United States v. Marco Proano

In the United States Court of Appeals For the Seventh Circuit ____________________ No. 17‐3466 UNITED STATES OF AMERICA, Plaintiff‐Appellee, v. MARCO PROANO, Defendant‐Appellant. ____________________ Appeal from the United States District Court for the Northern District of Illinois, Eastern Division. No. 1:16‐cr‐00590‐1 — Gary Feinerman, Judge. ____________________ ARGUED OCTOBER 23, 2018 — DECIDED JANUARY 7, 2019 ____________________ Before KANNE, HAMILTON, and ST. EVE, Circuit Judges. ST. EVE, Circuit Judge. On‐duty officer Marco Proano fired sixteen shots at a moving sedan filled with teenagers until the car idled against a light pole. He hit two passengers. The gov‐ ernment charged Proano with two counts of willful depriva‐ tion of constitutional rights, one for each injured passenger, and a jury convicted on both counts. 18 U.S.C. § 242 . Proano appeals, claiming both pretrial and trial errors. We affirm. 2 No. 17‐3466 I. Background A. The Shooting Around 5:00 p.m. on December 22, 2013, two Chicago Po‐ lice Department (CPD) officers, Ken Flaherty and Jonathan Morlock, stopped a gray Toyota Avalon on Chicago’s southside. The Toyota had just sped out of an alley. The driver fled on foot, leaving one passenger in the front seat and four or five (as far as Flaherty could tell) in the backseat. Morlock pursued the driver; Flaherty stayed with the Toyota. Before fleeing, the driver did not, apparently, put the car in park, and it rolled toward Flaherty and his squad car. The Toyota wedged itself between Flaherty’s squad car and another car parked on the street. Jaquon Grant had been in the passenger seat. He too tried to escape as the car rolled forward, but his legs got stuck be‐ tween Flaherty’s squad car and the Toyota. Grant tried to break free, and Flaherty assured him that when backup ar‐ rived Flaherty would assist him. Flaherty shouted commands to the other passengers—“stay still,” “quit moving”—but they did not obey. One passenger, thirteen‐year‐old Kevon Brown, attempted to flee, but stopped while hanging out of an open backseat window, with his head above the roof. Fla‐ herty dispatched for backup. Moments later, Proano and his partner, Guy Habiak, ar‐ rived in their squad car. Proano exited the car, with his weapon in one hand, cocked, and aimed at the Toyota. Sec‐ onds later, Delquantis Bates, who had been in the back seat of the Toyota, reached over the center console and pressed his hand on the gas pedal. The car, still wedged, revved but did not move. Bates then put the car in reverse and pressed the No. 17‐3466 3 pedal again. The Toyota jolted free and began to reverse. No one was in its path. As the car retreated, three things happened: a metal BB gun fell to the ground from the Toyota, Grant freed himself, and Proano began shooting at the Toyota. Flaherty quickly apprehended Grant. Habiak picked up the gun and handed it to Flaherty, saying “Gun. Here’s the gun. Here’s the gun.” And Proano continued to shoot at the Toyota as it stopped, pivoted, and rolled forward into a light pole. Ten of Proano’s sixteen bullets entered the Toyota. One bullet hit Bates’s shoulder, while others grazed his face. Two bullets hit another passenger, David Hemmans, in his leg and foot. No other of‐ ficer fired his weapon. After the shooting, Proano completed two Tactical Re‐ sponse Reports, forms upon which the CPD relies to docu‐ ment use‐of‐force incidents. On both forms, Proano admitted to firing his weapon sixteen times. He indicated that he did so because an “assailant” presented an “imminent threat of bat‐ tery” and so he “use[d] force likely to cause death or great bodily harm.” The assailant’s weapon, according to Proano’s reports, was an “automobile.” Proano did not identify the BB gun as a contributing factor in his decision to shoot in the re‐ ports. Still on the scene, however, Proano informed CPD de‐ tective Stanley Kalicki that he heard one of the other on‐site officers identify a gun. Proano also told Kalicki that he fired his weapon because he feared for Brown, who, according to Proano, “was being dragged” by the Toyota. Months later, Proano discussed the shooting with the In‐ dependent Police Review Authority (IPRA)—a now‐defunct body, which at the time investigated allegations of police 4 No. 17‐3466 misconduct.1 In March 2015, IPRA Investigator Dennis Prieto met with FBI agents to discuss the shooting. No one involved in that meeting believed that they discussed Proano’s state‐ ments to IPRA. The FBI agents received material from Prieto, including documents reflecting Proano’s statements to IPRA investigators. The FBI agents then passed those documents to the government’s “filter team” for review. B. The Prosecution On September 15, 2016, a grand jury returned a two‐count indictment against Proano for willfully depriving Bates and Hemmans of their constitutional rights—namely, their Fourth Amendment right to be free from unreasonable force—in vi‐ olation of 18 U.S.C. § 242 . Proano filed a motion to