RECOMMENDED FOR FULL-TEXT PUBLICATION Pursuant to Sixth Circuit I.O.P. 32.1(b) File Name: 19a0055p.06 UNITED STATES COURT OF APPEALS FOR THE SIXTH CIRCUIT RICKY JACKSON (17-3840); KWAME AJAMU, fka Ronnie Bridgeman, and WILEY EDWARD BRIDGEMAN (17-3843), Plaintiffs-Appellants, ┐ │ │ │ │ > v. CITY OF CLEVELAND; JEROLD ENGLEHART; KAREN LAMENDOLA, Guardian Ad Litem on behalf of Frank Stoiker; ESTATE OF EUGENE TERPAY, Administrator; ESTATE OF JAMES T. FARMER, Administrator; ESTATE OF JOHN STAIMPEL, Administrator, Defendants-Appellees. │ │ │ │ │ │ │ │ │ ┘ Nos. 17-3840/3843 Appeal from the United States District Court for the Northern District of Ohio at Cleveland. No. 1:15-cv-00989—Christopher A. Boyko, District Judge. Argued: June 14, 2018 Decided and Filed: March 28, 2019 Before: KEITH, ROGERS, and BUSH, Circuit Judges. _________________ COUNSEL ARGUED: Elizabeth C. Wang, LOEVY & LOEVY, Boulder, Colorado, for all Appellants. William M. Menzalora, CITY OF CLEVELAND, Cleveland, Ohio, for Appellee City of Cleveland. Stephen W. Funk, ROETZEL & ANDRESS, LPA, Akron, Ohio, for Appellees Karen Lamendola and the Estates of Eugene Terpay, James Farmer, and John Staimpel. ON BRIEF: Elizabeth C. Wang, LOEVY & LOEVY, Boulder, Colorado, for Appellant Ricky Jackson. Terry H. Gilbert, Jacqueline C. Greene, FRIEDMAN & GILBERT, Cleveland, Ohio, David E. Mills, THE MILLS LAW OFFICE LLC, Cleveland, Ohio, for Appellants Kwame Ajamu and Wiley Bridgeman. William M. Menzalora, CITY OF CLEVELAND, Cleveland, Ohio, for Appellee City of Cleveland. Stephen W. Funk, ROETZEL & ANDRESS, LPA, Nos. 17-3840/3843 Jackson, et al. v. City of Cleveland, et al. Page 2 Akron, Ohio, for Appellees Karen Lamendola and the Estates of Eugene Terpay, James Farmer, and John Staimpel. BUSH, J., delivered the opinion of the court in which ROGERS, J., joined, and KEITH, J., joined in all except Section II(C)(2). KEITH, J. (pp. 53–54), delivered a separate concurring opinion. _________________ OPINION _________________ JOHN K. BUSH, Circuit Judge. Appellants Ricky Jackson, Wiley Bridgeman, and Kwame Ajamu served a long time in prison for a crime they did not commit. For Jackson, it was thirty-nine years; for Bridgeman, thirty-seven years; for Ajamu, twenty-five years. They each spent close to two and a half of those years on death row. These men cannot get back any of the time they lost or erase the things they experienced. The best they can hope for is a remedy of damages under 42 U.S.C. § 1983 and Ohio law. This appeal concerns whether their complaints state sufficient facts for certain claims not to be dismissed and whether the men have presented enough evidence for other claims to overcome summary judgment. In 1975, Jackson, Ajamu, and Bridgeman were convicted of murder. Their convictions were based largely on the purportedly eyewitness testimony of Edward Vernon, who then was thirteen years old. In 2014, nearly forty years later, Vernon recanted, disclosing that police officers had coerced him into testifying falsely. Vernon’s recantation led to the overturning of appellants’ convictions. The exonerated men filed suit in the Northern District of Ohio, alleging § 1983 claims based on alleged violations of their constitutional rights by the officers and the City of Cleveland (“Cleveland”), along with state-law claims for indemnification against Cleveland. This appeal requires us to untangle a knot of legal issues surrounding the district court’s grant of appellees’ motions for judgment on the pleadings and for summary judgment and its denial of appellants’ motions to amend their complaints. We AFFIRM the district court’s grant of summary judgment as to the § 1983 claims based on conspiracy, but we REVERSE and REMAND the Nos. 17-3840/3843 Jackson, et al. v. City of Cleveland, et al. Page 3 district court’s (1) judgment on the pleadings as to the indemnification claims; (2) denial of appellants’ motions to amend their complaints to substitute the administrator of the estates of the deceased officers as a party in their place; (3) summary judgment as to § 1983 claims arising from violations of Brady v. Maryland, 373 U.S. 83 (1963), fabrication of evidence, and malicious prosecution; and (4) summary judgment as to claims against Cleveland based on Monell v. Department of Social Services, 436 U.S. 658 (1978). I. BACKGROUND A. Factual Background As befits this stage of the litigation, we recite the relevant facts in the light most favorable to the plaintiffs, who are appellants here. See Ciminillo v. Streicher, 434 F.3d 461, 464 (6th Cir. 2006). In 1973, the Cleveland Division of Police promulgated General Police Order 19-73 (“GPO 19-73”), entitled “PRETRIAL DISCOVERY RIGHTS OF DEFENSE ATTORNEYS AND COURTS IN CRIMINAL CASES.” R. 101-7, PageID 1630.1 GPO 19-73 provided that “defense counsel may be entitled” to several types of evidence, including “[e]vidence favorable to the defendant.” Id. But it also included a section entitled “EXCEPTION TO THE FOREGOING,” which contained the following provision: “The foregoing does not authorize the discovery or the inspection of . . . statements made by witnesses or prospective witnesses to state agents.” Id. The Manual of Rules used by the Division of Police (the “Manual”) did not otherwise instruct officers in handling potentially exculpatory information and did not mention Brady, as the Manual’s last update had occurred before Brady was decided. As described later in this opinion, some testimony suggests that Cleveland police officers may have received no formal training in their Brady obligations, and may not have known that Brady imposed any obligations upon them. 1All record citations are citations to the record in No. 17-3840 (Jackson’s suit) unless otherwise indicated. Citations to the record in Ajamu and Bridgeman’s suit will be prefaced by “No. 17-3843.” Nos. 17-3840/3843 Jackson, et al. v. City of Cleveland, et al. Page 4 Deposition testimony also reveals that, regardless of how officers understood their obligations under Brady, violations of those obligations were common. Although it was generally understood that anything in a detective’s file that was pertinent to a case “should go to the prosecutor,” it was up to individual officers whether they followed this policy, and they did not always do so. R. 103, PageID 3794. The general practice at the time, followed in “every case,” was for detectives to provide prosecutors with only “arrest reports, witness forms and written statements taken by the Statement Unit,” and “photos,” while omitting to turn over other evidence, including potentially exculpatory evidence, unless it was specifically requested by the prosecutor. Id. at PageID 3672–75. Deposition testimony describes this as a “practice,” which “happened more than it should,” of “detectives not [turning] over all the evidence to prosecutors.” R. 104, PageID 3970. Some detectives took a more proactive role by “manipulating the evidence” before giving it to prosecutors. Id. at PageID 3967. This was done, one officer testified, “because winning the case was what it was all about. It wasn’t about what was fair, it wasn’t about what was honest, it was about winning.” Id. at PageID 3967–68. Against this backdrop of evidence of incomplete Brady knowledge and frequent Brady violations, the record tells the following story. On May 19, 1975, Edward Vernon, then twelve years old, was riding the bus home from school when he heard two gunshots. Being twelve, Vernon exited the bus at the earliest opportunity and ran to where he believed the shots originated. Coming upon the scene, Vernon found a gunshot victim, but nothing to indicate who was responsible for the shooting. After police had secured the area, Vernon left and met up with a friend who told Vernon that the perpetrators were Ricky Jackson, Kwame Ajamu (then known as Ronnie Bridgeman), and Wiley Bridgeman (collectively, “Plaintiffs”). Vernon, a civic-minded youth, returned to the crime scene and told an officer that he knew who had committed the shooting, whereupon the officer recorded Vernon’s contact information. The next day, Detectives Eugene Terpay and James Farmer went to Vernon’s house and requested that he go down to the station to give a statement. As Vernon later recounted, when Nos. 17-3840/3843 Jackson, et al. v. City of Cleveland, et al. Page 5 his mother asked to accompany him to the station, the officers “told her, no, he’ll be all right, he’ll be all right.” R. 99-1, PageID 1183. At the station, Vernon told Terpay and Farmer that Plaintiffs had committed the shooting and gave their descriptions, which he was able to do because he knew them from the neighborhood. The following day, Terpay and Farmer again went to Vernon’s house and asked him for Plaintiffs’ addresses. Detective John Staimpel, along with his partner Frank Stoiker, was working the case with Terpay and Farmer. On May 25, Staimpel and another detective, whose name Vernon cannot remember, picked Vernon up at his house to bring him to a line-up. Vernon’s mother again asked to accompany him, and Vernon recalls the detectives saying, “[N]o, he’ll be all right. We’ll bring him back after the lineup.” Id. at PageID 1189. The detectives brought Vernon to the line-up and, as he recollects, asked him if “I see anybody that I recognize up there,” which Vernon interpreted as asking whether he had seen anyone in the line-up commit the shooting. Id. at PageID 1190. Vernon replied that he had not. Ricky Jackson and Wiley Bridgeman, who had been arrested earlier in the day, were in that line-up. From Vernon’s point of view, he had been forthright up until this point: he had honestly told the detectives that (he thought) he knew who had committed the crime, but he had never said that he had actually witnessed the crime, and so when he was asked at the line-up whether he saw anyone whom he had seen commit the crime, he said no. The two detectives then brought Vernon into a room, whereupon Staimpel accused Vernon of lying, threatened to send his parents to jail for perjury, banged on a table, and used racial pejoratives to describe Vernon. (Vernon and Plaintiffs are African-Americans.) After Vernon began to cry, Staimpel said, “[W]e’ll fix it,” and the detectives left the room. Id. at PageID 1191. When the detectives returned, they gave Vernon a piece of paper, explained to him that it said he had failed to identify Jackson and Bridgeman in the line-up because he had been scared of their retaliating, and told Vernon to sign it, which Vernon did. Stoiker signed a police report dated May 25, 1975, which described Stoiker and Staimpel’s picking Vernon up and taking him to the line-up, Vernon’s failing to identify Jackson and Bridgeman, and Vernon’s explaining this failure as due to his being “very afraid” of Plaintiffs. R. 114-19, PageID 5113. Nos. 17-3840/3843 Jackson, et al. v. City of Cleveland, et al. Page 6 The day after the line-up, Terpay and Farmer again spoke with Vernon. The detectives brought Vernon to the station, where he told them that he had not witnessed the crime and that he had never said that he had witnessed the crime. Terpay was wroth, yelling at Vernon and accusing him of having lied when he had gone to the line-up and said “that this is not them.” R. 99-1, PageID 1194. Terpay threatened to send Vernon’s parents to jail for perjury, and Vernon agreed to testify that he had seen Plaintiffs commit the crime. A police report dated May 28, 1975 indicates that Stoiker and Staimpel “[c]onsulted with [the prosecutor] who issued papers charging [Plaintiffs] with [homicide].” R. 114-28, PageID 5321. Prior to Jackson’s trial, Terpay coached Vernon regarding his testimony and afterwards reviewed the trial transcript with Vernon to ensure that his testimony in the trials of Bridgeman and Ajamu was consistent. Plaintiffs were convicted at separate trials. They were sentenced to death, but their sentences were later reduced to life imprisonment. For nearly forty years, Vernon struggled with the knowledge that his testimony had put Plaintiffs in prison. He later testified, “Through out [sic] the years this case has . . . be[en] heavy on my emotion, my everything.” R. 99-2, PageID 1234. “I wanted to come forward throughout the years, but I was scared, scared to come forward and tell the truth . . . with this battle in my mind, battle in my spirit, battle in my heart . . . . I’m battling with this . . . pretty much all my life . . . .” R. 99-1, PageID 1203. The years did not lessen the turmoil in Vernon’s mind. One day in 2013, Vernon finally disburdened his conscience. He was lying in a hospital bed, stricken with hypertension and kidney failure, when his pastor visited and told him that Innocence Project attorneys were seeking to exonerate Plaintiffs. Vernon testified: So, after he stated that and I said, okay, well, you know what—I got up out of the bed and I just broke down and I cried on his shoulder and I said, well, I’m ready to tell the truth, I’m ready to come forward and tell the truth that Ricky Jackson did not commit the crime that he went to prison for. Id. at PageID 1203–04. Nos. 17-3840/3843 Jackson, et al. v. City of Cleveland, et al. Page 7 Vernon formally recanted his testimony against each of the three Plaintiffs in November 2014. After the recantation, the prosecutor for Cuyahoga County, where Cleveland is located, admitted that there was “no evidence tying any of the three convicted defendants to the crimes” and that “[t]hey have been victims of a terrible injustice.” R. 116, PageID 6302–03. B. Procedural History On May 19, 2015, Jackson filed suit against Terpay, Farmer, Stoiker, 2 Staimpel, and Cleveland (collectively, “Defendants”), as well as others,3 alleging a multitude of state and federal claims. Bridgeman and Ajamu filed suit against the same defendants on July 2, 2015. On October 1, 2015, Cleveland moved for judgment on the pleadings, under Federal Rule of Civil Procedure 12(c), as to the state-law claims in both complaints. The district court granted both motions. In November 2015, Jackson, and Bridgeman and Ajamu in their parallel lawsuit, all moved for leave to file amended complaints substituting the administrator of the estates of the deceased Defendants (the “administrator”) for those Defendants. (J. Reid Yoder is the administrator of all of the estates.) The district court denied those motions as futile, reasoning that a § 1983 claim brought in Ohio does not survive a defendant’s death. On January 27, 2017, Stoiker (the only living individual Defendant) moved for summary judgment in both lawsuits, arguing that he was not involved in any unconstitutional activity and that, even if he was, he is protected by qualified immunity. On the same date, Cleveland also moved for summary judgment as to the Monell claims, arguing that Plaintiffs had failed to provide evidence sufficient for a jury to find Cleveland liable. The district court granted both motions for summary judgment. 2Karen Lamendola is acting as guardian ad litem on behalf of Stoiker and is therefore the named defendantappellee representing Stoiker’s interests. We continue to refer to Stoiker as a Defendant for narrative convenience. 