Frantz Bernard v. East Stroudsburg University

NOT PRECEDENTIAL UNITED STATES COURT OF APPEALS FOR THE THIRD CIRCUIT _____________ No. 16-1598 _____________ FRANTZ BERNARD; TIMOTHEUS HOMAS; ANTHONY ROSS; DEJEAN MURRAY; WILLIAM BROWN; JERRY SALTER v. EAST STROUDSBURG UNIVERSITY; DARRELL T. COVINGTON, as Members of the Board of Trustees of East Stroudsburg University; AMY SCHAEFFER WELCH, as Members of the Board of Trustees of East Stroudsburg University; TRUDI Q. DELINGER, as Members of the Board of Trustees of East Stroudsburg University; HARRY F. LEE, as Members of the Board of Trustees of East Stroudsburg University; HUSSAIN G. MALIK, as Members of the Board of Trustees of East Stroudsburg University; NANCY V. PERRETTA, as Members of the Board of Trustees of East Stroudsburg University; L. PATRICK ROSS, as Members of the Board of Trustees of East Stroudsburg University; DAVID M. SANKO, as Members of the Board of Trustees of East Stroudsburg University; ROBERT WILLEVER, as Members of the Board of Trustees of East Stroudsburg University; ELI BERMAN, as Members of the Board of Trustees of East Stroudsburg University; ROBERT J. DILLMAN, Individually and as President of East Stroudsburg University; ISAAC W. SANDERS, Individually and as Vice President for University Advancement of East Stroudsburg University; KENNETH BORLAND, Individually and as Provost of East Stroudsburg University; VICTORIA L. SANDERS, Individually and as Associate Vice President for Special Projects, Diversity and Equity of East Stroudsburg University Frantz Bernard; Timotheus Homas; Anthony Ross, Appellants _____________ APPEAL FROM THE UNITED STATES DISTRICT COURT FOR THE MIDDLE DISTRICT OF PENNSYLVANIA (Civ. Action No. 3:09-CV-00525) District Judge: Honorable Robert D. Mariani ______________ Submitted Under Third Circuit L.A.R. 34.1(a) July 14, 2017 ______________ Before: GREENAWAY, JR., SHWARTZ, and RESTREPO, Circuit Judges. (Opinion Filed: August 16, 2017) ______________ OPINION* ______________ GREENAWAY, JR., Circuit Judge. I. INTRODUCTION Appellants are former students at East Stroudsburg University (“ESU” or “the University”), a public university that is part of the Pennsylvania State System of Higher Education (“PASSHE”). Each alleges that Appellee Isaac Sanders, who served as the * This disposition is not an opinion of the full Court and pursuant to I.O.P. 5.7 does not constitute binding precedent. 2 Vice-President for Advancement at ESU and the CEO of the East Stroudsburg University Foundation, sexually assaulted or sexually harassed them while they were students. They further allege that top university officials (together, the “University Appellees”)—Robert Dillman, then the President of ESU1; Kenneth Borland, then the Provost and briefly Acting President; and Victoria Sanders, the Assistant to the President and the Associate Vice-President for Special Projects (and no relation to Isaac Sanders)—failed to protect them from Sanders, in violation of Title IX, 42 U.S.C § 1983, and the related conspiracy provisions of 42 U.S.C. §§ 1985-86. Appellants now challenge aspects of the District Court summary judgment decisions, as well as two evidentiary issues. We find no basis to disturb the District Court’s rulings. We will affirm. II. BACKGROUND In August 2007, Appellant Frantz Bernard accused Sanders of sexual assault, beginning in May of that year. Those accusations were quickly brought to the attention of the University Appellees: first Victoria Sanders, then Dillman. Prior to this, the University Appellees had received no concrete accusations that Sanders had sexually assaulted students, although there may have been rumors circulating among some members of the University community. The University separated Bernard from Sanders, whom Bernard had been working for, and began an internal investigation, conducted by Arthur Breese, the Director of Diversity and Campus Mediation. Breese interviewed 1 Dillman is no longer a party to this case. The District Court found that after Dillman passed away, plaintiffs did not substitute a proper party, as required by Fed. R. Civ. P. 25(a), and dismissed him from the suit. This ruling is not challenged on appeal. 