dismiss the indictment. He ar‐ gued that the FBI agents’ meeting with Investigator Prieto led to the disclosure of statements protected by Garrity v. New Jer‐ sey, 385 U.S. 493 (1967) (a decision we explain below), and that the disclosure tainted the prosecution against him. The dis‐ trict court held a hearing, in which Prieto and the FBI agents, Larissa Camacho and Eugene Jackson, testified. The district court then denied Proano’s motion on two independent grounds. First, it found no evidence of “seepage or taint” of Proano’s Garrity‐protected statements in the prosecution. Sec‐ ond, the district court found “legitimate independent sources” for any information the government could have gleaned from Proano’s Garrity‐protected statements, — 1 — Because Proano’s statements to IPRA are immune under Garrity v. New Jersey, 385 U.S. 493 (1967), and because resolution of this appeal does not require delving into the specifics of Proano’s statements, we are intention‐ ally vague about their contents. No. 17‐3466 5 assuming the government had seen them. The case proceeded to trial. Before trial, Proano moved in limine to exclude evidence relating to his training on CPD policies and procedures. He submitted that such evidence was irrelevant and that the gov‐ ernment’s witnesses—Sergeant Larry Snelling and officer Vincent Jamison, neither of whom recalled training Proano— lacked sufficient personal knowledge to testify regarding the training Proano received. The government filed a reciprocal motion in limine, asking the court to exclude evidence of Proano’s training on when use of force was appropriate under state law. The district court resolved the relevance of both pieces of evidence with the same stroke—both parties could use their respective training‐related evidence either to help prove or disprove that Proano acted willfully. Proano, specif‐ ically, could argue that his actions comported with his state‐ law training, and so he thought them reasonable; the govern‐ ment, meanwhile, could argue that Proano’s actions violated his CPD training, which could be relevant to his state of mind. The district court reserved ruling on the foundational ques‐ tion. It also deferred on resolving questions about the rele‐ vance of certain training‐related evidence, such as whether Proano was trained not to fire into a crowd or buildings. Trial began on August 21, 2017, and lasted six days. Among other government witnesses, Flaherty, Bates, and Brown described what occurred the night of the shooting; Kalicki explained Proano’s statements after the shooting; and CPD Sergeant Timothy Moore detailed Proano’s post‐shoot‐ ing reports. Snelling and Jamison also testified. Snelling described the CPD’s use‐of‐force policies, which he taught at the academy. He could not testify with certainty 6 No. 17‐3466 that he taught Proano. He was, however, familiar with the training that recruits received when Proano was a recruit at the academy, because trainers used a “common curriculum” and often sat in on each other’s classes to ensure consistent messaging. Regarding the CPD’s training, Snelling testified that use of deadly force is appropriate only when an assailant is likely to cause death or serious physical injury. Snelling tes‐ tified that recruits learn not to shoot into buildings, windows, or openings without clear visibility, and they learn not to shoot into crowds. Snelling also testified that recruits learn not to fire at a moving vehicle unless doing so is necessary to protect the life of another. Jamison testified about firearms training, which he over‐ saw at the academy. He too could not recall whether he per‐ sonally trained Proano, but Jamison stated that every firearms instructor at the academy works from the same preapproved lesson plans. Jamison testified that recruits learn to hold a weapon with two hands absent necessary circumstances, and they learn never to point a gun merely as a show of force. CPD firearms instructors also taught recruits to stop and “assess” whether a threat is ongoing after firing a few shots, though he also testified that recruits learn to shoot “to eliminate the threat.” In addition to this testimony, the government introduced video footage of the shooting. The footage, taken from Proano’s dashcam, showed the shooting unobstructed and in its entirety. The jury saw real‐time and slow‐motion versions of the footage. At the close of evidence, the parties discussed and debated jury instructions. Only one instruction‐related debate is rele‐ vant here: Proano proposed a (lengthy) instruction on No. 17‐3466 7 “willfulness,” the necessary mens rea for § 242, but the district court rejected it as redundant and confusing. The district court, instead, instructed the jury on willfulness using an in‐ struction it crafted with the parties’ input. After closing arguments and deliberation, the jury con‐ victed Proano on both counts. The district court later denied Proano’s posttrial motion, ruling in part that the court had properly admitted Snelling’s and