked that the concerns underlying the holding of Whitley apply whenever guards use force to keep order: Whether the prison disturbance is a riot or a lesser disruption, corrections officers must balance the need “to maintain or restore discipline” through force against the risk of injury to inmates. Both situations may require prison officials to act quickly and decisively. Likewise, both implicate the principle that “‘[p]rison administrators ... should be accorded wide-ranging deference in the adoption and execution of policies and practices that in their No. 17-2295 21 judgment are needed to preserve internal order and discipline and to maintain institutional security.’” In recognition of these similarities, we hold that whenever prison officials stand accused of using excessive physical force in violation of the Cruel and Unusual Punishments Clause, the core judicial inquiry is that set out in Whitley: whether force was applied in a good-faith effort to maintain or restore discipline, or maliciously and sadistically to cause harm. Hudson, 503 U.S. at 6–7 (citations omitted). In Hudson, the Court also made clear that not “every malevolent touch by a prison guard gives rise to a federal cause of action.” 503 U.S. at 9. Instead, the Eighth Amendment’s prohibition against cruel and unusual punishments excludes from constitutional recognition de minimis uses of force that are not of the kind that would be “repugnant to the conscience of mankind.” Hudson, 503 U.S. at 9–10 (quoting Estelle v. Gamble, 429 U.S. 97, 106 (1976)). But the Court also clarified that “[w]hen prison officials maliciously and sadistically use force to cause harm, contemporary standards of decency always are violated. This is true whether or not significant injury is evident.” Hudson, 503 U.S. at 9. The extent of injury is relevant to the Eighth Amendment inquiry because it provides some indication of the amount of force applied, and because it may suggest whether the use of force was plausibly necessary in a particular situation; but the absence of serious injury is not the end of the analysis. Wilkins v. Gaddy, 559 U.S. 34, 37 (2010). 22 No. 17-2295 B. In this case, the threshold question of whether the force applied was more than de minimis is easily answered. The firing of two shotguns toward the crowd, or into the ceiling of a crowded dining hall no doubt qualifies as more than a de minimis use of force, as the district court found. See McCottrell v. Correctional Officers, No. 1:15-cv-03208, slip. op. at 9 (N.D. Ill. May 22, 2017) (finding that discharging a firearm is plainly more than a de minimis use of force). The prison itself had installed a “shot box” device on the ceiling to minimize the known risks of ricochet from warning shots discharged in that space, and both defendants aimed away from this safety device. The resulting injuries and the pain inflicted confirm that the force was significant. See Lewis, 581 F.3d at 475 (pain is the barometer by which we measure claims of excessive force, and the use of a stun gun, which sends an electric pulse through the body causing immobilization, disorientation, loss of balance, and weakness, is more than a de minimis application of force). Setting aside for a moment the plaintiffs’ claim that the guards may have fired directly toward the crowd, the dissent suggests that firing two shotguns loaded with buckshot into the ceiling of a crowded dining hall cannot be deemed to be malicious and sadistic or even characterized as an intentional application of force without a showing that a guard “intended to hit or harm someone with his application of force.” That standard is met here. A jury may infer intent to make physical contact from the nature of the act taken, and spraying buckshot into a crowded room, either directly towards people or by the known force of ricochet, is circumstantial evidence that gives No. 17-2295 23 rise to an inference that the guards intended to injure the persons seated below. As we discuss infra, both defendants averred that they fired into the ceiling, avoiding the device intended to minimize ricochet. That action of purposely avoiding the shot box gives rise to an inference that the guards intended to make physical contact. That they might not have cared which particular inmate (or how many inmates) they hit is not relevant to the question of whether they intentionally applied force. Discharging shotguns loaded with buckshot in a crowded dining hall was a “force applied” that was more than de minimis. From consideration of the Whitley factors, “inferences may be drawn as to whether the use of force could plausibly have been thought necessary, or instead evinced such wantonness with respect to the unjustified infliction of harm as is tantamount to a knowing willingness that it occur.” Whitley, 475 U.S. at 321. See also Farmer v. Brennan, 511 U.S. 825, 835–36 (1994) (in a claim for excessive force during a prison disturbance, the claimant must show that officials applied force maliciously and sadistically for the very purpose of causing harm, or that officials used force with a knowing willingness that harm occur). Firing a shotgun into the ceiling and away from the shot boxes in a crowded dining hall gives rise to an inference that the guards used force with a “knowing willingness” that harm would occur. Whitley, 475 U.S. at 321. Although Whitley afforded deference to prison officials during a disturbance, the Court did “not insulate from review actions taken in bad faith and for no legitimate purpose.” 