3 Sanders, Bernard, and the two witnesses they identified who were willing to speak. Breese found the situation to be “one person’s word against another” and offered no clear conclusion as to what had occurred. D.C. Dkt. No. 94 ¶ 76; D.C. Dkt. No. 109 ¶ 76; The final report was sent to Dillman in December 2007, and Dillman and Breese discussed the report in a very brief telephone conversation at that point. Based upon the report, Dillman determined that there was insufficient evidence to support Bernard’s allegations. Throughout the litigation, Appellants have argued that the Breese investigation was insufficient, perhaps by design. The next year, in June, multiple additional students, including Appellants Timotheus Homas and Anthony Ross, came forward with allegations that Sanders engaged in sexual assault and sexual harassment. Appellants’ allegations all concerned acts taken before August 2007, when Bernard first complained to the University. ESU, now working in closer conjunction with the state PASSHE system, quickly suspended Sanders, then in September 2008 hired an outside law firm to conduct an investigation. The University sent Sanders a termination letter in October, effective in December 2008. The outside law firm first sent Dillman an interim report on its findings in September 2008 (the “PASSHE Report”), and then sent PASSHE a final version in February 2009. This investigation was much broader-ranging than Breese’s. This lawsuit was filed in March of 2009. Over the course of the litigation, various plaintiffs and defendants were dismissed from the case; those decisions are not challenged on appeal. The District Court granted summary judgment for the University Appellees but allowed the § 1983 suits against Isaac Sanders, for direct violations of the 4 right to bodily integrity, to proceed to trial. A jury found for Sanders on all counts and after extensive post-trial motion practice, this appeal commenced.2 Because the appeal concerns disparate aspects of the litigation, we discuss additional facts as needed, as well as the appropriate standards of review, in our analyses of Appellants’ four claims on appeal. III. DID THE DISTRICT COURT ERR IN GRANTING SUMMARY JUDGMENT FOR THE UNIVERSITY APPELLEES? Appellants assert that summary judgment was improperly granted. Our review of a grant of summary judgment is “plenary” and the court should “apply the same test the district court should have utilized initially.” Giles v. Kearney, 571 F.3d 318, 322 (3d Cir. 2009). Summary judgment should be granted only when “there is no genuine dispute as to any material fact and the movant is entitled to judgment as a matter of law.” Fed. R. Civ. P. 56(a). As the District Court properly recognized, liability under Title IX requires that “an official of the school district who at a minimum has authority to institute corrective measures on the district’s behalf has actual notice of, and is deliberately indifferent to, the teacher’s misconduct.” Gebser v. Lago Vista Indep. Sch. Dist., 524 U.S. 274, 277 (1998). “Actual notice” must be based on more than a “possibility.” Bostic v. Smyrna Sch. Dist., 418 F.3d 355, 361 (3d Cir. 2005). Deliberate indifference requires a response (or failure to respond) that is “clearly unreasonable in light of the known circumstances.” Davis ex 2 The District Court had jurisdiction over this case pursuant to 28 U.S.C. §§ 1331 and 1343(a)(3), (4). We have appellate jurisdiction pursuant to 28 U.S.C. § 1291. 5 rel. LaShonda D. v. Monroe Cty. Bd. of Educ., 526 U.S. 629, 648 (1999); see also Doe v. Mercy Catholic Med. Ctr., 850 F.3d 545, 566 (3d Cir. 2017). Supervisors like the University Appellees are also only liable under Title IX if their acts of deliberate indifference “cause students to undergo harassment or make them liable or vulnerable to it.” Davis, 526 U.S. at 645 (internal quotation marks omitted). Supervisory liability under § 1983, for violations of the due process right to bodily integrity, cannot be based on vicarious liability. Bistrian v. Levi, 696 F.3d 352, 366 (3d Cir. 2012). The District Court determined—and Appellants do not challenge—that the relevant theory of supervisory liability is that a supervisor may be liable “if he or she participated in violating the plaintiff’s rights, directed others to violate them, or, as the person in charge, had knowledge of and acquiesced in his subordinates’ violations.” A.M. ex rel. J.M.K. v. Luzerne Cty. Juvenile Det. Ctr., 372 F.3d 572, 586 (3d Cir. 2004). In an acquiescence context like this one, a plaintiff must show “(1) contemporaneous knowledge of the offending incident or knowledge of a prior pattern of similar incidents, and (2) circumstances under which the supervisor’s inaction could be found to have communicated a message of approval.” C.H. ex rel. Z.H. v. Oliva, 226 F.3d 198, 202 (3d Cir. 2000). On appeal, Appellants do not contest that the District Court applied the proper legal framework to their claims. Rather, their claims on appeal are limited to the District Court’s analysis of the record. Although Appellants do not organize their argument in this way, they now cite four categories of fact they believe were not properly appreciated by the District Court and which should have precluded summary judgment: 1) Breese’s 6 report was incomplete and/or incompletely communicated to the University Appellees; 2) Breese was intentionally constrained and prevented from pursuing important avenues of investigation; 3) the administration was aware of rumors about Sanders prior to the Bernard allegations; and 4) a former