3Plaintiffs also named another former officer, Jerold Englehart, in their notices of appeal. However, in their appellate briefs, they expressly abandon their claims against Englehart. Jackson Br. at 33; Ajamu & Bridgeman Br. at 6. Nos. 17-3840/3843 Jackson, et al. v. City of Cleveland, et al. Page 8 II. DISCUSSION We review de novo a judgment on the pleadings under Federal Rule of Civil Procedure 12(c), applying the same standard we apply to review the grant of a motion to dismiss under Rule 12(b)(6). Warrior Sports, Inc. v. Nat’l Collegiate Athletic Ass’n, 623 F.3d 281, 284 (6th Cir. 2010). We therefore “construe the complaint in the light most favorable to the plaintiff and accept all allegations as true” to determine whether the “complaint . . . contain[s] sufficient factual matter, accepted as true, to state a claim to relief that is plausible on its face.” Doe v. Miami Univ., 882 F.3d 579, 588 (6th Cir. 2018) (citations and internal quotation marks omitted). Federal Rule of Civil Procedure 15(a)(2) provides that leave to amend “should [be] freely give[n] . . . when justice so requires.” “We review a district court’s order denying a Rule 15(a) motion to amend for an abuse of discretion.” Rose v. Hartford Underwriters Ins. Co., 203 F.3d 417, 420 (6th Cir. 2000) (citing Gen. Elec. Co. v. Sargent & Lundy, 916 F.2d 1119, 1130 (6th Cir. 1990)). “A district court abuses its discretion when it relies on clearly erroneous findings of fact, uses an erroneous legal standard, or improperly applies the law.” United States v. Arny, 831 F.3d 725, 730 (6th Cir. 2016) (citation and internal quotation marks omitted). “We review a district court’s grant of summary judgment de novo.” Adair v. Charter Cty. of Wayne, 452 F.3d 482, 486 (6th Cir. 2006) (quoting Allen v. Mich. Dep’t of Corr., 165 F.3d 405, 409 (6th Cir. 1999)). Summary judgment is appropriate when “no genuine dispute as to any material fact” exists and the moving party “is entitled to judgment as a matter of law.” Fed. R. Civ. P. 56(a). “A genuine dispute of material fact exists ‘if the evidence is such that a reasonable jury could return a verdict for the nonmoving party.’” Peffer v. Stephens, 880 F.3d 256, 262 (6th Cir. 2018) (quoting Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 248 (1986)). At the summary judgment stage, “the evidence is construed and all reasonable inferences are drawn in favor of the nonmoving party.” Burgess v. Fischer, 735 F.3d 462, 471 (6th Cir. 2013) (citing Hawkins v. Anheuser-Busch, Inc., 517 F.3d 321, 332 (6th Cir. 2008)). Numerous decisions by the district court are now on appeal: (1) the district court’s dismissal, with prejudice, of Plaintiffs’ claims against Cleveland for indemnification under state law; (2) the district court’s denial of Plaintiffs’ motions to substitute the administrator of the Nos. 17-3840/3843 Jackson, et al. v. City of Cleveland, et al. Page 9 deceased Defendants’ estates as a defendant; (3) the district court’s grant of summary judgment to Stoiker on the § 1983 claims; and (4) the district court’s grant of summary judgment to Cleveland on the Monell claims. We address each issue in turn. A. Indemnification Claims The claims against Cleveland under Ohio Revised Code § 2744.07(B)4 seek indemnification for damages based on the alleged torts of the individual Defendants, who are former employees of Cleveland. Section 2744.07(B) provides that “a political subdivision shall indemnify and hold harmless an employee” found liable for that employee’s acts, so long as the employee was “acting in good faith” and “within the scope of employment.”5 The district court granted Cleveland’s motions for judgment on the pleadings, reasoning that § 2744.07(B) provides only a tortfeasor employee, and not a tort victim, with the right to bring a claim of indemnification against the tortfeasor’s employer. Plaintiffs argue that the district court erred in dismissing their indemnification claims with prejudice because the claims were not yet ripe and unripe claims, if they are to be dismissed, should only be dismissed without prejudice. Generally, a claim may not be adjudicated on its merits unless it is ripe. See Nat’l Rifle Ass’n of Am. v. Magaw, 132 F.3d 272, 284 (6th Cir. 1997). A claim is unripe when it “is anchored in future events that may not occur as anticipated, or at all.” Id. (citing Pac. Gas & Elec. Co. v. State Energy Res. Conservation & Dev. Comm’n, 461 U.S. 190, 200–01 (1983); Dames & Moore v. Regan, 453 U.S. 654, 689 (1981)). This prohibition comes both from the case or controversy requirement of Article III and from prudential considerations. See Brown v. 4At the time of the district court’s opinion, the relevant language appeared in § 2744.07(A)(2), so the district court cited that provision. See Jackson v. City of Cleveland, No. 1:15CV989, 2016 WL 3952117, at *2 (N.D. Ohio July 20, 2016). Because the Ohio Revised Code has since been amended, we cite the subsection in which the relevant language now appears, subsection (B). The language has not been changed in any way that would affect the district court’s, or our, analysis. 5Cleveland’s brief does not address whether the defendant officers were acting “in good faith” and within the scope of their employment for purposes of the indemnification claims, and the district court did not consider those issues. We will not address unargued principles of Ohio law on which the district court did not rule. Nos. 17-3840/3843 Jackson, et al. v. City of Cleveland, et al. Page 10 Ferro Corp., 763 F.2d 798, 801 (6th Cir. 1985) (noting that a ripeness analysis includes a discretionary determination beyond the Article III standing considerations). The ripeness doctrine exists “to prevent the courts, through premature adjudication, from entangling themselves in abstract disagreements.” Thomas v. Union Carbide Agric. Prods. Co., 473 U.S. 568, 580 (1985) (quoting Abbott Labs. v. Gardner, 387 U.S. 136, 148 (1967)). Application of this doctrine “requires that the court exercise its discretion to determine if judicial resolution would be desirable under all of the circumstances.” Brown, 763 F.2d at 801. Of primary importance is “whether the issues tendered are appropriate for judicial resolution,” and, if so, the degree of “hardship to the parties if judicial relief is denied” before the claim is allowed to ripen further. Young v. Klutznick, 652 F.2d 617, 625 (6th Cir. 1981) (quoting Toilet Goods Ass’n v. Gardner, 387 U.S. 158, 162 (1967)). Indemnification claims are frequently brought while unripe, depending as they often do on the favorable adjudication of underlying tort claims. Because of this, as a general matter, a claim for indemnification for damages that may be awarded on an underlying tort claim should not be adjudicated on the merits until the underlying claim is adjudicated. See, e.g., Safety Nat’l Cas. Corp. v. Am. Special Risk Ins. Co., 99 F. App’x 41, 43 (6th Cir. 2004) (finding unripe a claim of indemnification for fraudulent conveyance because, among other reasons, the underlying claim for fraudulent conveyance had not yet been adjudicated); see also Armstrong v. Ala. Power Co., 667 F.2d 1385, 1388–89 (11th Cir. 1982) (affirming dismissal of indemnity suits as premature prior to entry of judgment in underlying lawsuit); A/S J. Ludwig Mowinckles Rederi v. Tidewater Constr. Co., 559 F.2d 928, 932 (4th Cir. 1977) (holding indemnification issue not ripe prior to adjudication of underlying claims). Because the ripeness doctrine is discretionary, courts sometimes apply an exception for indemnification claims that have no possibility of success, regardless of the merits of the underlying claims. See, e.g., Cincinnati Ins. Co. v. Grand Pointe, LLC, No. 1:05-CV-161, 2006 WL 1806014, at *9 (E.D. Tenn. June 29, 2006) (collecting cases in support of the proposition that “a court may grant summary judgment on the issue of indemnification if it can determine the allegations in the complaint could under no circumstances lead to a result which would trigger the duty to indemnify” (citations and internal quotation marks omitted)). Nos. 17-3840/3843 Jackson, et al. v. City of Cleveland, et al. Page 11 The Sixth Circuit has not analyzed the propriety of this exception, and we need not do so now because, even if it is permissible for district courts to adjudicate indemnification claims with no possibility of success prior to the adjudication of underlying tort claims, this is not such a case. If Plaintiffs’ indemnification claims have no possibility of success, that would be because Ohio law provides that only the tortfeasor employees, and not the parties injured by them, may bring claims under Ohio Revised Code § 2744.07(B). The district court did an admirable job analyzing Ohio court cases before holding that Ohio law does so provide. See Jackson v. City of Cleveland, No. 1:15CV989, 2016 WL 3952117, at *2 (N.D. Ohio July 20, 2016). But the only cases available to the district court were from the Ohio courts of appeal, as the Ohio Supreme Court had yet to opine on the issue. The judgments of Ohio appellate courts not being binding on the Ohio Supreme Court, there remains a possibility that Plaintiffs’ indemnification claims could succeed: Plaintiffs would need to win their underlying tort action and, while that action was pending, the Ohio Supreme Court would need to adopt their interpretation of Ohio Revised Code § 2744.07(B). Although the latter eventuality may seem remote, it is far from impossible and, as it happens, the Ohio Supreme Court has accepted an appeal addressing this very issue. See Ayers v. Cleveland, 106 N.E.3d 65 (Ohio 2018) (Table). Because it is not impossible for Plaintiffs to prevail on their indemnification claims, those claims are not ripe for adjudication. As discussed above, in evaluating whether a claim is ripe, courts should determine (1) whether a matter is “appropriate for judicial resolution” and (2) whether the parties would undergo hardship “if judicial relief is denied” on their claim before it ripens further. Young, 652 F.2d at 625. Neither factor supports finding the indemnification claims are ripe here. First, interpreting Ohio Revised Code § 2744.07(B) is best avoided unless necessary. Federal courts generally avoid interpreting unsettled state law because state “courts are in the better position to apply and interpret” their own jurisdiction’s law. Travelers Indem. Co. v. Bowling Green Prof’l Assocs., PLC, 495 F.3d 266, 272 (6th Cir. 2007). As the Supreme Court Nos. 17-3840/3843 Jackson, et al. v. City of Cleveland, et al. Page 12 said in Railroad Commission of Texas v. Pullman Co., 312 U.S. 496 (1941) and repeated in Louisiana Power & Light Co. v. City of Thibodaux, 360 U.S. 25 (1959): Had we or they (the lower court judges) no choice in the matter but to decide what is the law of the state, we should hesitate long before rejecting their forecast of [state] law. But no matter how seasoned the judgment of the district court may be, it cannot escape being a forecast rather than a determination. Thibodaux, 360 U.S. at 27 (quoting Pullman, 312 U.S. at 499). Where, as here, adjudicating an issue of state law is unnecessary because the litigation is in its early stages, and state law is unsettled, the inappropriateness of deciding the issue in federal court weighs in favor of finding the claim unripe for adjudication in federal court. Second, that no harm will befall Cleveland if “judicial relief is denied” for the time being also weighs in favor of finding the indemnification claims unripe. Young, 652 F.2d at 625. The district court’s grant of Cleveland’s motions for judgment on the pleadings as to Plaintiffs’ indemnification claims did not release Cleveland from the litigation, as Plaintiffs still have Monell claims outstanding against Cleveland. The only effect that denying Cleveland’s motions, or holding them in abeyance, would have on the litigation would be to delay adjudication of the indemnification claims until a later stage in the litigation. At that point, the district court may be able to avoid interpreting Ohio Revised Code § 2744.07(B), because the Ohio Supreme Court may already have done so. The district court should interpret Ohio law only if the Ohio Supreme Court has not done so by the time the underlying § 1983 claims have been properly adjudicated on remand, and if those claims are found to have merit. The ripeness doctrine therefore requires that the indemnification claims not be adjudicated on the merits at the pleading stage, given the unsettled condition of state law. Because “a dismissal with prejudice operates as a rejection of the plaintiff’s claims on the merits,” the district court erred in dismissing those claims with prejudice. Mich. Surgery Inv., LLC v. Arman, 627 F.3d 572, 575 (6th Cir. 2010) (quoting United States v. One Tract of Real Prop., 95 F.3d 422, 425–26 (6th Cir. 1996)). Nos. 17-3840/3843 B. Jackson, et al. v. City of Cleveland, et al. Page 13 Plaintiffs’ Motions to Substitute Plaintiffs sought leave to amend their complaints to substitute the administrator of the estates of Defendants Terpay, Staimpel, and Farmer as a party in place of those Defendants, as they are now deceased. District courts “should freely give leave” to amend a complaint pre-trial “when justice so requires.” Fed. R. Civ. P. 15(a)(2). One permissible reason to deny leave is the “futility of [the] amendment[s].” Foman v. Davis, 371 U.S. 178, 182 (1962). The district court denied leave to amend, reasoning that § 1983 claims brought in Ohio do not survive the deaths of the tortfeasors, and, therefore, the requested amendments would be futile.6 On appeal, Defendants argue that the district court was correct, but also suggest an alternative ground for affirming—that Plaintiffs did not timely present their claims to the estates of the deceased Defendants. We address the survivability and timeliness arguments in turn. 1. Survival of § 1983 Claims Defendants first argue that the denial of Plaintiffs’ motions to amend should be affirmed because § 1983 claims do not survive the death of the tortfeasor in Ohio. 42 U.S.C. § 1988(a) provides that in actions to protect civil rights, where “the laws of the United States . . . are deficient in the provisions necessary to furnish suitable remedies and punish offenses against law, the common law, as modified and changed by the constitution and statutes of the State wherein the court having jurisdiction of such civil or criminal cause is held,” shall be applied, “so far as the same is not inconsistent with the Constitution and laws of the United States.” The Supreme Court has interpreted this statutory language as requiring a three-step process for determining which jurisdiction’s procedural law, such as provisions concerning statutes of limitations or the abatement of claims, is used in § 1983 suits. See Robertson v. Wegmann, 436 U.S. 584, 588–89 (1978). First, a district court must determine whether there is an applicable federal law that covers the issue, and, if there is, apply it. See id. Second, if there 6The district court also denied leave to amend on futility grounds with regard to Plaintiffs’ state-law claims against the deceased Defendants, but Plaintiffs do not appeal that ruling. Nos. 17-3840/3843 Jackson, et al. v. City of Cleveland, et al. Page 14 is no relevant federal law, then the district court must determine what the appropriate rule is in the state where the district court sits. See id. at 588. Third, the district court must determine whether the law of that state is “inconsistent with the Constitution and laws of the United States;” if there is no inconsistency, the state law is used, but if inconsistency exists, a federal common-law rule is used. Id. at 588–89. Because “[n]o federal statute or rule says anything about the survivorship of § 1983 claims,” Crabbs v. Scott, 880 F.3d 292, 294 (6th Cir. 2018), we turn to the relevant Ohio law, which provides: In addition to the causes of action which survive at common law, causes of action for mesne profits, or injuries to the person or property, or for deceit or fraud, also shall survive; and such actions may be brought notwithstanding the death of the person entitled or liable thereto. Ohio Rev. Code Ann. § 2305.21 (West 2019). Plaintiffs argued before the district court that their claims fall within “injuries to the person,” while Defendants argued that “injuries to the person” encompasses only physical injuries, and not the violation of rights alleged in this case. The district court agreed with Defendants, citing a district court case holding that under Ohio law, § 1983 claims similar to those brought by Plaintiffs did not involve “injuries to the person.” Tinney v. Richland Cty., No. 1:14 CV 703, 2014 WL 6896256, at *2 (N.D. Ohio Dec. 8, 2014), aff’d, 678 F. App’x 362 (6th Cir. 2017). After the district court’s judgment, however, Tinney was superseded by a published opinion of this circuit holding precisely the opposite. Crabbs expressly rejected the holding in Tinney, deciding instead that § 1983 claims are subject to the procedural rules of the state where they are brought that relate to personal injury actions, regardless of the specific type of injury alleged in the suit. See 880 F.3d at 296. When hearing a direct appeal, this court evaluates the merits of the case based on current law, not the law existing when the district court entered its judgment. See Chaz Concrete Co., LLC v. Codell, 545 F.3d 407, 409 (6th Cir. 2008). After Crabbs, all claims brought under § 1983 are to be treated as actions sounding in personal injury tort. Because Ohio Revised Code § 2305.21 provides that actions for personal injury survive the deaths of the tortfeasors, and that Nos. 17-3840/3843 Jackson, et al. v. City of Cleveland, et al. Page 15 statute does not conflict with the laws of the United States, see Crabbs, 880 F.3d at 295, § 1983 actions brought in Ohio survive the deaths of the tortfeasors. Therefore, through no fault of its own, the district court was in error as to its grounds for finding that the proposed amendments, substituting the administrator of the estates of Terpay, Staimpel, and Farmer for those Defendants, would be futile. 