475 U.S. at 322. A warning shot taken when there is no need to warn arguably has no legitimate purpose, and one purposely aimed away from a safety device raises an inference of bad 24 No. 17-2295 faith. But under the dissent’s version of the test, a guard could blindly unload a shotgun above a crowd of bystanders with impunity because making contact is not a certainty. Nothing in Whitley requires targeting. Whitley instead refers to a “security measure … undertaken to resolve a disturbance” and directs us to consider “whether the measure taken inflicted unnecessary and wanton pain and suffering,” a calculus that, in the prison disturbance setting, turns on “whether force was applied in a good faith effort to maintain or restore discipline or maliciously and sadistically for the very purpose of causing harm.” Whitley, 475 U.S. at 321. Firing buckshot toward a crowd of bystanders or into the ceiling above them (and away from the devices intended to reduce ricochet) may be comfortably characterized as a force applied during a security measure. See also Gomez v. Randle, 680 F.3d 859, 861–62, 864–65 (7th Cir. 2012) (bystander prisoner struck by buckshot fired into inmate population by guard on catwalk as guards on the ground sought to break up a fight between two unarmed inmates stated a claim for excessive force under the Eighth Amendment). The dissent also asserts that, in order to show that the guards intentionally shot the plaintiffs (as opposed to recklessly disregarding the risk of ricochet), the plaintiffs could have introduced evidence regarding the composition of the ceiling in order to demonstrate that ricochet was “very likely.” In that event, the dissent continues, “firing into the ceiling would be the functional equivalent of firing into the crowd.” Infra, at 6. There is no evidence in the record regarding the composition of the ceiling but there is evidence that the prison regarded the ceiling as a dangerous target: the prison itself No. 17-2295 25 installed shot boxes throughout the facility to reduce ricochet from warning shots. The installation of those boxes is evidence that the prison itself considered ricochet to be “very likely.” A jury may infer from the presence of the shot boxes that the guards were aware that a shotgun blast to the ceiling would ricochet into the crowd below. As we discuss below, in analyzing the fifth Whitley factor (efforts made to temper the severity of the force), the plaintiffs contend that “failure to hit the shot box might have been malicious.” Opening Brief at 26. The plaintiffs argued that the guns were either directed at the crowd or at the ceiling. They maintained that if the shots were fired into the ceiling, a reasonable jury could infer that the guards’ “failure to hit the shot box intended to catch buckshot was because they did not want the buckshot caught.” Opening Brief at 26. The plaintiffs also argued that a jury could find that the guards’ claims regarding the shot box (including their inability to hit it or their decisions to aim elsewhere, purportedly for safety reasons) were not credible, especially in light of their other misleading statements. Id. Thus, the plaintiffs do in fact cite circumstantial evidence that shooting at the ceiling (and away from the shot box) was functionally equivalent to sending buckshot into the crowd. This is more than recklessness, as the dissent asserts; a jury could conclude that this application of force “evinced such wantonness with respect to the unjustified infliction of harm as is tantamount to a knowing willingness that it occur.” Whitley, 475 U.S. at 321. It is obviously easier for the plaintiffs to prove that the guards met this standard if the guns were pointed at the crowd instead of the ceiling. But it is possible to prove that the standard was met even if the guns were pointed at the 26 No. 17-2295 ceiling and purposefully away from the shot box. From the evidence presented, the jury could infer both intent to make contact and malice. C. So we turn to the five factors that the Supreme Court set forth in Whitley, the controlling test for determining intent when more than de minimis force is applied in a prison disturbance.11 See Lewis, 581 F.3d at 476 (once the court has determined that the force applied was more than de minimis, the inquiry turns to the mind set of the individual applying the force). 