state official had warned the University of the shortcomings in its anti-discrimination policies. They also argue that the District Court should have contrasted the Breese Report and the PASSHE Report and thereby revealed the deliberate indifference with which the less-thorough first investigation was performed. None of these arguments suffice to overcome summary judgment, in large part because they do not speak to the knowledge or deliberate indifference of the University Appellees themselves.3 With respect to the Breese Report, Appellants cite various shortcomings of the investigation: Dillman did not talk through the evidence in the report with Breese; Breese violated the school policy for investigations by not explicitly stating in his report whether Sanders had “more likely than not” violated the school harassment policies, App. 291; and Breese found Bernard credible in his accusations. They also accuse the University, through Victoria Sanders, of limiting Breese’s investigation to exclude financial improprieties, anonymous letters sent to the university about Sanders, and a lewd illustration Sanders sent to Bernard. 3 Appellants also leave undisturbed an important aspect of the District Court’s summary judgment decision: that no evidence showed any discriminatory conduct after Bernard initially reported the alleged sexual abuse and sexual harassment, meaning that there were no injuries that could be causally traced to the University Appellees’ deliberate indifference. This waived argument appears to provide another basis for affirming the District Court’s summary judgment decision, at least in large part. 7 As the District Court discussed, these alleged shortcomings do not show deliberate indifference or acquiescence, as needed for liability. There is no indication in the record, or basis in the law, to conclude that Dillman was obligated to discuss Breese’s report with him at any length or that it was clearly unreasonable for him not to. Nor is there any indication that Dillman did not consider the report before reaching his determination. It is true that the school’s policy required an explicit finding, and arguably, Breese did not provide one.4 But in any event, Dillman, the relevant decisionmaker, was provided with the factual basis for the complaint and, as noted, took some action. Appellants put forward no evidence sufficient to create a genuine dispute that Dillman would have acted differently had Breese included a formal conclusion—particularly if his conclusion was equivocal. Nor is there evidence in the record sufficient to survive summary judgment that Breese’s investigation was hampered. The lewd illustration, notably, was provided to Breese by Victoria Sanders herself. This hardly shows an effort by Victoria Sanders or the University to obstruct the investigation. Rather, Breese testified that he left it out of his factfinding report because it was already known to, and more importantly, being addressed by, the University. Likewise, the University took clear action in response to 4 The District Court interpreted an emailed note from Breese that “it is difficult to ascertain if anything happened,” App. 234, as equivalent to a finding that it was not more likely than not that a violation occurred. On summary judgment, it may be impermissible to infer that this was a finding. Breese’s email concerned a draft report, which was later finalized, and sought advice on next steps; the quoted language could be read as describing interim uncertainty, not a final finding. It is also worth noting, as the District Court did, that Breese described the allegations as “one person’s word against another.” D.C. Dkt. No. 94 ¶ 76. 8 the anonymous letters, sent to the University and its trustees, accusing Sanders of sexual misconduct. These letters were not presented to Breese (it is disputed whether this was pursuant to a University policy against accepting anonymous letters not provided by the complainant or respondent in internal investigations). However, Dillman did not ignore the letters. Dillman spoke with the police about the letters’ allegations, to see if their content could be corroborated, and through ESU attorneys, forwarded some of the letters to the FBI. Finally, while it is undisputed that Breese was instructed not to investigate Sanders’ financial issues generally, Breese did investigate Sanders’ finances insofar as they related to Bernard, the subject of his investigation. Moreover, Breese was limited in this respect because a different, broader investigation of the financial aspects of the case was underway. None of these limitations indicated deliberate indifference by the University Appellees, who were aware of and responding to each piece of information ostensibly missing from Breese’s report. The rumors that Appellants cite do not rise to “actual notice” or “contemporaneous knowledge” of the