2. Timeliness of Plaintiffs’ Claims Against the Estates Defendants argue that we should affirm the district court on alternative grounds— namely, that the claims against the estates were not timely brought. Proper adjudication of this issue requires analysis of both Ohio and federal law. Defendants argue that Ohio estate law regarding the timely filing of claims defines which entities have the capacity to be sued, while Plaintiffs argue that those provisions are merely statutes of limitations. See Ohio Rev. Code §§ 2117.06, 2117.37. The points of contention do not end there, however. If Plaintiffs are correct that Ohio estate law merely establishes statutes of limitations, the parties also dispute whether those statutes or the general Ohio statute of limitations applies to § 1983 suits. On the other hand, if Defendants are correct that Ohio estate law defines which entities have the capacity to be sued, the parties also disagree over whether federal courts hearing § 1983 actions are bound by that definition, as well as whether an exception to that definition, provided in Ohio Revised Code § 2117.06(G), applies to the facts of this case. The district court did not address these issues, instead relying on its holding that the § 1983 claims did not survive the deaths of the deceased Defendants. 7 “It is the general rule that a federal appellate court does not consider an issue not passed upon below.” Lindsay v. Yates, 498 F.3d 434, 441 (6th Cir. 2007) (quoting United States v. Henry, 429 F.3d 603, 618 (6th Cir. 2005)). This directive is not jurisdictional, however, and “a departure from this general rule may be warranted when ‘the issue is presented with sufficient clarity and completeness and its 7The district court did address timeliness with regard to Plaintiffs’ state-law claims, but Plaintiffs do not challenge that analysis on appeal. The timeliness analysis required for the § 1983 claims differs from that required for the state-law claims: the former involves a three-step analysis to determine the applicable law, as described in section II(B)(1), supra. See Robertson, 436 U.S. at 588–89. The district court did not conduct this analysis. Nos. 17-3840/3843 Jackson, et al. v. City of Cleveland, et al. Page 16 resolution will materially advance the progress of this already protracted litigation.’” Katt v. Dykhouse, 983 F.2d 690, 695 (6th Cir. 1992) (quoting Pinney Dock & Transp. Co. v. Penn Cent. Corp., 838 F.2d 1445, 1461 (6th Cir. 1988)). We will follow the general rule and decline to address these issues in the first instance. These are thorny issues of first impression in this circuit, and because the district court has not yet addressed them, we do not believe they are “presented with sufficient clarity and completeness” for our review. Id. 3. Conclusion The district court erred in finding that Plaintiffs’ proposed amendments would be futile on the ground that § 1983 claims brought in Ohio do not survive the deaths of the tortfeasors, and we decline to address whether Defendants have presented an alternative ground on which the district court’s denial of Plaintiffs’ motions to amend could be affirmed. Because applying the wrong legal standard constitutes reversible error on abuse of discretion review, United States v. Arny, 831 F.3d 725, 730 (6th Cir. 2016), the district court’s denial of the motions to file amended complaints is REVERSED and REMANDED for further proceedings. C. Stoiker’s Motion for Summary Judgment We next address the district court’s grant of summary judgment to Stoiker on the § 1983 claims that Stoiker violated Plaintiffs’ Fourteenth Amendment right to due process by withholding exculpatory evidence, fabricating evidence, and conspiring to do the same, and Plaintiffs’ Fourth Amendment right to be free of malicious prosecution. If a police officer violates the Constitution, “42 U.S.C. § 1983 provides a civil remedy for those” injured by the violation. Peffer v. Stephens, 880 F.3d 256, 263 (6th Cir. 2018). But officers sued under the aegis of § 1983 are protected from liability by the doctrine of qualified immunity “insofar as their conduct does not violate clearly established statutory or constitutional rights of which a reasonable person would have known.” Pearson v. Callahan, 555 U.S. 223, 231 (2009) (quoting Harlow v. Fitzgerald, 457 U.S. 800, 818 (1982)). Qualified immunity does not apply if (1) “on the plaintiff’s facts,” a constitutional violation occurred, and (2) the alleged Nos. 17-3840/3843 Jackson, et al. v. City of Cleveland, et al. Page 17 violation was of “clearly established constitutional rights of which a reasonable person would have known.” Hoover v. Radabaugh, 307 F.3d 460, 465 (6th Cir. 2002) (quoting Dickerson v. McClellan, 101 F.3d 1151, 1158 (6th Cir. 1996)). The district court found that there was insufficient evidence for a reasonable jury to find that Stoiker had committed any of the alleged constitutional violations. We address each of the appealed determinations in turn. 1. Constitutional Violations a. Withholding Exculpatory Evidence The Due Process Clause of the Fourteenth Amendment provides that no state may “deprive any person of life, liberty, or property, without due process of law; nor deny to any person within its jurisdiction the equal protection of the laws.” U.S. Const. amend. XIV, § 1. In Brady v. Maryland, 373 U.S. 83, 87 (1963), the Supreme Court held “that the suppression by the prosecution of evidence favorable to an accused upon request violates due process where the evidence is material either to guilt or to punishment, irrespective of the good faith or bad faith of the prosecution.” Prosecutors are not the only state actors bound by Brady, and “police can commit a constitutional deprivation analogous to that recognized in Brady by withholding or suppressing exculpatory material.” Moldowan v. City of Warren, 578 F.3d 351, 379 (6th Cir. 2009). Brady claims have three elements: “[1] the evidence at issue must be favorable to the accused, either because it is exculpatory, or because it is impeaching; [2] that evidence must have been suppressed by the State, either willfully or inadvertently; and [3] prejudice must have ensued.” Strickler v. Greene, 527 U.S. 263, 281–82 (1999). The district court granted summary judgment to Stoiker on the ground that Plaintiffs had failed to present evidence sufficient for a reasonable jury to find the second element—that Stoiker had suppressed evidence. It did so for two reasons. First, it held that there was insufficient evidence for a jury to find that Stoiker was involved with the unconstitutional Nos. 17-3840/3843 Jackson, et al. v. City of Cleveland, et al. Page 18 activity at all, noting that Vernon had never identified Stoiker as one of the officers involved. 8 Second, even if Stoiker were involved, the district court held, there was insufficient evidence that he was aware of any exculpatory evidence, and an officer unaware of exculpatory evidence cannot suppress that evidence. We disagree with the district court’s reasoning. Given the evidence in the record, although a jury might ultimately find that Stoiker did not suppress evidence, it would not be unreasonable in finding that he had. Consider the following evidence. Vernon testified that Staimpel and another officer led him into a room after he failed to identify Plaintiffs at a line-up and coerced him into signing a false statement about that line-up. Staimpel testified at trial that (1) Stoiker was his partner and (2) Stoiker was present for the line-up. Based on Staimpel’s testimony, a reasonable jury could infer that (3) Stoiker was present during the post-line-up interview of Vernon and (4) Stoiker was present when Vernon signed his false statement explaining his “fear” of Plaintiffs. In addition, the record contains a police report, signed by Stoiker and dated the day that Vernon testified he was coerced into signing a false statement by two detectives, detailing the version of the line-up and subsequent interview that Vernon alleges were fabricated. The district court is correct that the report does not say that Stoiker was involved in that line-up and interview, but a jury is “allowed to make reasonable inferences from facts proven in evidence having a reasonable tendency to sustain them,” Galloway v. United States, 319 U.S. 372, 396 (1943), and it is reasonable to infer that a detective who signs a report was involved in the events recounted in that report. Because of this proof, a reasonable jury could find that Stoiker was present when Vernon was coerced into signing the allegedly false statement, in which he claimed that he had failed to identify Jackson and Bridgeman in the line-up because he was afraid of them. And if Stoiker was present when Vernon was coerced into signing the allegedly false statement, he knew that 8“The only evidence that points to Stoiker’s involvement are the signatures on the statement and the report. However, even if those are Stoiker’s signatures, Plaintiff has not cited to any policy, practice, or procedure about the meaning or effect of signature [sic]. Therefore, the Court is left to speculate as to what the signature meant.” Jackson v. City of Cleveland, CASE NO. 1:15CV989, 2017 WL 3380456, at *3 (N.D. Ohio Aug. 4, 2017). Nos. 17-3840/3843 Jackson, et al. v. City of Cleveland, et al. Page 19 Vernon had not given fear of Plaintiffs as his true reason for not identifying them—in other words, that the statement was false. That knowledge was exculpatory evidence. See Giglio v. United States, 405 U.S. 150, 154 (1972) (“When the reliability of a given witness may well be determinative of guilt or innocence, nondisclosure of evidence affecting credibility falls within [Brady’s disclosure] rule.” (citation and internal quotation marks omitted)). If Stoiker was present, he also knew that detectives coerced Vernon’s statement, which was a related, but separate, piece of exculpatory evidence. See id. In addition, a reasonable jury could find that Stoiker was aware of a third piece of exculpatory evidence. While his possible awareness of this evidence is less clear-cut than of the above-mentioned pieces of exculpatory evidence, it would not be unreasonable for a jury to infer that Stoiker knew Vernon had said he had not seen Plaintiffs commit the crime with which they were charged. Whether or not Stoiker was told by other officers that Vernon had not seen Plaintiffs commit the murder, Vernon stated that he was asked at the line-up “if I could recognize anyone there who was at the shooting” and that he answered that question in the negative. 9 R. 99-3, PageID 1236. Stoiker and the district court interpret that question as asking whether Vernon recognized anyone in the line-up. Vernon interpreted it as asking whether he had seen anyone in the line-up commit the crime.10 If Vernon’s interpretation is correct, then the officers present—which a reasonable jury could find included Stoiker—knew that Vernon was claiming he had not seen Plaintiffs commit the crime when he answered “No” to their question. A reasonable jury could find that Vernon, the only witness to the events who has testified to the contents of that conversation, interpreted the question correctly. As Stoiker did not disclose any of this evidence to prosecutors, a reasonable jury could find that Stoiker suppressed exculpatory evidence in violation of Brady. 9As mentioned in section I(A), supra, Vernon alternatively recalled that he may have been asked if “I see anybody that I recognize up there.” R. 99-1, PageID 1190. 10Vernon later explained that he answered in the negative because “I didn’t seen happen [sic] on May 19, 1975.” R. 99-2, PageID 1233. Nos. 17-3840/3843 Jackson, et al. v. City of Cleveland, et al. Page 20 As to the third element of a Brady claim, a reasonable jury could find that Plaintiffs suffered prejudice as a result of the alleged suppression. To show prejudice, Plaintiffs must show that the allegedly suppressed evidence was “material;” in other words, “that there is a reasonable probability that the suppressed evidence would have produced a different verdict.” Strickler, 527 U.S. at 280, 281. Because Vernon’s coerced statement formed the core of the prosecution’s case, there is a reasonable likelihood that, had the juries in Plaintiffs’ trials known that that statement was fabricated and coerced, or that Vernon had orally stated that he had not seen the shooting, the juries would not have convicted Plaintiffs. Therefore, a reasonable jury could find all three elements of a Brady claim satisfied. b. Fabricating Evidence The Due Process Clause of the Fourteenth Amendment is also “violated when evidence is knowingly fabricated and a reasonable likelihood exists that the false evidence would have affected the decision of the jury.” Gregory v. City of Louisville, 444 F.3d 725, 737 (6th Cir. 2006) (citing Stemler v. City of Florence, 126 F.3d 856, 872 (6th Cir. 1997)). Stoiker presents two arguments for why summary judgment was appropriate on this issue. First, he contends there was insufficient evidence that he was involved in the fabrication of Vernon’s statement. Second, Stoiker argues there was insufficient evidence that the fabricated statement affected the decision of the jury. The district court agreed with both of Stoiker’s arguments. Turning to the first argument, a reasonable jury could find, as discussed in section II(C)(1)(a), supra, that Stoiker was in the room when Vernon was initially intimidated, left the room with Staimpel, and then returned to the room with Staimpel, at which point Staimpel coerced Vernon into signing the statement. This does not necessarily entail that Stoiker participated in the creation of the false statement, but a reasonable jury could infer that Stoiker either drafted, or assisted Staimpel in drafting, the false statement. If Stoiker was actively involved in the fabrication of the false statement, he knowingly fabricated evidence. Nos. 17-3840/3843 Jackson, et al. v. City of Cleveland, et al. Page 21 As for the second argument, there is, in fact, sufficient evidence for a reasonable jury to conclude that the false statement influenced the juries at Plaintiffs’ trials. True, as the district court noted, although the statement was introduced in evidence by the defense at Jackson’s trial, it was used only by defense counsel in an attempt to impeach Vernon’s testimony and “it is unclear whether the jury in the Jackson trial had the statement while they were deliberating.”11 Jackson v. City of Cleveland, CASE NO. 1:15CV989, 2017 WL 3380456, at *3 (N.D. Ohio Aug. 4, 2017). Also, the statement was not admitted in evidence at Bridgeman’s or Ajamu’s trial at all. And it is fair to conclude, as the district court reasoned, that “Vernon’s live testimony,” not the statement, “led to the conviction[s] in all three trials.” Id. But the relevant question is not whether the fabricated evidence was shown to the jury; it is whether the statement affected the decision of the jury. For example, a fabricated search warrant affidavit, used to obtain evidence later shown to the jury, can form the basis for a fabrication-of-evidence suit. See Webb v. United States, 789 F.3d 647, 670 (6th Cir. 2015). And fabricated evidence that “is used as [the] basis for a criminal charge” can form the basis for a § 1983 claim because, absent that evidence, there would have been no jury. Halsey v. Pfeiffer, 750 F.3d 273, 294 n.19 (3d Cir. 2014). A reasonable jury in the present case could find that the fabricated statement impacted the juries’ decisions in the criminal trials in at least two ways. First, the prosecutor testified that his understanding of Vernon’s statement was based on the copy in the police report and that, if he had known what had actually happened on the day of the line-up, he would have declined to prosecute: he does not, as he put it, “believe in prosecuting innocent people.” R. 114-29, PageID 5350. The prosecutor did not speak to Vernon prior to bringing charges, and so the false statement constituted the entire basis for his understanding of Vernon’s involvement. If Staimpel and Stoiker had not fabricated Vernon’s statement, therefore, charges would not have been brought, and, of course, a jury that is never empaneled is a jury that does not return a guilty verdict. 