Jamison’s testimony and that the government had laid an adequate foundation for their respective testimony. The district court then sentenced Proano to sixty months in prison. This appeal followed. II. Discussion Proano challenges four issues on appeal: (1) the denial of his Garrity motion; (2) the admission of training and policy evidence; (3) the accuracy of the jury instruction on willful‐ ness; and (4) the sufficiency of the evidence. We take each challenge in turn. A. The Garrity Motion Proano first claims that the government violated his rights under Garrity. The Fifth Amendment assures defendants that they will not be compelled to testify against themselves. Gar‐ rity expounded upon that general right in a particular context, that of public‐official investigations. Garrity held that when a public official must choose between cooperating in an internal investigation or losing his job, the statements he makes dur‐ ing the investigation are compelled, and, as such, they cannot later be used against the official in a criminal trial. 385 U.S. at 500 . In deciding whether the government violated Proano’s rights under Garrity, we review the district court’s legal 8 No. 17‐3466 conclusions de novo and its factual findings for clear error. United States v. Cozzi, 613 F.3d 725, 728 (7th Cir. 2010). Proano’s Garrity challenge does not get far. He spoke to IPRA under the threat of job loss, and his statements were thus compelled and Garrity‐protected. But after that his chal‐ lenge fails, in three different ways. First, federal investigators and prosecutors cannot misuse Garrity‐protected statements if they are never exposed to the statements. See id. at 732 . The government, in this case, set up a filter team to receive and review IPRA’s materials. The filter team then redacted any protected statements before handing the materials over to the prosecution team. No evidence sug‐ gests that this process was flawed or that Garrity‐protected statements slipped through. The district court also found that, despite some ambiguous evidence, the FBI agents and the IPRA investigator did not discuss Proano’s statements during the March 2015 meeting. As a result, the district court found that there was no “seepage or taint” of Proano’s Garrity‐pro‐ tected statements to the FBI agents or the prosecution team. Proano’s only challenge to this conclusion of fact is to rehash evidence thoroughly considered and weighed by the district court. That is not clear error. Second, even if the prosecution could have accessed Proano’s protected statements, there is no constitutional vio‐ lation if the government can establish “a legitimate source wholly independent of the compelled testimony” for the evi‐ dence. Kastigar v. United States, 406 U.S. 441, 460 (1972); see also United States v. Velasco, 953 F.2d 1467, 1474 (7th Cir. 1992). Here, the district court specifically found that the dashcam video, other witness accounts, and police reports all provided independent bases from which the prosecution could have No. 17‐3466 9 learned of the facts Proano described in his Garrity‐protected statements. Proano makes no attempt to demonstrate that these findings were clearly erroneous. Third, Proano misunderstands Garrity’s protections. He argues a syllogism: Investigator Prieto knew of the Garrity‐ protected statements, and Prieto’s meeting with the FBI agents prompted the federal investigation; thus, the investi‐ gation made derivative use of the Garrity‐protected state‐ ments. This reasoning is interrupted by the district court’s findings that Prieto did not disclose Garrity‐protected state‐ ments and that, even if he did, there were independent sources for the information. What survives of Proano’s argu‐ ment is only the theory that Prieto tainted the prosecution with Garrity‐protected statements simply by knowing of the statements and meeting with FBI agents. That does not fol‐ low, as a matter of logic or law. As we said in Cozzi, we “are not concerned with how” an investigator who knows of Gar‐ rity‐protected statements “may have influenced the federal investigation, but rather how [the defendant’s] statements in‐ fluenced the investigation.” Cozzi, 613 F.3d at 731 (emphasis in original). The district court did not clearly err in concluding that Proano’s statements did not reach the investigators and prosecutors. B. Admissibility of the Training and Policy Evidence Proano next claims that the district court made three er‐ rors in admitting evidence of his training and the CPD’s poli‐ cies. He argues: (1) the evidence was irrelevant; (2) it was un‐ fairly prejudicial and confusing; and (3) Snelling and Jamison did not establish an adequate foundation to testify to the evi‐ dence. We review evidentiary rulings for an abuse of discre‐ tion. United States v. Parkhurst, 865 F.3d 509, 513 (7th Cir. 2017). 