1. The first factor is the need for the application of force, and on this point, there are material disagreements of fact. The plaintiffs produced evidence that the fight was nothing more than a minor shoving match between two unarmed combatants, and that the tussling inmates had been separated 11 The district court suggested that, in Fillmore v. Page, 358 F.3d 496 (7th Cir. 2004), the Seventh Circuit had “distilled” the analysis of Whitley and Hudson into a two-part inquiry of whether the force was more than de minimis, and whether the actions of the defendant officers were designed expressly for the purpose of punishing or humiliating the plaintiff inmates. Fillmore appropriately identified those as the operative, overarching questions in a case involving the use of force during a prison disturbance, but Fillmore did not purport to distill the five-part test set forth in Whitley. Fillmore in fact listed the five factors as the means of determining the officers’ intent and purpose in such a case. 358 F.3d at 504. Fillmore faithfully applied Whitley and Hudson when it affirmed a finding that the force applied was de minimis. Because that threshold had not been met, there was no need to progress through the five-part test to determine the officers’ intent. No. 17-2295 27 and subdued and were in the process of being handcuffed when the shots were fired. In the plaintiffs’ view, there was no need for the warning shots because the officers on the ground had already successfully used pepper spray to secure the situation. Some of the factual conclusions of the internal affairs report bolster this view. The report found that the fighting inmates had already been separated, that neither was armed with a weapon, that no staff members had been struck, and that the fighting inmates were brought to compliance with the application of pepper spray before the shots were fired. The defendant guards, on the other hand, told an evolving story about the sequence of events, at first reporting that the inmates were fighting and struggling with staff members who were trying to separate them at the moment the shots were fired, then later stating that the inmates had been separated before the shots were fired, and then finally that the inmates had been separated and only one inmate continued to struggle before the shots were then fired. A jury could infer from the shifting stories told by the guards that they were not truthful regarding the need for the application of force. See Reeves, 530 U.S. at 147; Ray v. Clements, 700 F.3d 993, 1025 n.7 (7th Cir. 2012) (witness’s changing story supports inference of lying). If the plaintiffs are correct about the sequence of events, then there was arguably no need for warning shots at all. Lewis, 581 F.3d at 477 (where the exact sequence of events leading to a stun gun’s use is strongly disputed, the court is required to view the facts in the light most favorable to prisoner opposing summary judgment). Unlike Whitley, there was no ongoing dangerous and volatile situation. See Lewis, 581 F.3d at 477–78 (where an inmate asserts that he was merely lying on his bunk, 28 No. 17-2295 weak and sluggish from a hunger strike, without sufficient time to respond to an order to stand, the inmate displayed no aggressive or threatening behavior that would have justified application of a stun gun without warning). But if the defendants’ various versions of the event can be reconciled and are correct, then from the guards’ perspective, the force could plausibly have been thought necessary. Whitley, 475 U.S. at 321; Wilkins, 559 U.S. at 37. The evidence supports an inference that the shots were fired after the fight on the floor was under control. Only a jury can resolve the factual dispute regarding the state of the conflict and the need for force at the time the shots were fired. But, construing the facts in favor of the plaintiffs, there was no need for any warning shots at the time they were fired, which gives rise to an inference of malice and sadism on the part of the guards. 2. The second factor, the relationship between the need for force and the amount of force employed, is closely tied to the first factor. Obviously, if there was no need for the warning shots, then those shots were significantly disproportionate to the need for force. The guards on the floor were already applying force, pulling the offenders apart and applying pepper spray when they did not comply with handcuffing procedures. Again, the stories told by the plaintiffs and the guards are materially different, and the prison itself concluded that the additional force applied by the defendants was not needed to secure the situation. That is not to say that the officers’ violation of prison policies regarding the use of force was evidence of an unconstitutional amount of force. We have found in the Fourth Amendment context, for example, that No. 17-2295 29 “the violation of police regulations or even a state law is completely immaterial as to the question of whether a violation of the federal constitution has been established.” Thompson v. City of Chicago, 472 F.3d 444, 454 (7th Cir. 2006). But the plaintiffs here are not attempting to equate a violation of department policy with a constitutional transgression. Rather, they are citing the factual conclusions of an official investigation into this very incident as support for their assertion that, because the fight had been resolved, there was no further need for the use of force, including in particular the warning shots fired by the defendants. See Daniel v. Cook County, 833 F.3d 728, 740 (7th Cir. 2016). As we noted in footnote 6 above, the district court concluded that the report was admissible under Federal Rule of Evidence 803(8), as “factual findings from a legally authorized investigation,” a conclusion that the defendants did not contest on appeal. Thus, the plaintiffs are free to use the factual findings of the report to demonstrate that there are genuine issues