11Although the possibility is ultimately unnecessary to our holding on the fabrication-of-evidence claims, we note that because Vernon’s statement was introduced in evidence at Jackson’s trial, a reasonable jury could infer that the jury that convicted Jackson had access to the statement at some point in their deliberations. Nos. 17-3840/3843 Jackson, et al. v. City of Cleveland, et al. Page 22 A jury in the present case also could find that the falsified statement caused the criminal verdicts because the statement coerced Vernon to testify in conformance with it. Unlike Staimpel’s baseless threat to prosecute Vernon’s parents if Vernon failed to sign a statement saying that he had seen Plaintiffs commit the crime, Vernon would have faced a real threat of prosecution for perjury had his testimony conflicted with his earlier signed statement. See Osburn v. State, 7 Ohio 212, 214–15 (1835) (admitting as evidence of perjury a paper signed by the defendant). A reasonable jury could therefore find both that Stoiker participated in the fabrication of Vernon’s statement and that there is a reasonable probability the statement affected the juries at Plaintiffs’ trials. c. Conspiracy to Withhold and Fabricate Evidence To make out a claim for conspiracy to deprive them of their due process rights, Plaintiffs must show “that (1) a single plan existed, (2) the conspirators shared a conspiratorial objective to deprive the plaintiffs of their constitutional rights, and (3) an overt act was committed in furtherance of the conspiracy that caused the injury.” Robertson v. Lucas, 753 F.3d 606, 622 (6th Cir. 2014) (internal quotation marks omitted) (quoting Revis v. Meldrum, 489 F.3d 273, 290 (6th Cir. 2007)). For the reasons discussed above, a reasonable jury could find that Stoiker and Staimpel planned to draft a false statement and coerce Vernon into signing that statement and that they committed an overt act in furtherance of that plan. Further supporting the conspiracy claim, it would not be unreasonable for a jury to infer that the detectives planned to withhold the existence of their acts from prosecutors for the purpose of tipping the scales against Plaintiffs, as informing prosecutors of the coercion would have rendered their actions meaningless. However, the inquiry does not end there. We must also determine whether an individual can be held liable for conspiracy when the alleged conspiracy was undertaken by agreement with another individual or individuals employed by the same entity as the defendant. Nos. 17-3840/3843 Jackson, et al. v. City of Cleveland, et al. Page 23 The intracorporate conspiracy doctrine, which states that if “all of the defendants are members of the same collective entity, there are not two separate ‘people’ to form a conspiracy,” has been applied to 42 U.S.C. § 1985(3) by this court. Johnson v. Hills & Dales Gen. Hosp., 40 F.3d 837, 839–40 (6th Cir. 1994) (quoting Hull v. Cuyahoga Valley Joint Vocational Sch. Dist. Bd. of Educ., 926 F.2d 505, 510 (6th Cir. 1991)). Section 1985(3) creates a cause of action for a conspiracy between two or more persons to deprive another of the equal protection of the laws. We have also held that the doctrine applies in § 1985(2) suits. Doherty v. Am. Motors Corp., 728 F.2d 334, 339 (6th Cir. 1984). 42 U.S.C. § 1985(2) creates a cause of action for a conspiracy to, among other actions, obstruct justice or to intimidate a party, witness, or juror. But this circuit has never decided whether the intracorporate conspiracy doctrine also applies to suits under § 1983. See DiLuzio v. Vill. of Yorkville, 796 F.3d 604, 615 (6th Cir. 2015) (noting that “the Sixth Circuit has never held that the intracorporate conspiracy doctrine applies to municipal government officials in a § 1983 action and the district courts within our circuit are split on this question”). To determine whether there is a genuine dispute of material fact as to whether Stoiker conspired to violate Plaintiffs’ constitutional rights, therefore, we must resolve this issue of first impression. We are aware of only one circuit, the Eleventh, that has squarely addressed the issue and has determined that the intracorporate conspiracy doctrine applies in § 1983 actions as in § 1985 actions. See Grider v. City of Auburn, 618 F.3d 1240, 1261 (11th Cir. 2010); Rehberg v. Paulk, 611 F.3d 828, 854 (11th Cir. 2010). By contrast, we are aware of no circuit that has applied the doctrine in § 1985 actions but declined to apply it in § 1983 actions.12 We join the Eleventh Circuit and hold that the intracorporate conspiracy doctrine applies in § 1983 suits to bar conspiracy claims where two or more employees of the same entity are alleged to have been acting within the scope of their employment when they allegedly conspired together to deprive the plaintiff of his rights. See Grider, 618 F.3d at 1261–62; cf. Johnson, 12Some courts have held that the doctrine does not apply in the civil rights context at all. See, e.g., Brever v. Rockwell Int’l Corp., 40 F.3d 1119, 1126–27 (10th Cir. 1994). Nos. 17-3840/3843 Jackson, et al. v. City of Cleveland, et al. Page 24 40 F.3d at 841 (“[W]hen employees act outside the course of their employment, they and the corporation may form a conspiracy under 42 U.S.C. § 1985(3).”). We so hold because the considerations that support applying the intracorporate conspiracy doctrine in § 1985 suits pertain equally to the § 1983 context, and we discern no logical distinction upon which to treat § 1983 conspiracy claims differently. Cf. Hull, 926 F.2d at 509–10 (holding that the intracorporate conspiracy doctrine applies to § 1985(3) claims and stating “that this court’s opinion in Doherty [which applied the doctrine to § 1985(2)—not § 1985(3)—claims] is dispositive of this issue”). Recognizing that district courts within this circuit have split on the question,13 we will explain why the reasons for applying the doctrine to § 1983 outweigh the reasons for not doing so. The intracorporate conspiracy doctrine was recognized in antitrust and civil rights cases based on the legal notion of corporations as “persons.” See Copperweld Corp. v. Indep. Tube Corp., 467 U.S. 752, 769 n.15 (1984); Doherty, 728 F.2d at 339. When employees of a corporation act to further the purposes of that “person,” principles from the law of agency dictate that those employees be treated not as separate “persons” but as part of the same “person.” See Hull, 926 F.2d at 509–10; Doherty, 728 F.2d at 339. We have recognized the relevance of these principles to suits against employees of local government entities as well as against employees of private corporations. See Hull, 926 F.2d at 509–10. Furthermore, the Supreme Court has made clear that municipalities are “persons” for purposes of § 1983. Monell v. Dep’t of Soc. Servs., 436 U.S. 658, 690 (1978). Because the intracorporate conspiracy doctrine follows from the legal definition of “person,” which includes local governments, the doctrine has been developed to deal with the 13Compare Vaduva v. City of Xenia, No. 3:17-cv-41, 2017 WL 4773076, at *3 (S.D. Ohio Oct. 23, 2017) (applying the intracorporate conspiracy doctrine in § 1983 suit); Gillespie v. City of Battle Creek, 100 F. Supp. 3d 623, 631–32 (W.D. Mich. 2015) (same); Wright v. Bloomfield Twp., No. 12-15379, 2014 WL 5499278, at *15–16 (E.D. Mich. Oct. 30, 2014) (same); Pardi v. Cty. of Wayne, No. 12–12063, 2013 WL 1011280, at *14–15 (E.D. Mich. Mar. 14, 2013) (same); Audio Visual Equip. & Supplies, Inc. v. Cty. of Wayne, No. 06-10904, 2007 WL 4180974, at *5–6 (E.D. Mich. Nov. 27, 2007) (same); Adcock v. City of Memphis, No. 06-2109, 2007 WL 784344, at *4–5 (W.D. Tenn. Mar. 13, 2007) (same); Turner v. Viviano, No. 04-CV-70509-DT, 2005 WL 1678895, at *13 (E.D. Mich. July 15, 2005) (same), with Tinney v. Richland Cty., No. 1:14 CV 703, 2015 WL 542415, at *12 (N.D. Ohio Feb. 10, 2015) (declining to apply the doctrine in a § 1983 suit), aff’d on other grounds, 678 F. App’x 362 (6th Cir. 2017); Kinkus v. Vill. of Yorkville, 476 F. Supp. 2d 829, 839–40 (S.D. Ohio 2007) (same), rev’d on other grounds, 289 F. App’x 86 (6th Cir. 2008). Nos. 17-3840/3843 Jackson, et al. v. City of Cleveland, et al. Page 25 question whether there are two separate persons to form a conspiracy. See Hull, 926 F.2d at 510. The doctrine’s application to other civil rights statutes has not been premised upon any factor unique to those statutes. We therefore see no reason to decline to apply the doctrine to § 1983. Section 1983 creates a cause of action against any “person” who deprives a plaintiff of his rights, just like § 1985. Therefore, if § 1985 cannot be violated by an alleged conspiracy where the alleged conspirators are all employees of the same entity acting within the scope of their employment, neither can § 1983. Furthermore, we decline to adopt the rationale that because “[§] 1985 is in its essence a conspiracy statute[,] [while] [§] 1983 is not,” the intracorporate conspiracy doctrine applies to the former but not the latter. Kinkus v. Vill. of Yorkville, 476 F. Supp. 2d 829, 840 (S.D. Ohio 2007). Although § 1983 does not expressly contemplate a cause of action for conspiracy, once we have recognized such a cause of action—which we have, see, e.g., DiLuzio, 796 F.3d at 615– 16—the question whether a conspiracy can exist where all alleged conspirators work for the same entity, and are alleged to have been acting in the scope of their employment, naturally arises. That inquiry is identical under § 1983 and § 1985. After all, we did not apply the intracorporate conspiracy doctrine in § 1985 actions on a theory that the text of that particular statutory provision demanded it. Instead, we simply adopted a conspiracy jurisprudence that developed outside the civil rights context. See Hull, 926 F.2d at 509. Nor do we see any reason to limit application of the doctrine to cases in which a municipality is alleged to have conspired with one or more of its employees, in contrast to cases in which two or more employees are alleged only to have conspired with each other. We have made clear that “members of the same legal entity cannot conspire with one another as long as their alleged acts were within the scope of their employment.” Jackson v. City of Columbus, 194 F.3d 737, 753 (6th Cir. 1999) (emphasis added) (citing Johnson, 40 F.3d at 840), abrogated on other grounds by Swierkiewicz v. Sorema N. A., 534 U.S. 506 (2002); accord Hull, 926 F.2d at 510. In Hull, we applied the doctrine to bar the plaintiff’s § 1985(3) claim alleging conspiracy against “a school district superintendent, the executive director of the district, and a school administrator, all of whom [were] employees or agents of the Board [of Education].” 926 F.2d at 510. The plaintiff did not allege that the school board itself was a conspirator, but we noted that Nos. 17-3840/3843 Jackson, et al. v. City of Cleveland, et al. Page 26 “[s]ince all of the defendants [were] members of the same collective entity, there [were] not two separate ‘people’ to form a conspiracy.” Id. Finally, we have recognized an exception to the intracorporate conspiracy doctrine in § 1985(3) suits where the defendants were alleged to have been acting outside the scope of their employment, see Johnson, 40 F.3d at 841, and we have indicated that the exception would apply equally in the § 1983 context were we to apply the doctrine in § 1983 suits, see DiLuzio, 796 F.3d at 616. Accordingly, the intracorporate conspiracy doctrine applies to § 1983, and we assume that adopting the doctrine entails adopting the exception. Cf. DiLuzio, 796 F.3d at 616. But the scope-of-employment exception is unsupported by the record here because Plaintiffs have alleged that Stoiker and the other individual Defendants were acting “within the scope of their employment.” R. 86, PageID 1018; No. 17-3843, R. 53, PageID 707. Therefore, as a matter of law, Stoiker cannot be liable for conspiracy in violation of § 1983 where he is alleged to have conspired with other employees of the same government entity, in the scope of their employment, to violate Plaintiffs’ rights. The district court’s grant of summary judgment to Stoiker on the conspiracy claims is AFFIRMED. d. Malicious Prosecution The Fourth Amendment begins: “The right of the people to be secure in their persons, houses, papers, and effects, against unreasonable searches and seizures, shall not be violated.” U.S. Const. amend. IV. Amongst other protections, this guarantee affords people the right to be free of unjust prosecution. See Mills v. Barnard, 869 F.3d 473, 479–80 (6th Cir. 2017).14 A malicious-prosecution claim has four elements: “(1) that a criminal prosecution was initiated against the plaintiff and that the defendant ma[d]e, influence[d], or participate[d] in the decision to prosecute; (2) that there was a lack of probable cause for the criminal prosecution; 14Although we now analyze constitutional claims for malicious prosecution under the Fourth Amendment, “[p]rior to January 1994 . . . this circuit analyzed [such claims] as accruing under the Fourteenth rather than the Fourth Amendment.” Spurlock v. Satterfield, 167 F.3d 995, 1006 (6th Cir. 1999) (citations omitted). We ceased doing so after the Supreme Court held in Albright v. Oliver, 510 U.S. 266, 271, 273–75 (1994) that maliciousprosecution claims must be asserted under the Fourth Amendment rather than the Fourteenth. In so holding, the Albright Court recognized “the Fourth Amendment’s relevance to the deprivations of liberty that go hand in hand with criminal prosecutions.” Id. at 274 (citing Gerstein v. Pugh, 420 U.S. 103, 114 (1975)). Nos. 17-3840/3843 Jackson, et al. v. City of Cleveland, et al. Page 27 (3) that, as a consequence of a legal proceeding, the plaintiff suffered a deprivation of liberty . . . apart from the initial seizure; and (4) that the criminal proceeding must have been resolved in the plaintiff’s favor.”15 Id. at 480 (alterations in original) (internal quotation marks omitted) (quoting Sykes v. Anderson, 625 F.3d 294, 308–09 (6th Cir. 2010)). There being no dispute that Plaintiffs suffered a deprivation of liberty or that the criminal proceedings were resolved in their favor, we need only address the first two elements. i. Stoiker Influenced or Participated in the Decision to Prosecute The first element of the malicious-prosecution claim is met when an officer “could reasonably foresee that his misconduct would contribute to an independent decision that results in a deprivation of liberty” and the misconduct actually does so. Sykes, 625 F.3d at 316 (quoting Higazy v. Templeton, 505 F.3d 161, 177 (2d Cir. 2007)). This element is met when an officer includes “misstatements and falsehoods in his investigatory materials” and those materials influence a prosecutor’s decision to bring charges. Id. A reasonable jury could find that Stoiker’s misconduct influenced the decision to bring charges against Plaintiffs for two reasons. First, Stoiker and Staimpel “[c]onsulted with [the prosecutor] who issued papers charging [Plaintiffs] with [homicide].” R. 114-28, PageID 