10 No. 17‐3466 We will reverse a ruling only if no reasonable person would agree with the district court’s view. United States v. Ajayi, 808 F.3d 1113, 1121 (7th Cir. 2015).2 1. Relevance Under Rule 401 Proano’s position on the relevance of the training and pol‐ icy evidence has evolved during this appeal. In his papers, he thought such evidence irrelevant to show intent as a matter of law. At oral argument, he submitted that the specific evidence used at trial was inadmissible. Both positions are mistaken. To be admissible, evidence must be relevant. Fed. R. Evid. 402. To be relevant, evidence must tend to make a fact of con‐ sequence at trial more or less probable. Fed. R. Evid. 401. This is a “low threshold.” Tennard v. Dretke, 542 U.S. 274, 285 (2004); see also United States v. Boros, 668 F.3d 901, 907 (7th Cir. 2012). The Federal Rules of Evidence do not permit only decisive, controlling, or the “most” probative evidence. United States v. McKibbins, 656 F.3d 707, 711 (7th Cir. 2011). If evidence can help jurors answer the questions they must ask, the Rules per‐ mit its admissibility in the absence of a rule or law to the con‐ trary. Fed. R. Evid. 402; United States v. Causey, 748 F.3d 310, 316 (7th Cir. 2014). The government charged Proano with violating 18 U.S.C. § 242 . Section 242 prohibits the willful deprivation of rights under color of law, and individuals (Bates and Hemmans in‐ cluded) have the Fourth Amendment right to be free from un‐ reasonable uses of deadly force. Tennessee v. Garner, 471 U.S. 1 , 10–11 (1985). The two most pressing questions for the jury, — 2 — Because the district court did not abuse its discretion in admitting evi‐ dence of Proano’s training, for reasons that follow, we need not consider whether any error was harmless. Fed. R. Crim. P. 52(a). No. 17‐3466 11 then, were whether Proano used deadly force unreasonably and, if so, whether he did so willfully. The district court ruled that Proano’s intent—willful or not—was the fact that could be made more or less probable by evidence of Proano’s train‐ ing. That decision was not an abuse of discretion. We have before recognized that evidence of departmental policies can be relevant to show intent in § 242 cases. United States v. Aldo Brown, 871 F.3d 532, 538 (7th Cir. 2017); United States v. David Brown, 250 F.3d 580, 586 (7th Cir. 2001). Other circuit courts have as well. United States v. Christopher A. Brown, 654 F. App’x 896, 910 (10th Cir. 2016); United States v. Rodella, 804 F.3d 1317, 1338 (10th Cir. 2015); United States v. Dise, 763 F.2d 586, 588 (3d Cir. 1985). Those decisions, ex‐ pressly or impliedly, acknowledge that an officer’s training can help inform his state of mind in certain circumstances. If, for example, an officer has been trained that officers should do certain things when confronted with tense situations, and he does those things, the fact that he acted in accordance with his training could make it less likely that he acted willfully. See Aldo Brown, 871 F.3d at 538 . And vice versa: If, as here, an officer has been trained that officers should not do several things when confronted with tense situations, yet he does those things anyway, the fact that he broke from his training could make it more likely that he acted willfully. The district court correctly accounted for both sides of the coin, admitting both Proano’s and the government’s proposed training‐re‐ lated evidence. Proano nevertheless argues that the government’s evi‐ dence of his training was inadmissible, relying mostly on Thompson v. City of Chicago, 472 F.3d 444 (7th Cir. 2006). Thompson concerned 42 U.S.C. § 1983 , and it held that the 12 No. 17‐3466 CPD’s General Orders (essentially, formal policy statements) were not relevant to proving whether force was constitu‐ tional. Thompson, 472 F.3d at 454 . This is because the Fourth Amendment, not departmental policy, sets the constitutional floor. Id. at 454 ; see also Whren v. United States, 517 U.S. 806 , 815–16 (1996); Scott v. Edinburg, 346 F.3d 752 , 760–61 (7th Cir. 2003). Since Thompson, however, we have clarified that there is no per se rule against the admission of police policies or training. Aldo Brown, 871 F.3d at 537–38; see also Florek v. Vil‐ lage of Mundelein, 649 F.3d 594 , 602–03 (7th Cir. 2011) (regard‐ ing expert testimony). We explained in Aldo Brown that such a rule would be especially excessive in the § 242 context, where an officer’s intent is at issue and the defendant has a constitutional right to present a defense. 871 F.3d at 538 . Thompson did not address whether evidence of police policy or training can be relevant to intent; § 1983, unlike § 242, is a civil statute that lacks a specific‐intent requirement. See Kings‐ ley v. Hendrickson, 135 S. Ct. 2466, 2472 (2015). Thompson there‐ fore offers no guide here. Still, Proano presses, even if some evidence of training may be relevant, the government’s evidence in this case was not because it concerned CPD‐specific training. Proano seizes on language from Aldo Brown, which said that evidence of “widely used standardized training or practice[s]” could be relevant to show an officer’s intent in § 242 cases. 