relevant misconduct, as the District Court correctly held. Appellants cite to a statement from police officer Randy Nelson. Nelson recounted that he had heard that Sanders was gay, that he was “weird and a little different,” that he closed the door to his office at times, that Sanders would leave his office late at night with another man, and most specifically, that Sanders was once caught in another male’s car, at night, in the dark. App. 441. None of these rumors—even if they could be taken as reliable— concerns sexual assault of a student. And Nelson never stated, or even implied, that he or 9 anyone else had communicated these rumors to Dillman or other top administrators. The University Appellees have not been shown to have known of these rumors, much less to have been deliberately indifferent to reports of Sanders’ sexual assaults or harassment. Likewise, Appellants put forward statements by Charmaine Clowney, a state official in charge of university harassment policies, that the ESU anti-harassment policy was “weak and ineffective” and the worst of the policies of the fourteen state universities in the system. App. 506. But Gebser requires deliberate indifference not merely to the general possibility of sexual harassment, but deliberate indifference “to . . . the teacher’s misconduct.” 524 U.S. at 277. Supervisory liability under § 1983, in this context, similarly requires participation in or acquiescence to the subordinate’s violations. A.M. ex rel. J.M.K., 372 F.3d at 586. Clowney’s statements reveal nothing about the top University officials’ knowledge of or reaction to Sanders’ misconduct. Finally, the District Court did, in fact, refer to the PASSHE Report, at least to the extent that it was entered into the record. It did not, as Appellants would have liked, compare the scope of that report to the scope of the Breese Report as evidence of the latter’s insufficiency. But highlighting that comparison is simply another attempt to show that the Breese Report was so flimsy that it must have been a cover-up. For the reasons set forth already, and at greater length by the District Court, the record does not support this. Additional reference to a document written after Appellees’ relevant actions occurred does not change that conclusion. At the end of the day, Appellants may—or may not—have shown genuine disputes of material fact as to whether someone could have known about Sanders’ 10 behavior or whether Breese’s report could have been wider-ranging and more complete. But they did not show that the University officials who were the defendants in this case were deliberately indifferent to actual notice of Sanders’ misconduct, nor that they acquiesced in it.5 IV. DID THE DISTRICT COURT ERR IN ITS TREATMENT OF PAGES MISSING FROM THE SUMMARY JUDGMENT RECORD? Appellants next argue that the District Court erred by granting summary judgment without having read the full PASSHE Report, which had not been completely entered into the record at that point. In September 2010, the parties signed a Confidentiality Agreement governing the PASSHE Report (as well as the final, 2009 version of that report). The Confidentiality Agreement contemplated the filing of the report as an exhibit under seal in the litigation, but did not purport to file the report at that time. At a pre-trial hearing in October 2014, it became clear to the parties that the full report had never been introduced into the record, perhaps because the plaintiffs were confused about the procedures for filing the report under seal or perhaps due to technical difficulties with 5 We also affirm the District Court’s grant of summary judgment on Appellants’ conspiracy claims. The Court found no evidence of a meeting of the minds, as required for liability, nor evidence of class-based discriminatory animus. See Farber v. City of Paterson, 440 F.3d 131, 134 (3d Cir. 2006) (setting forth standard for liability under § 1985); Rogin v. Bensalem Twp., 616 F.2d 680, 696 (3d Cir. 1980) (“transgressions of § 1986 by definition depend on a preexisting violation of § 1985”). Appellants barely respond to this finding. To the extent that Appellants allege that evidence that the University “covered up” Sanders’ actions shows the existence of a conspiracy, this still would not show a meeting of the minds, much less an agreement to discriminate on the basis of a protected class. 