5321. Although the record does not indicate the contents of that consultation, it is reasonable to infer that it involved false statements about Vernon’s identification of Plaintiffs and that this consultation influenced the prosecutor’s decision to bring charges against Plaintiffs. Second, the prosecutor’s only knowledge of Vernon’s involvement when deciding to bring charges was based on Vernon’s statement, a statement that a jury could reasonably find to have been fabricated by Stoiker. And the prosecutor later testified that had he known about what 15There are two types of § 1983 claims, both sounding in the Fourth Amendment, that are sometimes referred to as “malicious prosecution” claims. One is for the wrongful institution of legal process (which is the type most properly called a “malicious prosecution” claim) and the other is for continued detention without probable cause. See Cleary v. Cty. of Macomb, 409 F. App’x 890, 898 (6th Cir. 2011); see also Gregory, 444 F.3d at 747–49 (stating that claims for continued detention without probable cause are not properly considered “malicious prosecution” claims, but recognizing that courts’ use of terminology varies). Although Plaintiffs are not always clear as to their intended theory of liability, they and Stoiker state the test for, and perform their analysis under, the wrongful institution of legal process theory, and we will do the same. Nos. 17-3840/3843 Jackson, et al. v. City of Cleveland, et al. Page 28 actually happened on the day of the line-up—because Stoiker and Staimpel had told him during conversation, or because they had drafted an accurate statement for Vernon, or because Stoiker had drafted an accurate report concerning that day’s events—the prosecutor would not have proceeded to trial. ii. There Was a Lack of Probable Cause for the Criminal Prosecution When a grand jury returns an indictment against a defendant, this creates a “presumption of probable cause,” which is rebuttable by showing that: (1) [A] law-enforcement officer, in the course of setting a prosecution in motion, either knowingly or recklessly ma[de] false statements (such as in affidavits or investigative reports) or falsifie[d] or fabricate[d] evidence; (2) the false statements and evidence, together with any concomitant misleading omissions, [we]re material to the ultimate prosecution of the plaintiff; and (3) the false statements, evidence, and omissions d[id] not consist solely of grand-jury testimony or preparation for that testimony (where preparation has a meaning broad enough to encompass conspiring to commit perjury before the grand jury). King v. Harwood, 852 F.3d 568, 587–88 (6th Cir. 2017). As discussed above, a reasonable jury could find that Stoiker falsified or fabricated evidence, and that the evidence did not consist solely of grand-jury testimony or preparation for that testimony. Stoiker argues, however, that even if he fabricated Vernon’s false statement, that statement could not have been material to the grand jury’s determination of probable cause, as it was not presented to the grand jury. Although Vernon may have testified to the grand jury in conformance with his fabricated statement, Stoiker argues, it was Vernon’s testimony, not the earlier statement, that impacted the grand jury’s decision. But a careful reading of King shows that fabricated evidence can be material to a grand jury’s determination of probable cause without being presented to the grand jury. If only evidence presented to a grand jury could be material to that grand jury’s decision, plaintiffs would be faced with the Scylla and Charybdis of either admitting that the fabricated evidence was not material or claiming that it was material because it was presented to the grand jury, thereby gracing the fabricator with the absolute immunity afforded to grand jury testimony. See id. at 589. Nos. 17-3840/3843 Jackson, et al. v. City of Cleveland, et al. Page 29 Instead, plaintiffs can show that a fabrication was material to the grand jury’s determination by showing “that the officer has made knowing or reckless false statements or has falsified or fabricated evidence in the course of setting a prosecution in motion.” Id. Here, according to the prosecutor, had Stoiker not fabricated Vernon’s statement, there would have been no grand jury. But even had there been one, Vernon would not have testified falsely before it. Stoiker’s fabrication was therefore material to the grand jury’s determination because it “was material to the ultimate prosecution” of Plaintiffs.16 Id. at 587–88. 2. Qualified Immunity The statute now codified at 42 U.S.C. § 1983 was originally passed in 1871. It was not until the second half of the twentieth century that the Supreme Court recognized that § 1983 admitted of an implicit doctrine, born of the common law, known as qualified immunity. See Pierson v. Ray, 386 U.S. 547, 555, 557 (1967). Since then this doctrine has grown considerably, but not without its critics. See, e.g., Kisela v. Hughes, 138 S. Ct. 1148, 1162 (2018) (Sotomayor, J., dissenting) (arguing that a “one-sided approach to qualified immunity transforms the doctrine into an absolute shield for law enforcement officers, gutting the deterrent effect of the Fourth Amendment”). Qualified immunity has outgrown its original justifications, which were “rooted in historical analogy” and “based on the existence of common-law rules in 1871.” Wyatt v. Cole, 504 U.S. 158, 170 (1992) (Kennedy, J., concurring). Responding to the many and varied suits brought under § 1983, the judiciary recrafted that limited version of the doctrine of qualified immunity in an effort to protect public officials “from undue interference with their duties and from potentially disabling threats of liability.” Elder v. Holloway, 510 U.S. 510, 514 (1994) (quoting Harlow v. Fitzgerald, 457 U.S. 800, 806 (1982)). We therefore no longer “attempt[] to locate [the qualified immunity] standard in the common law as it existed in 1871,” Ziglar v. Abbasi, 137 S. Ct. 1843, 1871 (2017) (Thomas, J., concurring), but instead attempt to determine whether a defendant, by his conduct, “violate[d] clearly established statutory or constitutional rights of which a reasonable person would have known,” Harlow, 457 U.S. at 818. 16Stoiker does not argue that there was sufficient evidence to find probable cause to prosecute Plaintiffs absent Vernon’s testimony. Nos. 17-3840/3843 Jackson, et al. v. City of Cleveland, et al. Page 30 At issue in this appeal is whether, in 1975, the constitutional rights allegedly violated by Stoiker were sufficiently clearly established to deprive him of the protection of qualified immunity. It is a plaintiff’s burden to show that the right at issue was clearly established. Harris v. Klare, 902 F.3d 630, 637 (6th Cir. 2018). Although the Supreme Court “do[es] not require a case directly on point for a right to be clearly established, existing precedent must have placed the statutory or constitutional question beyond debate.” White v. Pauly, 137 S. Ct. 548, 551 (2017) (alteration in original) (internal quotation marks omitted) (quoting Mullenix v. Luna, 136 S. Ct. 305, 308 (2015)). In examining “existing precedent,” “we may rely on decisions of the Supreme Court, decisions of this court and courts within this circuit, and in limited instances, on decisions of other circuits.” Spurlock v. Satterfield, 167 F.3d 995, 1006 (6th Cir. 1999) (citations omitted); accord Hearring v. Sliwowski, 712 F.3d 275, 280 (6th Cir. 2013). The Supreme Court has recognized “that officials can still be on notice that their conduct violates established law even in novel factual circumstances” and has “rejected a requirement that previous cases be ‘fundamentally similar’” to the facts in a case to render qualified immunity inapplicable. Hope v. Pelzer, 536 U.S. 730, 741 (2002) (quoting United States v. Lanier, 520 U.S. 259, 263 (1997)); see also id. at 753–54 (Thomas, J., dissenting) (“Certain actions so obviously run afoul of the law that an assertion of qualified immunity may be overcome even though court decisions have yet to address materially similar conduct.” (internal quotation marks omitted)). And we have noted that “‘[g]eneral statements of the law’ are capable of giving clear and fair warning to officers even where ‘the very action in question has [not] previously been held unlawful.’” Smith v. Cupp, 430 F.3d 766, 776–77 (6th Cir. 2005) (second alteration in original) (quoting Anderson v. Creighton, 483 U.S. 635, 640 (1987)). a. Withholding Exculpatory Evidence In 1975, it was clearly established law that prosecutorial withholding of exculpatory evidence violates a criminal defendant’s Fourteenth Amendment right to due process. See Brady v. Maryland, 373 U.S. 83, 86–87 (1963). Multiple circuits had also recognized by that time that “Brady-derived” claims could be based on the conduct of law-enforcement officers—as distinct from prosecutors—who had allegedly withheld exculpatory evidence. See Clarke v. Burke, 440 F.2d 853, 855 (7th Cir. 1971) (“This is not to say that there can never be a due process Nos. 17-3840/3843 Jackson, et al. v. City of Cleveland, et al. Page 31 violation if the prosecutor does not know that the police has in its possession evidence possibly favorable to the defendant . . . . It has been held . . . that knowledge of the police is knowledge of the prosecutor . . . .” (citations omitted)); Smith v. Florida, 410 F.2d 1349, 1351 (5th Cir. 1969) (“[I]t makes no difference if the withholding is by the prosecutor or by officials other than the prosecutor.” (citing Barbee v. Warden, Md. Penitentiary, 331 F.2d 842 (4th Cir. 1964))); Barbee, 331 F.2d at 846; cf. Jackson v. Wainwright, 390 F.2d 288, 295, 298 (5th Cir. 1968) (noting that “lower federal courts” applying Brady “ha[d] emphasized the harm to the defendant rather than the prosecutor’s motive in failing to disclose exculpatory evidence” and finding a violation of the “duty to disclose . . . exculpatory statements” where “there was no evidence of the prosecutor’s bad faith or of overreaching by the prosecution” (emphasis added)); Curran v. Delaware, 259 F.2d 707, 713 (3d Cir. 1958) (finding a violation of the Fourteenth Amendment when a police officer perjured himself at trial, regardless of whether the prosecutor was aware of the perjury). In Barbee, decided the year after Brady, the Fourth Circuit addressed the habeas claim of a man who argued his conviction violated due process because law-enforcement officers had not disclosed the existence of ballistics and fingerprint reports that “cast grave doubt upon” his guilt. Barbee, 331 F.2d at 844. Responding to the state’s argument that the man was required, and had failed, to show that the prosecutor knew about the reports, the court stated: Nor is the effect of the nondisclosure neutralized because the prosecuting attorney was not shown to have had knowledge of the exculpatory evidence. Failure of the police to reveal such material evidence in their possession is equally harmful to a defendant whether the information is purposely, or negligently, withheld. And it makes no difference if the withholding is by officials other than the prosecutor. The police are also part of the prosecution, and the taint on the trial is no less if they, rather than the State’s Attorney, were guilty of the nondisclosure. If the police allow the State’s Attorney to produce evidence pointing to guilt without informing him of other evidence in their possession which contradicts this inference, state officers are practicing deception not only on the State’s Attorney but on the court and the defendant . . . . If the police silence as to the existence of the reports resulted from negligence rather than guile, the deception is no less damaging. Nos. 17-3840/3843 Jackson, et al. v. City of Cleveland, et al. Page 32 The duty to disclose is that of the state, which ordinarily acts through the prosecuting attorney; but if he too is the victim of police suppression of the material information, the state’s failure is not on that account excused. Id. at 846 (footnotes omitted). The above cases, decided prior to Plaintiffs’ trials, make clear that the duty to disclose evidence falls on the state as a whole and not on one officer of the state particularly, and it was therefore clearly established by the time of those trials that Stoiker had a Fourteenth Amendment obligation to disclose exculpatory evidence. It was also clearly established that impeachment evidence, such as the fact that a witness was coerced into making a fabricated statement, qualifies as exculpatory. See Giglio v. United States, 405 U.S. 150, 153–55 (1972) (holding that evidence that the government had procured an informant’s testimony by suggesting he could escape prosecution through cooperating was “material” evidence “affecting credibility” that should have been disclosed to the defense under Brady). Stoiker argues that a Seventh Circuit case shows that it is not clearly established even now that officers are under a Brady obligation to disclose their own or fellow officers’ fabrication of evidence. In Saunders-El v. Rohde, 778 F.3d 556, 558 (7th Cir. 2015), the plaintiff sued police officers under Brady, alleging that the officers failed to disclose that they had severely beaten the plaintiff and planted his blood at a crime scene. The Seventh Circuit held that the plaintiff had not alleged a violation of Brady because “Brady does not require the creation of exculpatory evidence, nor does it compel police officers to accurately disclose the circumstances of their investigations to the prosecution.” Id. at 562. Even if we were bound by Saunders-El, which we are not, it would not foreclose our holding. Because Brady and its progeny are concerned only with ensuring that a defendant receives a fair trial, “Brady is concerned only with cases in which the government possesses information which the defendant does not.” United States v. Graham, 484 F.3d 413, 417 (6th Cir. 2007) (citation omitted). And so if a defendant knows at the time of trial that the Nos. 17-3840/3843 Jackson, et al. v. City of Cleveland, et al. Page 33 government has fabricated evidence, as in Saunders-El,17 officers do not violate Brady by failing to tell prosecutors that evidence has been fabricated.18 Had Plaintiffs argued that Stoiker violated Brady only by failing to disclose that Vernon’s statement was inaccurate, Stoiker’s reliance on Saunders-El might be appropriate, as Plaintiffs already knew that Vernon’s statement was inaccurate. But Plaintiffs did not know that Vernon’s statement had been coerced, and that fact could have been used to impeach Vernon’s testimony at trial. Therefore, Stoiker had an obligation to disclose that fact to Plaintiffs. Finally, as discussed in section II(C)(1)(a), supra, an additional piece of exculpatory evidence that Stoiker may have possessed was the knowledge of Vernon’s un-coerced statement to Terpay and Farmer that he had not seen the shooting at all. That statement was exculpatory evidence separate from the fact that Vernon’s signed statement was false, and there is no evidence that Plaintiffs knew of Vernon’s exculpatory statement. Therefore, even if Saunders-El were controlling, we would hold that Plaintiffs had alleged a violation of clearly established rights with regard to Stoiker’s alleged withholding of exculpatory evidence of which Plaintiffs were not aware. Stoiker is not entitled to qualified immunity on the withholding-of-evidence claims. b. Fabricating Evidence It is difficult to countenance any argument that a law-enforcement officer in 1975 would not be “on notice [his] conduct [was] unlawful” when coercing a witness into perjuring himself in a capital trial. Hope, 536 U.S. at 739 (citation omitted). The obvious injustice inherent in fabricating evidence to convict three innocent men of a capital offense put Stoiker on notice that his conduct was unlawful. Cf. id. at 745 (stating, in evaluating qualified immunity in the Eighth Amendment context, that “[t]he obvious cruelty inherent in [tying a prisoner to a hitching post “for an extended period of time in a position that was painful, and under circumstances that were 17In describing the circumstances of the alleged fabrication of crime-scene evidence underlying his Brady claim, the Saunders-El plaintiff indicated that he had known about the fabrication all along. See Saunders-El, 778 F.3d at 558. 