871 F.3d at 538 . Proano characterizes the CPD’s training as “localized” and not “widely used,” and therefore not relevant. That char‐ acterization is suspect; the CPD is the second‐largest police force in the country. David B. Goode, Law Enforcement Policies and the Reasonable Use of Force, 54 WILLAMETTE L. REV. 371, 372 (2018). Regardless, neither Aldo Brown nor common sense lim‐ its the pool of admissible training‐related evidence of intent No. 17‐3466 13 to national, model, or interdepartmental standards. Assum‐ ing those standards exist,3 only evidence of training that the officer actually received can be relevant to his state of mind. Accord United States v. Trudeau, 812 F.3d 578 , 591–92 (7th Cir. 2016), cert. denied, 137 S. Ct. 566 (2016); United States v. Kokenis, 662 F.3d 919, 930 (7th Cir. 2011). Proano’s remaining arguments go to the weight of the ev‐ idence, not its relevance. He asserts that the prohibition on shooting into windows and crowds was not relevant because that training did not concern cars. But as the district court rea‐ sonably concluded, four or five people in the back of a car could constitute a crowd. Proano also asserts that his firearms training was not relevant because that training occurred in a controlled environment. Yet Jamison testified that the fire‐ arms training was not training for training’s sake, but rather it was intended to have real‐word application. Proano’s argu‐ ments were ones for the jury, not us. See United States v. Firi‐ shchak, 468 F.3d 1015, 1021 (7th Cir. 2006); Williams v. Jader Fuel Co., 944 F.2d 1388, 1403 (7th Cir. 1991). The probative value of an officer’s training, like most any evidence, depends on case‐specific factors. Those factors are too many to list, but no doubt included are the training’s re‐ cency and nature, representativeness of reasonable practices, standardization, and applicability to the circumstances the of‐ ficer faced. Whatever its ultimate strength, evidence of an of‐ ficer’s training can be relevant in assessing his state of mind. The district court carefully assessed the evidence and the — 3 — Proano emphasizes that only national standards can be relevant. The United States does not have a national police force, and Proano has not pointed us to what more widely used policies could have been relevant here. 14 No. 17‐3466 state‐of‐mind inquiry in this case, and it did not abuse its dis‐ cretion in admitting the evidence of Proano’s training. 2. Prejudice and Confusion Under Rule 403 A court may exclude relevant evidence if its probative value is substantially outweighed by risks of unfair prejudice or confusion. Fed. R. Evid. 403. Proano argues the evidence of his training presented those risks, and that the district court abused its discretion in not recognizing as much. Rule 403 speaks of what a district court “may” do, so we review a Rule 403 decision for abuse of discretion. More than that, a Rule 403 decision “is entitled to special deference” because only “in an extreme case are appellate judges competent to second‐ guess the judgment of the person on the spot, the trial judge.” United States v. Jackson, 898 F.3d 760, 764 (7th Cir. 2018). No risk of unfair prejudice or confusion substantially out‐ weighed the probative value of Proano’s training. Proano con‐ tends that the jury could have thought the training evidence mattered to whether his use of force was objectively reasona‐ ble, a question for which it is generally inadmissible. See Aldo Brown, 871 F.3d at 536–37; Thompson, 472 F.3d at 454 . But that risk was minimal. At the close of evidence, the district court instructed the jury: You have heard evidence about training the defend‐ ant received relating to the use of deadly force. You should not consider this training when you decide whether the defendant’s use of force was reasonable or unreasonable. But you may consider the training No. 17‐3466 15 when you decide what the defendant intended at the time he acted.4 See Rodella, 804 F.3d at 1338 (approving a similar instruction); see also United States v. Schmitt, 770 F.3d 524, 535 (7th Cir. 2014) (proper jury instructions can cure potential prejudice); United States v. Albiola, 624 F.3d 431, 440 (7th Cir. 2010) (same); Fed. R. Evid. 403 advisory committee’s notes (1972) (explaining that limiting instructions are a factor in weighing the danger of unfair prejudice). Proano also contends that the training evidence was un‐ fairly prejudicial because it invited the jury to convict for Proano’s failure to follow protocol. Proano was free to, and did, argue to the jury that his training had little applicability to the situation he faced, and the jury received the appropriate instructions about what was required to convict and how to use the evidence of his training. Proano identifies no grounds to assume the jury believed that violating CPD policy amounted to violating § 242. 