of material fact that must be resolved at trial. It is not the prison’s policies or the guards’ violations of those policies that are relevant. Only the factual findings may be considered, and those findings bolster the plaintiffs’ evidence that the shots were fired after the conflict was over, at a time when there may have been “no plausible basis for the officials’ belief that this degree of force was necessary.” Whitley, 475 U.S. at 323. See also Lewis, 581 F.3d at 478 (in light of the absence of any agitation or threat from an inmate, the short passage of time between a single order and the application of a stun gun, and a lack of warning as to the consequences of a failure to comply with the order, summary judgment in favor of the 30 No. 17-2295 defendant officer is not appropriate). Construing the facts in favor of the plaintiffs, the force applied was grossly disproportionate to the force that could plausibly have been thought necessary, again giving rise to an inference of malice and sadism. 3. We next consider the extent of injury inflicted. McCottrell and Clay each suffered physical and psychological wounds caused by the buckshot. In Clay’s case, the wound had to be closed with stitches and the buckshot remains in his arm. McCottrell bled from the wounds in his neck and leg, and required pain medication. Both men also suffered psychological trauma from the event and sought treatment for their emotional injuries. “Injury and force … are only imperfectly correlated, and it is the latter that ultimately counts.” Wilkins, 559 U.S. at 38. The defendant guards do not dispute the extent of the injuries or the nature of the force applied, and we can safely say that the injuries and the force employed were significant enough to bring the case within the constitutional realm. 4. The fourth factor is the extent of the threat to the safety of staff and inmates, as reasonably perceived by the responsible individuals on the basis of the facts known to them. The evidence on this point is also mixed. The guards presented evidence that the dining hall is the site of frequent fights, that makeshift weapons sometimes appear during these incidents, that the guards on the floor are vastly outnumbered by the inmates in the hall, and that prison fights can escalate quickly. No. 17-2295 31 No doubt even a shoving match between two inmates can quickly turn into something much more serious. The undisputed evidence demonstrated that even after guards were able to pull the inmates apart, they continued to struggle for some period of time until pepper spray brought about compliance. From beginning to end, the entire incident lasted less than a minute and the guards in the tower had to decide very quickly whether the warning shots were necessary. Under the standards set forth in Whitley, this factor initially appears to favor White and Williams. At the same time, other witnesses reported that the fight was over by the time the shots were fired, that no weapons had materialized, that no other inmates joined the fracas, and that no staff had been attacked. Unlike Whitley, there was no prison riot, no hostage, and no armed agitator. In combination with the finding that White and Williams gave inaccurate and evolving accounts of the sequence of events, a jury could infer that their perception of the threat was not reasonable. If the defendants saw that the fight was over and that the guards on the floor had the situation under control, then there was no penological justification for firing warning shots, giving rise to an inference of malice and sadism. Whitley, 475 U.S. at 322 (deference to prison officials does not insulate from review actions taken in bad faith and for no legitimate purpose); Fillmore, 358 F.3d at 504 (infliction of pain that is totally without penological justification is per se malicious). 32 No. 17-2295 5. The fifth and final factor is whether any efforts were taken to temper the severity of a forceful response. This factor is also disputed. As we discussed above, there is a reasonable inference that the guards fired directly toward the bystander inmates. Supra at p.4–9. That would indicate that little or no effort was taken to temper the severity of the force. Even the defendants’ evidence, as construed in favor of the plaintiffs, demonstrates that White and Williams purposely fired into the ceiling rather than the shot box, the device intended to reduce injuries from ricochet. Williams submitted an affidavit during summary judgment proceedings in which he stated that he “discharged a warning shot into the ceiling toward the nearest shot-box.” R. 75-4, at 2. Williams did not indicate whether he was trying to hit the shot box and missed, or purposely aimed at the ceiling, but the evidence raises an inference that he purposely shot at the ceiling. In his official “Incident Report,” Williams wrote that he “fired one warning round into the ceiling.” R. 85, at 54. In his interview with the investigator, he also stated that he fired a round “into the ceiling.” R. 85, at 52. Only in his affidavit did he add the detail that he fired the shot “toward the nearest shot-box.” Construing this account in favor of the plaintiffs, Williams chose the ceiling rather than the shot box for the warning shot. As discussed above, the