18This course. does not necessarily entail that such a situation would involve no other constitutional violations, of Nos. 17-3840/3843 Jackson, et al. v. City of Cleveland, et al. Page 34 both degrading and dangerous”] should have provided respondents with some notice that their alleged conduct violated [the prisoner’s] constitutional protection against cruel and unusual punishment”). More concretely, as far back as 1935, the Supreme Court recognized that the introduction of fabricated evidence violates “the fundamental conceptions of justice which lie at the base of our civil and political institutions.” Mooney v. Holohan, 294 U.S. 103, 112 (1935) (citing Hebert v. Louisiana, 272 U.S. 312, 316 (1926)). And in 1942, the Supreme Court held that when a witness perjures himself because of threats from police officers, the defendant suffers “a deprivation of rights guaranteed by the Federal Constitution.” Pyle v. Kansas, 317 U.S. 213, 216 (1942). The only difference between those cases and the present one is that those cases involved the use of fabricated evidence at trial, whereas this one involves the use of fabricated evidence to affect a jury in a manner other than by introducing the evidence at trial. 19 However, we have recognized that a Fourth Amendment claim based on the fabrication of evidence does not require that “false testimony [have been] given at trial.” Spurlock, 167 F.3d at 1007. And we can see no principled distinction, for purposes of qualified immunity, between such a claim and Plaintiffs’ claims here that they were deprived of their Fourteenth Amendment due process rights through the use of fabricated evidence. The alleged misconduct here is the use of the falsified statement to procure testimony in conformance with it—the same type of misconduct that we have previously found supported recovery for a constitutional tort, irrespective of the stage at which the fabrication tainted the proceeding. See id. As far as clearly established law in 1975 is concerned, several months before the events at issue in this case, this court stated that Mooney “made it clear that the Fourteenth Amendment right to due process prohibits a knowing and deliberate use by a state of perjured evidence in order to obtain a conviction.” Burks v. Egeler, 512 F.2d 221, 224 (6th Cir. 1975). More recently, we have cited Brady, Pyle, and Mooney in finding that a defendant officer could not 19Although we assume for the purpose of this analysis that the allegedly fabricated evidence did not affect the proceedings through being used at trial, we again note that Vernon’s statement was used by defense counsel at Jackson’s trial, and a reasonable jury could find that the statement was considered by Jackson’s jury in some way. Nos. 17-3840/3843 Jackson, et al. v. City of Cleveland, et al. Page 35 “seriously contend that a reasonable police officer would not know that [his] actions [including fabricating evidence] were inappropriate and performed in violation of an individual’s constitutional . . . rights.” Spurlock, 167 F.3d at 1005–06 (also citing Albright v. Oliver, 510 U.S. 266, 274 (1994)). In Spurlock (a malicious-prosecution case), the defendant argued that Albright, and the Sixth Circuit case explicitly finding that malicious prosecution violated clearly established rights, had been decided after his conduct and therefore did not put him on notice. See Spurlock, 167 F.3d at 1006 n.19 (discussing generally Albright and Smith v. Williams, 78 F.3d 585, 1996 WL 99329 (6th Cir. 1996) (unpublished table opinion)). Rejecting that argument, we stated that “the fundamental principle that an individual has a constitutional right to be free from malicious prosecution . . . was clearly established well before either of [the] cases [cited by the defendant] was decided.” Id. The reasoning in Spurlock is sound, and we follow it in holding that Stoiker was on notice in 1975 that it was unlawful for him to fabricate evidence. Stoiker is not entitled to qualified immunity on the fabrication-of-evidence claims. c. Malicious Prosecution Stoiker argues that he is entitled to qualified immunity because Plaintiffs “fail to identify a pre-1975 case that would clearly establish that a police officer could be held liable for malicious prosecution where he did not actively participate in the prosecution [and] did not testify before the grand jury or at trial.” Stoiker Br. at 51. Stoiker’s argument admits of two interpretations, one of which is possibly valid but has false premises and the other of which has true premises but is invalid. Stoiker might be arguing that the state of malicious prosecution law in 1975 was in flux and that it was not clear at that time that he could be liable under a malicious prosecution cause of action. That may be true, but it does not follow that he is protected by qualified immunity. Whether a defendant is protected by qualified immunity turns not on whether the defendant was on notice that his actions satisfied the elements of a particular cause of action, but instead on whether the defendant was on notice that his actions violated the laws of the United States. Recently, when presented with a similar argument to Stoiker’s, we responded: Nos. 17-3840/3843 Jackson, et al. v. City of Cleveland, et al. Page 36 [The defendant] spends a considerable portion of his brief illustrating why it is not clear that he should be liable for malicious prosecution, thus reasoning that he is entitled to qualified immunity. Yet, his claim that the contours of our jurisprudence concerning malicious prosecution are not entirely clear misses the point. Our inquiry is whether [the defendant’s] alleged actions—arresting and detaining [the plaintiff] based on false pretenses and then seeking an arrest warrant based on these false statements—violated [the plaintiff’s] clearly established constitutional rights. We conclude that they did. Miller v. Maddox, 866 F.3d 386, 395 (6th Cir. 2017), cert. denied, 138 S. Ct. 2622 (2018). In short, “the sine qua non of the ‘clearly established’ inquiry is ‘fair warning,’” Baynes v. Cleland, 799 F.3d 600, 612–13 (6th Cir. 2015) (quoting Hope, 536 U.S. at 741), and we ask only “whether it would be clear to a reasonable officer that his conduct was unlawful in the situation he confronted,” id. at 610 (quoting Saucier v. Katz, 533 U.S. 194, 202 (2001)). Stoiker’s argument may, on the other hand, be that it was not clear in 1975 that an officer who fabricated evidence but did not testify for the prosecution had violated the laws of the United States. If this were true, he would be protected by qualified immunity. It is not, and he is not. For Plaintiffs’ claims to survive summary judgment, it must have been clearly established that where an officer fabricates evidence against a defendant and then withholds exculpatory evidence from the prosecution, but does not testify at trial or a grand jury hearing, he is “influenc[ing]” the decision to initiate the prosecution in a way that violates the defendant’s constitutional rights. Mills v. Barnard, 869 F.3d 473, 480 (6th Cir. 2017). Stoiker cites no case requiring testimony as an element of a § 1983 claim for malicious prosecution and no case suggesting that testifying is required in order to influence the decision to prosecute. To the contrary, this court held long before 1975 that if officers arrested a suspect without a warrant (in violation of state law), and “subjected [that suspect] to fraudulent trial in a criminal case” that resulted in wrongful conviction, the officers caused the suspect “a deprivation of [her] liberty without due process of law.” McShane v. Moldovan, 172 F.2d 1016, 1019 (6th Cir. 1949). The court in McShane made no mention of whether the officers had testified against the suspect, and with good cause: the crux of the violation is the institution of judicial processes without probable cause, which does not require a testimonial act. Nos. 17-3840/3843 Jackson, et al. v. City of Cleveland, et al. Page 37 In conjunction with the cases cited in section II(C)(2)(b), supra, McShane is sufficient to have clearly established before May 1975 that an officer need not testify in order to violate a defendant’s right to due process. That the phrase “malicious prosecution” was not used in that case to describe the cause of action is immaterial; what matters are the actions allegedly taken by Stoiker, not the name we give to the claim used to seek redress for those actions. Stoiker is therefore not entitled to qualified immunity on the malicious-prosecution claims. 3. Conclusion For the foregoing reasons, we REVERSE the district court’s grant of summary judgment to Stoiker as to the Fourteenth Amendment claims for fabrication of evidence and for withholding of exculpatory evidence in violation of Brady, and the Fourth Amendment claims for malicious prosecution.20 But we AFFIRM the grant of summary judgment as to the conspiracy claims. D. Cleveland’s Motion for Summary Judgment The cause of action created by § 1983 may be exercised only against a “person who . . . causes to be subjected, any citizen of the United States or other person within the jurisdiction thereof to the deprivation of any rights, privileges, or immunities secured by the Constitution and laws.” 20The 42 U.S.C. § 1983. The Supreme Court has interpreted the word concurring opinion argues that we should not reach the qualified immunity question at this stage because “[t]his case fits into the narrow category of fact intensive cases where it is appropriate for the jury to decide the issue of qualified immunity only after they decide which party they believe.” Concurring Opinion at 53. We respectfully disagree. Although it is true that a genuine dispute of material fact exists as to whether Stoiker was “present and involved in” withholding exculpatory evidence, fabricating evidence, or initiating a malicious prosecution, Concurring Opinion at 54, it is also true that if the jury finds he was involved, the only question remaining will be a purely legal one: whether Plaintiffs’ rights were clearly established. See Mitchell v. Forsyth, 472 U.S. 511, 526 (1985) (recognizing that “whether the conduct of which the plaintiff complains violate[s] clearly established law” is an “essentially legal question”); see also Brandenburg v. Cureton, 882 F.2d 211, 215 (6th Cir. 1989) (noting that the question whether a defendant’s conduct violates clearly established rights “is a legal question to be decided by the court” (citation omitted)). Indeed, the concurring opinion appears to agree that “the law of withholding exculpatory evidence, fabricating evidence, and malicious prosecution was clearly established in this context as of 1975.” Concurring Opinion at 54. Because the legal question whether Plaintiffs’ rights were clearly established in 1975 has been answered, and because if the evidence is viewed in the light most favorable to Plaintiffs, Stoiker violated those rights, it is appropriate for this court to deny Stoiker qualified immunity at this stage of the litigation. See Harlow, 457 U.S. at 816 n.26, 818–19; Heflin v. Stewart Cty., 958 F.2d 709, 717 (6th Cir. 1992). Nos. 17-3840/3843 Jackson, et al. v. City of Cleveland, et al. Page 38 “person” broadly, and certain polities, including municipalities, are considered persons for purposes of § 1983 liability. Monell v. Dep’t of Soc. Servs., 436 U.S. 658, 690 (1978). Although “person” has been given a wide meaning under § 1983, when the person is a municipality, liability attaches only under a narrow set of circumstances: “A municipality may not be held liable under § 1983 on a respondeat superior theory—in other words, ‘solely because it employs a tortfeasor.’” D’Ambrosio v. Marino, 747 F.3d 378, 388–89 (6th Cir. 2014) (quoting Monell, 436 U.S. at 691). Instead, a plaintiff must show that “through its deliberate conduct, the municipality was the ‘moving force’ behind the injury alleged.” Alman v. Reed, 703 F.3d 887, 903 (6th Cir. 2013) (quoting Bd. of Cty. Comm’rs v. Brown, 520 U.S. 397, 404 (1997)). A plaintiff does this by showing that the municipality had a “policy or custom” that caused the violation of his rights. Monell, 436 U.S. at 694. There are four methods of showing the municipality had such a policy or custom: the plaintiff may prove “(1) the existence of an illegal official policy or legislative enactment; (2) that an official with final decision making authority ratified illegal actions; (3) the existence of a policy of inadequate training or supervision; or (4) the existence of a custom of tolerance or acquiescence of federal rights violations.” Burgess v. Fischer, 735 F.3d 462, 478 (6th Cir. 2013) (citation omitted). Plaintiffs argue that they have provided evidence sufficient to make out a Monell claim under the first theory, as GPO 19-73 caused the violation of their Brady rights, and the third theory, as Cleveland’s failure to train its officers in Brady caused the violation of their Brady rights. Cleveland disagrees, as did the district court.21 21Plaintiffs also argue that they have a third Monell claim based on Cleveland’s failure to adopt an adequate policy to prevent Brady violations. The district court ruled against Plaintiffs on this theory, finding that they had not established that Cleveland had failed to adopt adequate policies to train officers in Brady’s requirements. Jackson v. City of Cleveland, CASE NO. 1:15CV989, 2017 WL 3336607, at *4 (N.D. Ohio Aug. 4, 2017). We decline to analyze this theory separately. Plaintiffs cite no Sixth Circuit or Supreme Court case in support of their theory that they have a Monell claim—separate from their failure-to-train claim—based on Cleveland’s unconstitutional failure to adopt a policy. Instead, the relevant cases they cite are failure-to-train cases. See City of Canton v. Harris, 489 U.S. 378, 391 (1989); Gregory v. City of Louisville, 444 F.3d 725, 755 (6th Cir. 2006); Miller v. Calhoun Cty., 408 F.3d 803, 816–17 (6th Cir. 2005). That makes sense: the harm alleged and the analysis required under the failure-to-train theory are functionally indistinguishable from the harm Plaintiffs allege and the analysis they wish us to conduct under the failure-to-adopt-a-policy theory. Indeed, the district court stated Nos. 17-3840/3843 1. Jackson, et al. v. City of Cleveland, et al. Page 39 Official Policy “[T]o satisfy the Monell requirements a plaintiff must ‘identify the policy, connect the policy to the city itself and show that the particular injury was incurred because of the execution of that policy.’” Garner v. Memphis Police Dep’t, 8 F.3d 358, 364 (6th Cir. 1993) (quoting Coogan v. City of Wixom, 820 F.2d 170, 176 (6th Cir. 1987)). When proceeding under the first theory of Monell liability, under which a plaintiff must show an official policy or legislative enactment, the plaintiff must show that there were “formal rules or understandings—often but not always committed to writing—that [were] intended to, and [did], establish fixed plans of action to be followed under similar circumstances consistently and over time.” Pembaur v. City of Cincinnati, 475 U.S. 469, 480–81 (1986) (emphasis added). Plaintiffs argue that GPO 19-73 reflects just such a policy: it was a formal rule; it was promulgated by Cleveland; and it was illegal because it authorized Brady violations. Cleveland agrees that GPO 19-73 was a policy and that it was promulgated by Cleveland but argues that GPO 19-73 is consistent with Brady and, therefore, could not cause Brady violations.22 GPO 19-73 read in pertinent part as follows: TO THE MEMBERS OF THE DEPARTMENT In a letter to this Department, County Prosecutor John T. Corrigan has defined the legal rights of defense attorneys and courts to statements, reports and other items in criminal cases. His letter, as a part of this order, shall be considered an integral part of criminal case preparations procedures and all members shall comply with its provisions. R. 101-7, PageID 1630. that to prevail on their failure-to-adopt theory, Plaintiffs needed to show Cleveland was deliberately indifferent to the high likelihood of violations in the absence of a policy. See Jackson, 2017 WL 3336607, at *4 (citing Miller, 408 F.3d at 816–17). As we discuss below, Plaintiffs must make the same showing for their failure-to-train claim. 