3. Foundation Under Rule 602 Proano alternatively contends that, even if the training ev‐ idence was admissible, it was not admissible through Snelling and Jamison because their testimony lacked the proper foun‐ dation. Specifically, Proano argues, they both lacked “per‐ sonal knowledge of the training Proano received in 2006,” when he was in the academy. Rule 602 allows a witness to testify “to a matter only if … the witness has personal knowledge of the matter.” Fed. R. Evid. 602. “Evidence to 4 For good measure, district courts should provide this or a similar instruc‐ tion verbally before the parties offer a department’s policy or an officer’s training into evidence, as well as at the close of evidence. 16 No. 17‐3466 prove personal knowledge may consist of the witness’s own testimony.” Id. Snelling and Jamison could not recall whether they in‐ structed Proano at the academy, but that did not make their testimony inadmissible under Rule 602. Snelling taught use‐ of‐force procedures in (and before) 2006. By virtue of that po‐ sition, Snelling was aware of the use‐of‐force training that re‐ cruits generally received while Proano was enrolled. He testi‐ fied that he was aware of what his colleagues taught at the time and the academy’s common curriculum because of his “cross‐training,” a practice that ensures consistency among instructors. Jamison, too, taught while Proano attended the academy. He testified about the academy’s weapons training based on his familiarity with the academy’s preapproved les‐ son plans and syllabi, from which all firearms instructors teach. Jamison further described the firearms principles to which he testified as “basic,” “typical,” and “standard” at the academy. Snelling and Jamison thus had personal knowledge of the matters to which they testified, regarding the academy’s stock training in 2006. It was for the jury to determine whether Proano in fact received that training. Even if it were otherwise and, as Proano submits, Snelling and Jamison purported to testify regarding the training Proano actually received, the testimony was still admissible. Personal knowledge can in‐ clude reasonable inferences drawn from a witness’s observa‐ tions and firsthand experiences. Widmar v. Sun Chem. Corp., 772 F.3d 457, 460 (7th Cir. 2014); see also Visser v. Packer Eng’g Assocs., Inc., 924 F.2d 655, 659 (7th Cir. 1991); United States v. Giovannetti, 919 F.2d 1223, 1226 (7th Cir. 1990). Snelling and Jamison each offered enough evidence of the homogeneity in No. 17‐3466 17 the academy’s teachings (facts with which they had firsthand experience) to establish their respective personal knowledge of what Proano learned there (a reasonable inference). In ei‐ ther event, the district court did not abuse its discretion in ad‐ mitting Snelling’s and Jamison’s testimony. C. Willfulness Instruction Proano’s next challenge is to the district court’s jury in‐ struction on willfulness, the mens rea requirement of § 242. We review de novo whether an instruction fairly states the law, and we review the decision to give a particular instruc‐ tion for an abuse of discretion. United States v. Maldonado, 893 F.3d 480, 486 (7th Cir. 2018).5 Section 242 is a specific‐intent crime. Aldo Brown, 871 F.3d at 538 ; David Brown 250 F.3d at 584–85. It prohibits the willful deprivation of constitutional rights. United States v. Lanier, 520 U.S. 259, 264 (1997). The Supreme Court first addressed this crime (though at a time when it sat in a different part of the U.S. Code) in Screws v. United States, 325 U.S. 91 (1945) (plu‐ rality). Screws explained that a defendant need not “have been thinking in constitutional terms” to have willfully deprived another of a constitutional right. Screws, 325 U.S. at 106 . An officer does, though, have the requisite intent under § 242 when he “is aware that what he does is precisely that which the statute forbids.” Id. at 104; see also id. at 103–105 (an officer violates § 242 when he acts with “a specific intent to deprive — 5 — Proano submits that plain‐error review applies in assessing the district court’s willfulness instruction. See Fed. R. Crim. P. 52(a). That is a puzzling concession. Our review of the record suggests that he contested the in‐ struction adequately to preserve the issue for appeal, and the government does not contend otherwise. We, therefore, will not apply plain‐error re‐ view. 18 No. 17‐3466 a person” of constitutional rights or with “open defiance or in reckless disregard of a constitutional requirement”); see also Aldo Brown, 871 F.3d at 538 . In United States v. Bradley, 196 F.3d 762, 770 (7th Cir. 1999), we added that “to act ‘willfully’ in the § 242 sense, the defendant must intend to commit an act that results” in a constitutional deprivation. The district court in this case instructed the jury that Proano acted willfully if he “intended to deprive” Bates or Hemmans of their right to be free from unreasonable force. It explained further: The defendant acted intentionally if he used force knowing that the force he used was more than what a reasonable officer would have used under the cir‐ cumstances. The defendant did not act intentionally if he did not know