plaintiffs also raised a reasonable inference that the shots were directed toward the inmates. White conceded that he purposely “discharged a warning shot into the ceiling,” rather than at the shot box. He claimed that, prior to this incident, he “had been told by [his] colleagues that shooting into the ceiling, as opposed to the shot No. 17-2295 33 box, decreased the chance of ricochet.” R. 75-5, at 2–3. He asserted that he fired into the ceiling “in order to minimize the chance that anyone would be struck by ricocheted buckshot.” R. 75-5, at 3. In any event, this evidence, construed in favor of the plaintiffs, indicates that both White and Williams purposely fired into the ceiling rather than the shot box. If the jury determines that the shots were directed at the ceiling instead of toward the inmates, the jury must also assess the credibility of White’s purported reason for his choice. Williams has yet to explain why he did not fire into the shot box. Negligence (even gross negligence) is not actionable but neither defendant has claimed that he mistakenly missed the shot box. A jury could conclude that purposely shooting toward the inmates or into the ceiling of a crowded dining hall rather than the shot box “evinced such wantonness with respect to the unjustified infliction of harm as is tantamount to a knowing willingness that it occur.” Whitley, 475 U.S. at 321. Moreover, purposely avoiding the shot box served no penological purpose. That is, a warning shot aimed at the shot box fulfills the purpose of a warning shot, and no additional penological value is gained by purposely avoiding this safety device, again giving rise to an inference of malice and sadism. See Fillmore, 358 F.3d at 504. Aiming a warning shot into the shot box was the course for the guards if the purpose of the shot was a good faith effort to restore order. As the plaintiffs argue, purposely avoiding the shot box gives rise to an inference that the guards “did not want the buckshot caught.” Opening Brief at 26. 34 No. 17-2295 D. There are a number of contested issues of material fact that preclude the entry of summary judgment at this stage. To summarize, a finder-of-fact must first determine whether the guards intended to hit the inmates with buckshot when they discharged their shotguns over the crowded dining hall, a calculus that depends in part on the disputed direction of fire. Whether the guards fired in the direction of the inmates or at the ceiling, there are genuine disputes of material fact for virtually every factor in the Supreme Court’s five-part test, and we must vacate the judgment and remand for further proceedings.12 The facts, construed in favor of the plaintiffs, support an inference that the defendants acted maliciously and sadistically rather than in good faith to restore order. If the jury found both that the brawling inmates were subdued before the shots were fired and that the defendants perceived as much, then the jury could find that by purposely discharging shotguns toward the crowd or into the ceiling (rather than toward the shot box), the defendants acted maliciously and sadistically for the purpose of causing harm, and did so at a time when there was no need for any force. The jury would have to focus on what the defendants could see and actually did see before they discharged their firearms. But on this record, we cannot rule 12 In the final sentence of their brief on appeal, the defendants also assert that they were “at least entitled to qualified immunity.” Defendants’ Brief at 31. Perfunctory and undeveloped arguments are waived, and we need not consider this single-sentence argument further. Tobey v. Chibucos, 890 F.3d 634, 652 (7th Cir. 2018). No. 17-2295 35 out the possibility that the defendants saw that the fight was over, and that the combatants had been separated and subdued before the shots were fired. Failing to accurately depict the event in official reports and failing to aim for the very device intended to protect bystanders are facts that weigh in favor of the plaintiffs’ view that the officers’ actions were not a good faith effort to restore order but rather were undertaken maliciously and sadistically for the very purpose of causing harm. Hudson, 503 U.S. at 6; Whitley, 475 U.S. at 320–21. A jury would not be compelled to find that the officers acted with that intent, but it could so find. III. We acknowledge that the Supreme Court called for deference to prison officials making split-second decisions during disturbances. But a jury must determine whether the shots were fired during an ongoing struggle that threatened staff and other prisoners, or after the struggle was over. Whitley, 475 U.S. at 320–21. See also Lewis, 581 F.3d at 478 (although it is not difficult to imagine any number of scenarios that would justify the immediate and unadvertised use of summary force, where the inmate’s version of the facts calls into question the guard’s state of mind at the time force is applied, summary judgment is not appropriate). And a jury must determine why the officers chose to fire toward the crowd or into the ceiling rather than the shot box. In short, there are significant factual disputes that affect the analysis of every one of the five Whitley factors. We may not simply credit the claims