11 the Court. In any case, in the end only the first 24 of the 31 pages that plaintiffs had sought to file were included in the summary judgment record. This issue was addressed twice in the District Court. Plaintiffs first argued that the Court had not considered the PASSHE report in the reply brief for their Rule 59 motion. The District Court clarified at that point that it had, in fact, seen 24 out of 31 pages of the report and noted that it was the plaintiffs themselves who had inadvertently submitted an incomplete document. Then, in July 2016, plaintiffs filed a Rule 60 motion, which the District Court denied, seeking correction of the record to include the full PASSHE Report and relief from judgment. The District Court noted, among other things, that 1) the report was generally immaterial to its determinations about deliberate indifference; 2) the plaintiffs incorrectly alleged that the Court had failed to review any of the PASSHE Report, when in fact it had reviewed the first 24 pages; 3) the remaining seven pages of material provided no new, relevant information; 4) plaintiffs cited no case law supporting their request and did not specify which portion of Rule 60(b) supported their claim for relief (i.e., mistake, inadvertence, surprise, or excusable neglect); 5) plaintiffs had received ample notice of the problem and done nothing; and 6) the problem was at least substantially the fault of plaintiffs’ counsel. Appellants, however, do not challenge this Rule 60 motion. Rather, they now argue that the District Court should, after the October 2014 hearing, have sua sponte 12 reversed its summary judgment orders pursuant to Rule 59(e).6 A Rule 59 motion must rely on “(1) an intervening change in controlling law; (2) the availability of new evidence; or (3) the need to correct clear error of law or prevent manifest injustice.” Lazaridis v. Wehmer, 591 F.3d 666, 669 (3d Cir. 2010). Although Appellants do not specify, only the “manifest injustice” prong appears applicable here. We review a District Court’s denial of a Rule 59 motion for abuse of discretion. Id. Appellants offer no argument that the District Court was obligated to amend its judgment sua sponte. Thus, we see no basis for identifying anything in the Court’s proceeding towards trial, along the ordinary course of civil litigation, as a “clearly erroneous finding of fact, an errant conclusion of law or an improper application of law to fact.” Hagan v. Rogers, 570 F.3d 146, 152 (3d Cir. 2009). What is more, while Appellants do not challenge the District Court’s Rule 60 decision directly, we take its reasoning to helpfully explain the District Court’s use of its discretion in not reconsidering its summary judgment decision under Rule 59. And Appellants do not address, much less rebut, the District Court’s arguments: they do not show that the remaining seven pages would have been material to the Court’s decisionmaking, for example, and they do not respond to the point that the Court actually had engaged with the pages of the report it was provided. This leaves no basis for holding that a manifest injustice occurred. It was no abuse of discretion for the Court not to reconsider its 6 Appellants specifically claim to seek relief under Rule 59(e)(3). As there is no such subsection, we take their claim to be under Rule 59(e). 13 summary judgment decisions, sua sponte, in response to the plaintiffs’ failure to properly file its materials. V. DID THE DISTRICT COURT ERR IN EXCLUDING THE PASSHE REPORT AT TRIAL? Appellants next contest the District Court’s ruling that the PASSHE Report was not admissible at trial. The defendants attempted to exclude the report pursuant to Federal Rule of Evidence 803(8), which exempts certain public records from the hearsay rule. The District Court ruled against them on three grounds. First, it held that the Report was not a public record because its author, an outside law firm, was not a public agency. Second, the Court found that the report was not trustworthy—as Rule 803(8)(B) requires for admissibility—both because the version of the report at issue was an interim report, not yet finalized, and because it was replete with hearsay-within-hearsay. Third, the District Court determined that the admission of the report, which reaches conclusions as the ultimate issue at trial, would be highly prejudicial and threaten to usurp the jury’s role as factfinder. We review for abuse of discretion. United States v. Price, 458 F.3d 202, 205 (3d Cir. 2006) (“application of the relevant hearsay exceptions is subject to review for abuse of discretion.”). We need not decide whether reports commissioned by public agencies, but written by private entities, are public records under Fed. R. Evid. 803(8), a question which appears not to have been decided in this circuit. See Watts v. City of Hartford, No. 3:00CV0681 (RNC), 2004 WL 717132, at *4 n.9 (D. Conn. Mar. 31, 2004) (collecting 14 cases). The District Court’s other two reasons for deeming the report inadmissible are sufficient to affirm. The District Court’s two reasons for finding the Report untrustworthy both are well-grounded in our case law. See Coleman v. Home Depot, Inc., 306 F.3d 1333, 1342 n.4 (3d Cir. 2002) (listing both hearsay-within-hearsay and non-finality as indicators of untrustworthiness). Appellants offer other reasons that the Report could be considered trustworthy—the