that the force he used was more than what a reasonable officer would have used un‐ der the circumstances. Proano argues that these instructions reduced the needed mens rea and transformed § 242 into a general‐intent crime. We disagree. The district court’s instruction was consistent with Bradley. The court instructed the jury that it could convict only if Proano acted intending to violate constitutional rights. See Bradley, 196 F.3d at 769 (approving § 242 instruction that re‐ quired a finding of an act “with the intent” to deprive consti‐ tutional rights). The instruction then went a step beyond Brad‐ ley by defining what intent meant under § 242. It explained that Proano had the requisite intent if and only if Proano knew his force was not reasonable and used it anyway. There is no one definition of willfulness. Ratzlaf v. United States, 510 U.S. 135, 141 (1994); see also United States v. No. 17‐3466 19 Pulungan, 569 F.3d 326, 329 (7th Cir. 2009) (“‘Willfully’ is a no‐ toriously plastic word.”); Pattern Criminal Jury Instructions of the Seventh Circuit 4.11 (2012 ed.) (willfulness is a statute‐ specific term). Yet the district court’s definition tracked how most authorities understand the term. See, e.g., United States v. Dobek, 789 F.3d 698, 700 (7th Cir. 2015) (willfulness in criminal law often requires knowledge that one is violating the law) (citations omitted). The Model Penal Code, as a prime exam‐ ple, states that willfulness is met “if a person acts knowingly with respect to the material elements of the offense.” Model Penal Code § 2.02(8); see also United States v. Ladish Malting Co., 135 F.3d 484, 487 (7th Cir. 1998). More important than the gen‐ eral understanding of willfulness, though, the instruction— requiring a finding that Proano acted “knowing” his actions were an unreasonable use of force—is consistent with Screws, which required that a defendant be “aware” he is doing what the statute forbids in the § 242 context. Screws, 325 U.S. at 104 ; see also Aldo Brown, 871 F.3d at 538 .6 The instruction thus did not, as Proano insists, permit a conviction based only on the — 6 — Other circuit courts have described willfulness in the § 242 context some‐ what differently. See United States v. Cowden, 882 F.3d 464, 474 (4th Cir. 2018) (defining willfulness as “the particular purpose of violating a pro‐ tected right … or recklessly disregard[ing] the risk that he would do so”) (citations and alterations omitted); United States v. House, 684 F.3d 1173 , 1199–1200 (11th Cir. 2012) (defining willfulness similarly); United States v. McRae, 795 F.3d 471, 479 (5th Cir. 2015) (defining willfulness as conduct done “voluntarily and intentionally and with the specific intent to do something the law forbids”); United States v. Reese, 2 F.3d 870, 885 (9th Cir. 1993) (“the requisite specific intent is the intent to use more force than is necessary under the circumstances”). Those definitions are not so dissim‐ ilar from the district court’s definition to cast doubt on our conclusion that the district court fairly stated the law. 20 No. 17‐3466 unreasonable use of force. Proano’s contrary reading takes “knowing” out of the instruction. Proano also insists that the willfulness instruction is mis‐ leading when “juxtaposed” with the instruction on how the jury could use Proano’s training‐related evidence (i.e., for in‐ tent but not objective‐reasonableness purposes). These in‐ structions, he claims, “blur the distinct” objective and subjec‐ tive parts of § 242 and are at odds with one another. We again fail to see how. Both instructions were clear, concise, and guided the jury in determining Proano’s subjective intent in shooting at the Toyota. To that end, the instructions were complimentary: one defined intent; the other told the jury what evidence it could use in assessing intent. The district court’s instructions provide no grounds to reverse. D. Sufficiency of the Evidence Proano’s final contention is that trial failed to produce suf‐ ficient evidence to convict him. We can overturn the jury’s verdict only if in viewing the record in the government’s favor it is “devoid of evidence from which a reasonable jury could find guilt beyond a reasonable doubt.” United States v. Wrobel, 841 F.3d 450, 454 (7th Cir. 2016). A defendant bears the burden of convincing the court that “no rational trier of fact could have found him guilty.” United States v. Warren, 593 F.3d 540, 546 (7th Cir. 2010). We have often said that this is a heavy bur‐ den—indeed a “nearly insurmountable” one. E.g., Maldonado, 893 F.3d at 484 . Proano has not met that burden. He first asserts that there was insufficient evidence to prove that his actions were not objectively reasonable. Reasonableness depends on the total‐ ity of the circumstances. Plumhoff v. Rickard, 134 S. Ct. 2012 , No. 17‐3466 21 2020 (2014). This calls for a balanced inquiry into “the nature and quality of the intrusion on the individual’s Fourth Amendment interests against the countervailing government interests at