of White and Williams that they believed the shots were necessary to restore order and defer to that claim when there is evidence that appears to contradict their assertion of 36 No. 17-2295 good faith. See McGreal v. Ostrov, 368 F.3d 657, 677 (7th Cir. 2004) ("It is rarely appropriate on summary judgment for a district court to make a finding on state of mind."); Stumph v. Thomas & Skinner, Inc., 770 F.2d 93, 97 (7th Cir. 1985) (“‘Summary judgment is notoriously inappropriate for determination of claims in which issues of intent, good faith and other subjective feelings play dominant roles.’”) (quoting Pfizer, Inc. v. International Rectifier Corp., 538 F.2d 180, 185 (8th Cir.1976), cert. denied, 429 U.S. 1040 (1977)). Summary judgment is not appropriate when there is evidence in the record that could support a reasonable jury finding that the defendants acted maliciously and sadistically for the very purpose of causing harm rather than in a good faith effort to restore order. Anderson, 477 U.S. at 255–56; Whitley, 475 U.S. at 320–21. VACATED AND REMANDED. No. 17-2295 37 BARRETT, Circuit Judge, dissenting. As the plaintiffs briefed and argued this case, they would have lost it. They argued that the defendants violated the Eighth Amendment by dis- charging their weapons into the ceiling, rather than into the shot box, immediately after the fight had been broken up and there was no plausible need for that use of force. But they did not argue—and had no evidence to prove—that the defend- ants intentionally hit anyone. That deficit should have been fatal, because if the officers did not intend to hit anyone, they could not have done so “maliciously and sadistically,” as Whitley v. Albers requires. 475 U.S. 312, 320 (1986). The guards may have acted with deliberate indifference to inmate safety by firing warning shots into the ceiling of a crowded cafeteria in the wake of the disturbance. In the context of prison disci- pline, however, “deliberate indifference” is not enough. The Supreme Court has drawn a clear distinction between the standard applicable to claims challenging the conditions of confinement and the standard applicable to claims chal- lenging the use of excessive force. In Farmer v. Brennan, it ex- plained that prisoners challenging the conditions of their con- finement must show that officials were deliberately indiffer- ent to an excessive risk to inmate health or safety, and that “deliberate indifference to a substantial risk of serious harm to a prisoner is the equivalent of recklessly disregarding that risk.” 511 U.S. 825, 836 (1994). Put differently, deliberate in- difference means criminal recklessness. Id. at 839–40 (“[S]ub- jective recklessness as used in the criminal law is a familiar and workable standard that is consistent with the Cruel and Unusual Punishments Clause as interpreted in our cases, and 38 No. 17-2295 we adopt it as the test for ‘deliberate indifference’ under the Eighth Amendment.”). The Court emphasized, however, that “‘application of the deliberate indifference standard is inappropriate’ in one class of prison cases: when ‘officials stand accused of using exces- sive physical force.’” Id. at 835 (citation omitted); see also Whit- ley, 475 U.S. at 320 (holding that the “deliberate indifference” standard does not apply in the context of prison security). In that context, “the question whether the measure taken in- flicted unnecessary and wanton pain and suffering ultimately turns on ‘whether force was applied in a good faith effort to maintain or restore discipline or maliciously and sadistically for the very purpose of causing harm.’” Whitley, 475 U.S. at 320–21 (citation omitted). The Court has described this as a “very high state of mind” requirement—one much higher than the criminal recklessness necessary to show deliberate indifference. Farmer, 511 U.S. at 836. The Whitley factors are designed to evaluate this “very high state of mind” by smok- ing out why the guard inflicted harm: was it to maintain dis- cipline or for the satisfaction of hurting the prisoner? A guard who injures a prisoner by using force in the absence of a rea- sonable belief that a threat exists and without any effort to temper the severity of the force can be thought to knowingly will “the unjustified infliction of harm.” Whitley, 475 U.S. at 321 (emphasis added). In other words, those circumstances sup- port an inference that the guard acted in extreme bad faith rather than “in a good faith effort to maintain or restore disci- pline.” Id. at 320. No. 17-2295 39 An inmate cannot satisfy the “malicious and sadistic” standard without showing that a guard intended to hit or harm someone with his application of force. After all, if the guard did not intentionally apply force to a prisoner, how could he have had a malicious and sadistic intent to cause him pain? In the absence of an intent to exert force on someone, there is nothing for the Whitley factors to probe.1 That is pre- sumably why the cases treat the intent to make physical con- tact as a given—I have been unable to find any case, in either the Supreme Court or our circuit, in which an excessive force claim did not involve force directly and intentionally applied to the plaintiff. (That is true, incidentally, of both successful and unsuccessful claims.) In Whitley, a guard deliberately shot the plaintiff in the knee, following orders to “shoot low at any prisoners climbing the stairs” behind an officer engaged in an effort to free a hostage. 