22Cleveland also argues that Plaintiffs cannot show that GPO 19-73 caused any Brady violations because they cannot demonstrate any Brady violations to begin with. As discussed in section II(C)(1)(a), supra, that argument is unavailing. Nos. 17-3840/3843 Jackson, et al. v. City of Cleveland, et al. Page 40 The letter from the County Prosecutor, incorporated into GPO 19-73, read in pertinent part as follows: The Ohio Supreme Court has recently promulgated Criminal Rules of Procedure . . . . Particularly, Rule 16 is going to be the concern of police departments and prosecutors. .... NO POLICE DEPARTMENT IS REQUIRED OR SHALL GIVE TO DEFENSE COUNSEL AND/OR ANY COURT ANY RECORD, PAPER, STATEMENT, REPORT OR TANGIBLE OBJECT OF A CRIMINAL CASE. Under proper circumstances under this rule, by application to the Prosecuting Attorney and/or the court, the defense counsel may be entitled to the following: 1. 2. 3. 4. 5. 6. Statement of a defendant or co-defendant, written, recorded, or a summary of an oral statement. Defendant’s prior felony record. Inspection of [physical evidence] material to the preparation of the defense or intended for use by the Prosecuting Attorney as evidence. Reports of results of physical or mental examinations, scientific tests or experiments. Names and addresses of witnesses. Evidence favorable to the defendant. EXCEPTION TO THE FOREGOING: The foregoing does not authorize the discovery or the inspection of reports, memoranda, or other internal documents made by the Prosecuting Attorney or his agents (police departments are his agents) in connection with the investigation or prosecution of the case, or of statements made by witnesses or prospective witnesses to state agents. Id. (emphases added). Plaintiffs argue that GPO 19-73 is appropriately read as providing that defendants are generally entitled to favorable evidence, but that the entitlement does not apply if the favorable evidence is in the form of witness statements. The individual Defendants were therefore acting in conformance with GPO 19-73 when they failed to turn over to prosecutors Vernon’s uncoerced, exculpatory statement that he had not seen the shooting, even though this withholding violated Plaintiffs’ Brady rights. Nos. 17-3840/3843 Jackson, et al. v. City of Cleveland, et al. Page 41 In rebuttal, Cleveland argues, first, that the language of GPO 19-73 did not permit officers to withhold exculpatory evidence from prosecutors, and, second, that even if it did, it must be read in conjunction with other rules by which officers were bound in order to determine whether Cleveland had a policy of permitting the withholding of exculpatory evidence from prosecutors. Because a city may be liable under Monell for a policy of permitting constitutional violations regardless of whether the policy is written, see Pembaur, 475 U.S. at 480–81, we ask whether GPO 19-73 and the other rules were inconsistent with a policy of withholding evidence in violation of Brady. If they were not inconsistent, then a genuine issue of material fact exists as to whether Cleveland had such a policy, and summary judgment was improper on Plaintiffs’ first Monell theory. a. The Text of GPO 19-73 Cleveland notes that the only language in GPO 19-73 discussing officers’ disclosure obligations simply made it clear that they were not permitted to give any physical evidence directly to a defendant; it did not say officers were permitted to withhold exculpatory evidence from prosecutors. Under this reading of GPO 19-73, the purpose of the order was to ensure that officers did not give evidence directly to defendants and, perhaps as something of an addendum, let officers know what prosecutors might have to do with the evidence officers gave them. The section permitting disclosure of exculpatory witness statements, after all, argues Cleveland, was prefaced by the statement that “by application to the Prosecuting Attorney and/or the court, the defense counsel may be entitled” to various evidence, and that language appears to concern the interaction between prosecutors and defendants, not officers and prosecutors. Cleveland’s interpretation of GPO 19-73 may be plausible, but it is not the only reasonable interpretation. The incorporated letter from the County Prosecutor informed Cleveland police officers that “Rule 16 is going to be the concern of police departments and prosecutors,” not just prosecutors. (emphasis added). The incorporated letter recreated in part the text of Rule 16, which dealt only with what sorts of evidence required disclosure, not with who would do the disclosing. That GPO 19-73 told officers that Rule 16 was their concern and Nos. 17-3840/3843 Jackson, et al. v. City of Cleveland, et al. Page 42 provided the text of that Rule, which described which evidence must be disclosed, could suggest that GPO 19-73 was promulgated for the purpose of ensuring officers knew what particular evidence they had to disclose to prosecutors so that the prosecutors could then disclose that evidence to the defense. Under this reading, GPO 19-73 did more than simply inform officers that they should not give evidence directly to defendants; it also served as the directive to officers as to what evidence they should, and should not, give to prosecutors. Particular to the last point, the “EXCEPTION TO THE FOREGOING” provision could be read to direct officers not to turn over to prosecutors the documents described in that provision, which included “statements made by witnesses or prospective witnesses to state agents.” Therefore, one reasonable reading of GPO 19-73 is that it (1) spoke to police officers about their disclosure obligations and (2) informed them that they did not need to disclose exculpatory witness statements to the Prosecutor’s Office. Because GPO 19-73 can be read as consistent with a policy of not disclosing exculpatory witness statements, we turn to the other written rules to see whether any of those foreclosed such a policy. b. Other Rules Cleveland’s second argument for why a reasonable jury could not read GPO 19-73 as embodying a policy of allowing officers to withhold exculpatory witness statements from prosecutors is stronger, but ultimately unavailing. Instead of looking at GPO 19-73 in a vacuum, Cleveland urges us to consider the text of GPO 19-73 in the context of other rules and regulations. These other sources fall into two groups: sections of the Division of Police’s Manual of Rules and the full text of Rule 16. i. The Manual The Manual contained a number of rules that were applicable to all Cleveland police officers in 1975. Cleveland points to four rules as requiring, when read in conjunction with GPO 19-73, that officers disclose exculpatory witness statements to prosecutors. While all laudable, each of these policies can, however, reasonably be interpreted as consistent with a reading of GPO 19-73 that permitted officers to withhold exculpatory witness statements from prosecutors. Nos. 17-3840/3843 Jackson, et al. v. City of Cleveland, et al. Page 43 Rule 14 Case Preparation and Fraud Unit .... (2) The officer in charge shall cause statements to be taken from persons brought to the Unit in the course of criminal investigations; and shall see that such statements are properly filed and preserved. These statements shall be available only to the officers and members of the Division of Police who are interested in the presentation of a particular case, to the office of the County Prosecutor or the Law Department of the City of Cleveland. Under no circumstances shall they be given or exhibited to any other person without the written consent of the Chief of Police. R. 102-2, PageID 1910–11. Rule 14 only applied to statements given by persons brought to the Case Preparation and Fraud Unit, and there was no requirement that all witnesses be brought to that unit when giving a statement.23 Nor did it require officers to affirmatively disclose exculpatory statements to prosecutors; it required only that detectives make statements available. This reading is consistent with testimony that it was the practice of Cleveland detectives to withhold evidence not contained in arrest reports, witness forms, or written statements unless it was specifically requested by prosecutors. Rule 66 [No Title] (1) Officers and members prosecuting persons charged with a crime shall thoroughly familiarize themselves with all of the facts and details concerning such case, so that all of the evidence may be properly presented to the court. Id. at PageID 1941. Rule 66 said nothing about disclosure of evidence to prosecutors, much less exculpatory evidence. More importantly, it only required familiarity with the facts and details of a case insofar as such familiarity was required to “properly present” those facts to a court. As discussed above, GPO 19-73 can reasonably be interpreted as not requiring officers to disclose exculpatory witness statements to prosecutors for disclosure in turn to defendants (or, presumably, to courts). 23Vernon does not appear to have been brought to the Case Preparation and Fraud Unit. Nos. 17-3840/3843 Jackson, et al. v. City of Cleveland, et al. Page 44 Read in conjunction with GPO 19-73, then, Rule 66 can reasonably be interpreted as not applying to exculpatory witness statements. Rule 77 [No Title] (1) Officers and members shall report on all matters referred to or investigated by them. Such reports may be either verbal or written, as the officer in charge may direct. (2) They shall, before reporting off duty, make such written reports as may be required on all matters coming to their attention or assigned to them for investigation. If the investigation has not been completed before he reports [sic] off duty, he shall make a report stating the progress made. (3) He shall address his written reports to his superior officer and shall sign the reports, giving his full name and rank, title or number. When required, such reports shall be examined and signed by a superior officer. Written reports shall be forwarded to the commanding officer. Id. at PageID 1945–46. Rule 77 required that officers make reports concerning their cases. It did not require that such reports contain all exculpatory information that they may have learned. Importantly, it did not require that any reports be disclosed to prosecutors, and it allowed reports to be made verbally to a superior officer, a method well suited to the withholding of information from prosecutors. Rule 78 [No Title] Written and verbal reports, testimony in court and conversation of any kind affecting the Division of Police, its officers, members, employees or persons under its jurisdiction shall be truthful and unbiased. Id. at PageID 1946. Even read in conjunction with GPO 19-73 and the other rules discussed herein, the limitations of Rule 78 render it incapable of carrying the weight with which Cleveland burdens it. It required not that all reports be complete but only that they be truthful and unbiased. 24 Nor 24Consider, for example, two officers who have coerced a witness into making what they believe to be a truthful statement: a report detailing that statement but excluding the coercion would comport with Rule 78. More insidiously, consider two officers who have coerced a witness into making a statement that they know to be false and who file a report stating that the witness made that statement. That report would be both true and unbiased and therefore consistent with Rule 78: the witness did, after all, make the (false) statement contained in the report. Nos. 17-3840/3843 Jackson, et al. v. City of Cleveland, et al. Page 45 did it require that any reports, which may have been made verbally to a superior officer, be disclosed to prosecutors. None of the rules contained within the Manual, taken individually or collectively, are inconsistent with an interpretation of GPO 19-73 that permits officers to withhold exculpatory information from prosecutors. These rules can be read by a reasonable jury as consistent with a policy of permitting the withholding of exculpatory evidence in violation of Brady. ii. Ohio Rule of Criminal Procedure 16 Cleveland also argues that the full text of Rule 16 made clear that defendants were entitled to all exculpatory evidence, including witness statements. The version of Rule 16 in force in 1975 read in pertinent part as follows: (B) Disclosure of evidence by the prosecuting attorney. (1) Information subject to disclosure. .... (2) (f) Disclosure of evidence favorable to defendant. [description of exculpatory evidence] (g) In camera inspection of witness’ [sic] statement. [description of procedure for in camera inspection of witness statements] Information not subject to disclosure. Except as provided in subsections (B)(1)(a), (b), (d), (f), and (g), this rule does not authorize the discovery or inspection of reports, memoranda, or other internal documents made by the prosecuting attorney or his agents in connection with the investigation or prosecution of the case, or of statements made by witnesses or prospective witnesses to state agents. Proposed Ohio Rules of Criminal Procedure, 46 Ohio BAR 817, 849–52 (1973) (emphases added). Unlike GPO 19-73 and the rules in the Manual, Rule 16 made it quite clear that defendants were entitled to exculpatory witness reports. Rule 16(B)(2) excepted witness statements from the general disclosure requirements, using language almost identical to that used in GPO 19-73. Unlike GPO 19-73, however, Rule 16(B)(2) included an additional clause, Nos. 17-3840/3843 Jackson, et al. v. City of Cleveland, et al. Page 46 excepting exculpatory witness statements from the exception for witness statements generally. Rule 16(B)(2) thereby placed exculpatory witness statements back into the universe of mandatory disclosure. As a result, the full text of Rule 16 made it clear that prosecutors were obligated to disclose all exculpatory witness statements to defendants. This fact does not, however, save Cleveland, at least not at this stage of the litigation. Cleveland has provided no evidence that Cleveland required that its officers follow the official version of the Ohio Rules of Criminal Procedure, that those rules were followed by Cleveland officers, or that Cleveland officers were even aware of those rules. There is evidence, of course, that Cleveland officers were bound by Rule 16: GPO 19-73 can reasonably be interpreted as directing that Cleveland officers follow the requirements of Rule 16. But GPO 19-73 could be read as directing Cleveland officers to follow a modified version of Rule 16—the version included in GPO 19-73.25 And that version differed in at least one material way from the official Rule 16. Both GPO 19-73 and the official Rule 16 included a paragraph excepting witness statements from the general disclosure requirements. But although, as mentioned, that paragraph in the official Rule 16 contained an additional clause in paragraph (B)(2) excepting (by reference to the requirements of (B)(1)(f) and (g)) exculpatory witness statements from that exception—and the paragraph in the official Rule thereby required that exculpatory witness statements be disclosed—the version of Rule 16 in GPO 19-73 omitted the clause in that paragraph excepting exculpatory witness statements. The paragraph in the version of Rule 16 included in GPO 19-73 merely provided: The foregoing [disclosure requirements] do[] not authorize the discovery or inspection of reports, memoranda, or other internal documents made by the Prosecuting Attorney or his agents (police departments are his agents) in connection with the investigation or prosecution of the case, or of statements made by witnesses or prospective witnesses to state agents. 