stake.” Graham v. Connor, 490 U.S. 386, 396 (1989). We must view the events through the lens of the officer in the moment, not with 20/20 hindsight. Flournoy v. City of Chicago, 829 F.3d 869, 874 (7th Cir. 2016) (citing Graham, 490 U.S. at 396 ). The law, of course, allows for “the fact that police offic‐ ers are often forced to make split‐second judgments.” Graham, 490 U.S. at 397 . Deadly force is generally reasonable when a reasonable officer in the same circumstances would believe that the assailant’s conduct put someone in the “immediate vicinity in imminent danger of death or serious bodily in‐ jury.” Horton v. Pobjecky, 883 F.3d 941, 949 (7th Cir. 2018) (ci‐ tation omitted). Based on the totality of the circumstances, Proano argues, it was reasonable for him to believe that Brown and the other passengers were in mortal danger and to act accordingly. He emphasizes the chaos of the scene: Brown hanging out of the Toyota, Grant stuck between Flaherty’s squad car and the Toyota, the occupants’ refusal to show their hands, and the car reversing. The dashcam video, however, provided ample grounds for the jury to conclude that there was no danger posed to anyone and, thus, no need for lethal force. The video showed Brown sitting up out of a window, not being dragged. The car reversed at a mild pace, and quickly slowed, redi‐ rected, and butted against a light pole. No bystander was near it. Yet Proano shot and continued to shoot even after the Toyota stopped its retreat. Nor does the BB gun’s presence at the scene show that the jury erred in its conclusions. The officers’ first awareness of 22 No. 17‐3466 the gun was when it fell to the ground, and there was no evi‐ dence that any passenger threatened an officer with a weapon. Cf. Garner, 471 U.S. at 11 (if a suspect threatens an officer with a weapon, deadly force may be reasonable). Kal‐ icki’s testimony, moreover, suggested that the BB gun fell out at the same time the shooting started. That fact, combined with Proano’s immediate show of force and Proano’s post‐ shooting reports, which did not identify the BB gun, could have reasonably led the jury to reject the idea that Proano fired in reaction to the weapon. Proano relatedly posits that the government’s reliance on the dashcam video, particularly its slow‐motion version, is (1) inconsistent with the totality analysis required under the Fourth Amendment and (2) distorts the in‐the‐moment expe‐ rience Proano felt. The jury heard, and clearly rejected, these arguments. The jury saw at trial and had in deliberations the real‐time dashcam video. It may well have reviewed that video and found that no interpretation of the circumstances supported the notion that someone was in danger. Even if circumstances were sufficient to give rise to a lethal threat reasonably requiring deadly force, a jury still could have decided that Proano’s reaction was unreasonable. Proano argues that the number of rounds he fired—sixteen— is irrelevant, because officers reasonably shoot until the threat is eliminated. That is correct in principle, see Plumhoff, 134 S. Ct. at 2022 , but wrong in application. The jury could have con‐ cluded (easily) that Proano continued to apply lethal force even after the threat subsided. After the vehicle stopped re‐ versing and began inching toward the light pole, Proano con‐ tinued to fire several more shots into its side. See, e.g., Becker v. Elfreich, 821 F.3d 920, 928 (7th Cir. 2016) (it is “well‐ No. 17‐3466 23 established that police officers cannot continue to use force once a suspect is subdued”). Proano next asserts that there was insufficient evidence to prove that he willfully used unreasonable force. Again, how‐ ever, the dashcam video provided grounds for the jury to con‐ clude otherwise. The brazenness of Proano’s actions alone could have supported the jury’s conclusion: despite the car not threatening anyone’s safety, Proano fired sixteen shots at it, including several after the car began idling. See Bradley, 196 F.3d at 769 (sufficient evidence of willfulness when an officer fired at a car “to stop … [its] flight” which, in the circum‐ stances, was “clearly unreasonable and excessive”). Add to that how Proano, viewing the record in the government’s fa‐ vor, disregarded training by: using his gun, cocked, as an im‐ mediate show of force; discharging it into a group of people; shooting at something into which he did not have visibility; and never reassessing the situation until his magazine was empty. The jury also could have disregarded Proano’s justifi‐ cations as inconsistent with the video evidence. Specifically, although Proano reported concern for Brown, who he said was being “dragged” by the Toyota, the jury could have con‐ cluded that assertion was flatly not believable in light of the video, which showed Brown propped up out of the window (and thus not “dragged”). In all, there was sufficient evidence to convict Proano on both counts. III. Conclusion For these reasons, we AFFIRM the district court’s judg‐ ment.