475 U.S. at 316. In Hudson v. McMillian, the guard “punched Hudson in the mouth, eyes, chest, and stomach while [another guard] held the inmate in place and kicked and punched him from behind.” 503 U.S. 1, 4 (1992). See also Wilkins v. Gaddy, 559 U.S. 34, 35 (2010) (the officer “slammed [the plaintiff] onto the concrete floor” and “pro- ceeded to punch, kick, knee, and choke [him].”). Our own ex- cessive-force cases have been brought against officers who 1 That is not to say that the guard must intend to exert force on the partic- ular inmate who is injured. See Gomez v. Randle, 680 F.3d 859, 862, 864–65 (7th Cir. 2012) (reversing the dismissal of a claim against a guard who fired into a crowd rather than at one specific inmate). The point is that the guard must maliciously and sadistically exert force on someone. And the officer in Gomez, who fired “into the inmate population,” id. at 862, exerted force on the inmates at whom he fired. 40 No. 17-2295 intentionally shot, tasered, physically assaulted, or pepper sprayed prisoners. See, e.g., Rice ex rel. Rice v. Corr. Med. Servs., 675 F.3d 650, 668 (7th Cir. 2012) (pepper spray); Lewis v. Downey, 581 F.3d 467, 475–78 (7th Cir. 2009) (taser); Fillmore v. Page, 358 F.3d 496, 501–05 (7th Cir. 2004) (assault); Kinney v. Indiana Youth Center, 950 F.2d 462, 464–65 (7th Cir. 1991) (gun- shot). Faced with this hole in the plaintiffs’ case, the majority gives them an assist by changing the case. It offers two theo- ries of why the plaintiffs can show that the guards intended to shoot the inmates. First, it insists that the guards might have shot directly into the crowd of inmates, rather than at the ceiling. Second, it suggests that the guards might have in- tentionally tried to hit the inmates by bouncing buckshot off of the ceiling. Yet the plaintiffs did not raise the first argument at all before us, and the second argument consists of one oblique sentence in their brief. And in any event, there is in- sufficient evidence in the record to permit a reasonable juror to draw either conclusion. In the district court, the plaintiffs relied on the hearsay tes- timony of two other inmates to support the claim that the guards shot directly into the crowd. The defendants, in con- trast, swore in their affidavits that they aimed at the ceiling. The district court resolved that issue against the plaintiffs. It stated: Plaintiffs have speculated that the Officer De- fendants were aiming at them, but it is undis- puted that they did not see where the Officer Defendants were aiming and there is no No. 17-2295 41 admissible evidence that would create a genu- ine factual dispute regarding where the Officer Defendants aimed. … Plaintiffs have attempted to create a factual dispute by asserting that two inmates, identified only as “Rico” and “Fuzz,” told McCottrell that they observed one of the Officer Defendants aiming at the inmates. This statement, however, is inadmissible hearsay, and the Court may not consider it at the sum- mary judgment stage. As a result, there is no ev- idence suggesting that the Officer Defendants aimed at the inmates. Memorandum Opinion and Order at 12 (record citations omitted). The plaintiffs have not challenged either the district court’s evidentiary ruling or its conclusion that the plaintiffs had failed to create a factual dispute on this issue. That was a prudent choice. The district court’s resolution of the hearsay issue was plainly correct, and without that testimony, the plaintiffs have no evidence that the officers shot into the crowd. It bears emphasis that this is so even if the officers lied in their affidavits about the direction of the shots. A plaintiff cannot discharge her burden simply by tearing the defend- ant’s case down; “[i]nstead, the plaintiff must present affirm- ative evidence in order to defeat a properly supported motion for summary judgment.” Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 256 (1986). Although the plaintiffs do not ask us to reverse this ruling, the majority, reading between the lines of the brief, maintains 42 No. 17-2295 that “the plaintiffs’ arguments … encompass both the possi- bility that the shots were fired into the ceiling ... or toward the inmates.” Maj. Op. at 8. Yet in the absence of an explicit fac- tual assertion that the officers fired into the crowd, much less an explicit request that we reverse the district court, it is safe to say that the plaintiffs have waived this argument on ap- peal. The majority resurrects the issue on its own, and—fault- ing the district court for failing to appreciate the significance of circumstantial evidence—concludes that the plaintiffs can rest on the fact that shots hit them to prove that the guards fired at them. That argument should have lost even if the plaintiffs had made it. One can draw two plausible inferences from the circumstantial evidence of the plaintiffs’ injury: the buckshot hit them either directly or by ricochet. The officers’ affidavits and the prison’s internal affairs report, which as- serts that the “[i]nmates were struck with pellets from the