25GPO 19-73 did not indicate that the version of Rule 16 it included was in any way different from the official Rule 16 or suggest that officers either consult the official text of Rule 16 or consult prosecutors as to the duties of officers or prosecutors under Rule 16. Nos. 17-3840/3843 Jackson, et al. v. City of Cleveland, et al. Page 47 R. 101-7, PageID 1630. It was this modified version of Rule 16—the one that could plausibly be interpreted as allowing the withholding of exculpatory witness statements—by which a reasonable jury could find Cleveland officers were bound. c. Conclusion GPO 19-73 and the rules in Cleveland’s police manual, read together, could be understood to authorize Cleveland officers to withhold exculpatory witness statements from prosecutors. It is for a jury to consider GPO 19-73 and the rules in the Manual in light of Cleveland’s actual practices and determine whether Cleveland had a policy of permitting Brady violations. Because Cleveland does not contest that it promulgated GPO 19-73 or that the individual Defendants were acting in conformance with GPO 19-73 when they withheld Vernon’s exculpatory statements, a reasonable jury could find Cleveland liable under Monell. See Garner, 8 F.3d at 364–65. Because a genuine issue of material fact exists as to whether Cleveland had a policy of permitting Brady violations, the district court’s grant of summary judgment to Cleveland on Plaintiffs’ first Monell theory was improper. 2. Failure to Train Plaintiffs also argue that in 1975, Cleveland had “a policy of inadequate training or supervision” of its officers as to their obligation to disclose exculpatory evidence under Brady. Burgess v. Fischer, 735 F.3d 462, 478 (6th Cir. 2013). In order to show that a municipality is liable for a failure to train its employees, a plaintiff “must establish that: 1) the City’s training program was inadequate for the tasks that officers must perform; 2) the inadequacy was the result of the City’s deliberate indifference; and 3) the inadequacy was closely related to or actually caused the injury.” Ciminillo v. Streicher, 434 F.3d 461, 469 (6th Cir. 2006) (citing Russo v. City of Cincinnati, 953 F.2d 1036, 1046 (6th Cir. 1992)). Nos. 17-3840/3843 Jackson, et al. v. City of Cleveland, et al. Page 48 Cleveland argues that Plaintiffs lack evidence sufficient for a jury to find that either of the first two requirements is met. We address each requirement in turn. a. Adequacy of Cleveland’s Brady Training When determining whether a municipality has adequately trained its employees, “the focus must be on adequacy of the training program in relation to the tasks the particular officers must perform.” City of Canton v. Harris, 489 U.S. 378, 390 (1989). Plaintiffs argue that Cleveland’s training program was deficient in that it failed to train officers in their Brady obligations and that, to the degree that those officers received training in the disclosure of evidence to prosecutors, they were trained to withhold exculpatory evidence. Cleveland disagrees, citing deposition testimony to show that officers received official training in their disclosure obligations as well as unofficial on-the-job training in their disclosure obligations. The district court agreed with Cleveland. It is undisputed that Cleveland officers received, and were trained in, the Manual. But as discussed in section II(D)(1)(b)(i), supra, those rules could be read as insufficient to inform officers of their disclosure obligations, as none of the rules in the Manual explicitly mandated disclosure of exculpatory witness statements to prosecutors. The only rule from the Manual that came close to requiring disclosure of witness statements to prosecutors was Rule 14, which applied only to witness statements made to the Case Preparation and Fraud Unit, and it only required that those statements be available to prosecutors, not that they be proactively disclosed. Cleveland presents deposition testimony that “in the police academy . . . [y]ou were told to give [exculpatory evidence] to . . . a prosecutor” and that “Cleveland police officers are trained and instructed to turn over the entire product of their investigation to the prosecutor.” R. 103, PageID 3664, 3672. Cleveland also presents evidence that officers were trained on the job to disclose evidence. One officer testified that although the rule was not always followed, “the rule was, you should turn over all evidence acquired in an investigation to the prosecution.” R. 104, PageID 3949–50. That officer also testified: Nos. 17-3840/3843 Jackson, et al. v. City of Cleveland, et al. Page 49 Police officers that conduct interviews are instructed to write down the statement as close to what the witness said as possible, whether it is good or bad. But it is part of what was said, and it needs to be, the entire thing needs to be presented to the prosecutor, the entire thing, not parts of it, the entire thing. Id. at PageID 3972. But Plaintiffs provide testimony that conflicts with Cleveland’s account of the training received, both in the academy and on the job. One former officer testified that he was not “taught anything at police academy about police officers’ obligation to disclose Brady evidence” and that he did not remember having “ever attend[ed] any training concerning police officers’ obligation to disclose exculpatory evidence to the defense.” R. 114-20, PageID 5127–29. Another former officer testified that he could not recall having ever “attend[ed] any course, or receive[d] any training in which [he] learned that officers have an obligation to disclose exculpatory evidence to criminal defendants or prosecutors.” R. 114-35, PageID 5492– 93. A third former officer testified, “As far as training, I would have to say no” training was provided teaching officers they were required “to place any witness statements in the official file or otherwise make them available to criminal defendants, defense counsels, and prosecutors.” R. 102, PageID 1776. He also testified that there was “[n]o specific training” requiring police detectives “to disclose exculpatory evidence.” Id. at PageID 1777. The district court interpreted these statements as indicating only that no official training had been provided, not that no on-the-job training had been provided. Jackson v. City of Cleveland, CASE NO. 1:15CV989, 2017 WL 3336607, at *6 (N.D. Ohio Aug. 4, 2017). We think this a cramped interpretation of these statements. One officer testified that he could not recall having “receive[d] any training in which [he] learned that officers have an obligation to disclose exculpatory evidence.” (emphasis added). Another testified that he received “no [training] to place any witness statements in the official file.” (emphasis added). A reasonable jury could interpret this testimony as indicating that officers received no training, on-the-job or otherwise, in their Brady obligations generally or in their obligation to provide witness statements to prosecutors. Nos. 17-3840/3843 Jackson, et al. v. City of Cleveland, et al. Page 50 There is therefore a genuine issue of material fact as to whether Cleveland’s training of its officers in their disclosure obligations was sufficient, and summary judgment was inappropriate as to this issue. See Burgess, 735 F.3d at 471. b. Cleveland’s Deliberate Indifference “[D]eliberate indifference is a stringent standard of fault, requiring proof that a municipal actor disregarded a known or obvious consequence of his action.” Bd. of Cty. Comm’rs v. Brown, 520 U.S. 397, 410 (1997) (internal quotation marks omitted). “In other words, the risk of a constitutional violation arising as a result of the inadequacies in the municipal policy must be ‘plainly obvious.’” Gregory v. City of Louisville, 444 F.3d 725, 752 (6th Cir. 2006) (quoting Bd. of Cty. Comm’rs, 520 U.S. at 412). A plaintiff may meet this standard by showing either (1) “prior instances of unconstitutional conduct demonstrating that the City had notice that the training was deficient and likely to cause injury but ignored it” or (2) “evidence of a single violation of federal rights, accompanied by a showing that the City had failed to train its employees to handle recurring situations presenting an obvious potential for such a violation.” Campbell v. City of Springboro, 700 F.3d 779, 794 (6th Cir. 2012) (citing Plinton v. Cty. of Summit, 540 F.3d 459, 464 (6th Cir. 2008)). Plaintiffs do not contend that they can show Cleveland’s failure to train was deliberately indifferent via the first method.26 Instead, Plaintiffs argue that they satisfy the second method of showing deliberate indifference because the “likelihood that the situation [i.e., a situation requiring police to handle exculpatory evidence] will recur and the predictability that an officer lacking specific tools to handle that situation will violate citizens’ rights” mean that failing to train officers in their disclosure obligations demonstrates deliberate indifference to the “highly predictable consequence” that untrained officers will violate Brady. Jackson Br. at 65 (quoting Bd. of Cty. Comm’rs, 520 U.S. at 409–10). 26Finding that Plaintiffs had not shown Cleveland provided inadequate training to its officers, the district court did not address whether Plaintiffs could make out the deliberate-indifference element by either method. See Jackson, 2017 WL 3336607, at *5. Nos. 17-3840/3843 Jackson, et al. v. City of Cleveland, et al. Page 51 Plaintiffs cite Gregory in support of their conclusion. In Gregory, the plaintiff presented evidence that the defendant municipality had failed to train its officers in the handling of exculpatory evidence. See 444 F.3d at 753–54. This court reversed a grant of summary judgment to the municipality, holding that a “custom of failing to train its officers on the handling of exculpatory materials is sufficient to establish the requisite fault on the part of the [municipality]” for a deliberate-indifference claim. Id. at 754 (citing Bd. of Cty. Comm’rs, 520 U.S. at 407). Cleveland attempts to distinguish Gregory on the ground that in Gregory, the plaintiff had presented evidence sufficient for a jury to find that the defendant municipality had failed to train its officers, while in this case, Cleveland argues that the “undisputed evidence in the record establishes that the detectives in the City’s Homicide Detective Bureau received on-the-job training about the evidence that they were required to turn over to the prosecutor.” Cleveland Br. at 48. But, as discussed above, the evidence is not undisputed. Plaintiffs have provided testimony sufficient for a jury to find that Cleveland did not in fact train its officers in their disclosure obligations. Gregory therefore controls, and there is sufficient evidence for a reasonable jury to find that Cleveland was deliberately indifferent to the risk of Brady violations. 3. Conclusion Plaintiffs having shown that there are genuine issues of material fact both as to whether Cleveland had an official policy of permitting the withholding of exculpatory witness statements from prosecutors and as to whether Cleveland had a policy of failing to train its officers in their disclosure obligations, summary judgment was inappropriate on the Monell claims. See Burgess, 735 F.3d at 471.27 We therefore REVERSE the district court’s grant of summary judgment to Cleveland as to those claims. 27A failure-to-train claim under Monell also requires showing that the failure to train “was closely related to or actually caused [Plaintiffs’] injury,” Ciminillo, 434 F.3d at 469, but Cleveland does not dispute that element. Nos. 17-3840/3843 Jackson, et al. v. City of Cleveland, et al. Page 52 III. CONCLUSION In 1940, then-Attorney General Robert H. Jackson admonished prosecutors: “Your positions are of such independence and importance that while you are being diligent, strict, and vigorous in law enforcement you can also afford to be just. Although the government technically loses its case, it has really won if justice has been done.” 28 In the present case, by contrast, one law-enforcement officer testified that “winning the case was what it was all about. It wasn’t about what was fair, it wasn’t about what was honest, it was about winning.” R. 104, PageID 3967–68. If that sentiment explains the circumstances of Plaintiffs’ convictions, then those convictions were the result of a process that was the very antithesis of Jackson’s famous admonition. For the foregoing reasons, we AFFIRM the district court’s grant of summary judgment to Stoiker as to Plaintiffs’ § 1983 claims for conspiracy to fabricate evidence and withhold exculpatory evidence. We REVERSE and REMAND the district court’s (1) judgment on the pleadings for Cleveland as to Plaintiffs’ indemnification claims; (2) denial of Plaintiffs’ motions to amend their complaints to substitute the administrator of the estates of the deceased Defendants as a party in their place; (3) grant of summary judgment to Stoiker as to the § 1983 claims for withholding of exculpatory evidence in violation of Brady, fabrication of evidence, and malicious prosecution; and (4) grant of summary judgment to Cleveland as to the Monell claims. 28Robert H. Jackson, Attorney Gen. of the U.S., Address at the Second Annual Conference of United States Attorneys: The Federal Prosecutor (Apr. 1, 1940). Nos. 17-3840/3843 Jackson, et al. v. City of Cleveland, et al. Page 53 _________________ CONCURRENCE _________________ KEITH, Circuit Judge, concurring. I concur in the Court’s opinion, except for Section II(C)(2) regarding Officer Stoiker’s entitlement to qualified immunity. This case presents many factual disputes, and we are sending it back to the district court for trial to determine whether Officer Stoiker committed constitutional violations. Our circuit has been clear that the issue of qualified immunity may be submitted to a jury only if “the legal question of immunity is completely dependent upon which view of the [disputed] facts is accepted by the jury.” Miller v. Sanilac Cty., 606 F.3d 240, 247 (6th Cir. 2010) (quoting Humphrey v. Mabry, 482 F.3d 840, 846 (6th Cir. 2007)); see Brandenburg v. Cureton, 882 F.2d 211, 216 (6th Cir. 1989) (vacating and remanding for a new trial and leaving the issue of qualified immunity to be decided by the jury, even though the appellant raised the issue of qualified immunity on appeal). In Brandenburg, the court stated: [T]he jury becomes the final arbiter of appellant Sharp’s claim of immunity, since the legal question of immunity is completely dependent upon which view of the facts is accepted by the jury. If the jury determines that Sharp fired on Brandenburg without a belief that someone was in danger of serious bodily injury, then as a legal matter no reasonable officer could believe that such gunfire would not violate another’s constitutional rights. On the other hand, if the jury believes detective Sharp’s version of the facts, he will be entitled to qualified immunity. 882 F.2d at 216; see also McKenna v. Edgell, 617 F.3d 432, 437 (6th Cir. 2010) (“[W]here the legal question of qualified immunity turns upon which version of the facts one accepts, the jury, not the judge, must determine liability.” (quoting Champion v. Outlook Nashville, Inc., 380 F. 3d 893, 899 (6th Cir. 2004)). This case fits into the narrow category of fact intensive cases where it is appropriate for the jury to decide the issue of qualified immunity only after they decide which party they believe. The circumstances surrounding Officer Stoiker’s alleged involvement in the witness statement and related police report of then twelve-year-old Vernon fall squarely into this Nos. 17-3840/3843 Jackson, et al. v. City of Cleveland, et al. Page 54 category because Officer Stoiker’s liability is completely dependent upon the jury finding that there is sufficient evidence that he was present and involved in Vernon’s coercion. Because the law of withholding exculpatory evidence, fabrication of evidence, and malicious prosecution was clearly established in this context as of 1975, and we are already sending this case back to the district court, instead of the analysis this court provided in Section II(C)(2) above, a jury instruction on immunity could have resolved this issue at trial.