ric- ochet” of warning shots, support the ricochet theory.2 The plaintiffs introduced no admissible evidence in support of the 2 The majority minimizes the report’s repeated statements that the plain- tiffs were injured by ricochet, asserting that these statements reflect no more than an assumption about the direction of the shots. Maj. Op. at 6 n.4. It was not an assumption. The report concludes by stating that “[a]fter a total review of the evidence, statements, reports, and all available facts,” the investigators determined that White and Williams “used an unreason- able amount of force (warning shot)” in violation of Department policy. (Emphasis added). A “warning shot” is one intentionally aimed away from people. The report does not identify a single piece of evidence sug- gesting that there was direct fire, and such evidence would have been im- portant, to say the least, in an investigation of whether the officers used unreasonable force. No. 17-2295 43 “direct hit” theory.3 There is therefore insufficient evidence to permit a reasonable juror to find that fact in the plaintiffs’ fa- vor. As for the claim that the guards indirectly aimed at the plaintiffs by intentionally bouncing buckshot off the ceiling and into the crowd: this too is an argument that the plaintiffs do not develop and that the evidence does not support. To be sure, the plaintiffs contend that the guards fired into the ceil- ing and that the buckshot then ricocheted and hit them. But to show that the guards intentionally shot them—as opposed to recklessly disregarding the risk that the buckshot would ricochet—the plaintiffs needed to introduce some evidence that guards were intentionally, albeit indirectly, aiming to hit 3 According to the majority, my view is that “a court would be required to conclusively credit a witness’s sworn statement that it was not raining even in the face of the proverbial wet umbrella.” Maj. Op. at 7. That is certainly not my view. The overwhelming inference raised by the circum- stantial evidence of a wet umbrella is that it was raining outside; the infer- ence is nowhere close to as powerful here. I’m also unsure why the major- ity invokes “the force with which the buckshot arrived … and the number of persons injured” as additional circumstantial evidence that the guards aimed at the prisoners rather than at the ceiling. Maj. Op. at 5. As an initial matter, the record contains no evidence about the force with which the bullets hit the inmates, and the incident report, which determined that the inmates’ injuries were “minor,” is inconsistent with the proposition that the shots hit the inmates with great force. Moreover, the force of the hit and the number of injuries support the “direct hit” theory only if those two factors would be different with a hit from ricochet. Maybe they would be. To carry their burden on summary judgment, though, the plaintiffs would have had to introduce evidence to support that. 44 No. 17-2295 them.4 For example, the plaintiffs could have introduced evi- dence that the ceiling was composed of a material that made ricochet very likely; in that event, firing into the ceiling might be the functional equivalent of firing into the crowd. Rather than introducing such evidence, however, plaintiffs made the opposite argument: they insisted (as part of their effort to show that the guards shot directly into the crowd) that it is “‘unlikely that both defendants fired into [the] ceiling’ be- cause the ceiling is acoustic tile and would not have caused a ricochet.” Memorandum Opinion and Order at 5 n.3. The most that the plaintiffs say to support a “deliberate ricochet” theory is that “a reasonable jury could [] determine that Defendants’ decision to shoot in the first place and their failure to hit the shot box intended to catch buckshot was be- cause they did not want the buckshot caught.” The guards’ failure to hit the shot box is certainly evidence in the plaintiffs’ favor. But without any evidence of how likely ricochet was from the ceiling, the comparatively lower risk of ricochet from the shot box would not permit a reasonable jury to find that 4 The majority says that under my reading of Whitley, “a guard could blindly unload a shotgun above a crowd of bystanders with impunity be- cause making contact is not a certainty.” Maj. Op. at 24. But those facts— a guard emptying his shotgun above a crowd of bystanders—raise an in- ference that the guard intended for a bullet to hit someone. The facts in this case—two guards each firing a warning shot into the ceiling—do not, standing alone, raise the same inference. It is also worth observing that a guard who shoots a gun outside the context of prison discipline is subject to the “criminal recklessness” standard of Farmer v. Brennan, and there is no doubt that unloading a shotgun above a crowd of bystanders would be criminally reckless. 511 U.S. 825, 839-40 (1994). No. 17-2295 45 the officers pinballed buckshot off the ceiling with the intent to hit a prisoner. If the plaintiffs could win by showing that the guards reck- lessly put them at risk by firing warning shots into the ceiling after the fight on the floor was under control, I would agree that they could survive summary judgment. But because that is not the standard, I respectfully dissent.