quality of the investigators and the lack of bias by the PASSHE system, in particular—but they offer no persuasive response to the District Court’s concerns. On appeal, they do not acknowledge the interim nature of the report they sought to admit, much less provide a reason that its interim nature does not indicate a lack of trustworthiness. As to the hearsay-within-hearsay problem, they argue only that the layered hearsay issues were unavoidable and mitigated by the inclusion of hearsay- within-hearsay on all sides of the internal investigation. We do not generally consider more hearsay to be the solution to hearsay problems, and this does not convince us that the District Court’s determination was an abuse of discretion. Indeed, these arguments for the Report’s trustworthiness only exacerbate the Rule 403 problems that also concerned the District Court. The more the Report is made trustworthy because it compiles and analyzes hearsay from both sides, the more it resembles an independent arbitration of what occurred and the more it could distract the jury from its need to make independent credibility determinations on the basis of admissible evidence—or so the District Court could have determined. We find no abuse 15 of discretion in the District Court’s balancing of these various considerations under Rule 803(8) and its decision to exclude the Report. VI. DID THE DISTRICT COURT ERR IN EXCLUDING TESTIMONY ABOUT PAST SEXUAL MISCONDUCT? Mid-trial, the District Court held three evidentiary hearings to determine whether to allow evidence that Sanders had previously sexually assaulted three individuals other than the plaintiffs: Dejean Murray, William Brown, and Bryan Haskins. The Court allowed part of Haskins’ testimony to be admitted into evidence, none of Murray’s, and only a small amount of Brown’s deposition (not the parts relevant to showing a pattern of sexual assaults). Appellants challenge these evidentiary rulings. Our approach to the admissibility of past acts involving sexual assault is set out in Johnson v. Elk Lake School District, 283 F.3d 138 (3d Cir. 2002). First, the trial court must decide “whether a reasonable jury could find by a preponderance of the evidence that the past act was an ‘offense of sexual assault’ under Rule 413(d)’s definition and that it was committed by the defendant.” Id. at 154-55. Then, the trial court “may still exclude [the evidence] under Federal Rule of Evidence 403” if “its probative value is substantially outweighed by the danger of unfair prejudice, confusion of the issues, or misleading the jury, or by considerations of undue delay, waste of time, or needless presentation of cumulative evidence.” Id. at 155 (quoting Fed. R. Evid. 403). That balancing test should place “a thumb on the scale in favor of admissibility” if the past act is “demonstrated with specificity” and “is sufficiently similar to the type of sexual assault allegedly committed by the defendant.” Id. at 155-56. Otherwise, there is no 16 presumption of admissibility. Id. at 156. In any case, even in this context “the Rule 403 balancing inquiry is, at its core, an essentially discretionary one that gives the trial court significant latitude to exclude evidence.” Id. We review the District Court’s evidentiary ruling for abuse of discretion. Id. An offense of sexual assault is defined by Rule 413(d) to include, among other things, “contact, without consent, between any part of the defendant’s body—or an object—and another person’s genitals or anus,” and “contact, without consent, between the defendant’s genital or anus and any part of another person’s body.” Fed. R. Evid. 413. The District Court’s evidentiary rulings involved a close analysis of the three witnesses’ descriptions of how they were touched by Sanders and whether those descriptions constituted sexual assault under Rule 413. Dejean Murray testified, among other things, that Sanders touched him “[n]ear my crotch area,” App. 650, on his thigh and “towards my crotch,” App. 656, “on my crotch,” App. 653, and in the “general area” of, but not touching, his penis, App. 657. William Brown testified that Sanders put his hand on his thigh, “moving it up . . . towards my genital area and kind of like rubs it a little bit.”7 App. 463. He then testified that Sanders “reach[ed] for my zipper” at which point he moved Sanders’ hand away. App. 463. 7 Brown’s testimony, unlike the other two potential witnesses, was put forward in the form of a deposition transcript, as he could not be reached by Appellants’ counsel. 17 Bryan Haskins testified that Sanders touched his buttocks while Haskins was stepping out of a car. He testified that the touch was “near the center of my butt cheeks,” App. 609, near the anus. He also testified that Sanders touched his arm inappropriately, stared at him while he used a urinal, and touched him “on my knee rubbing up my thigh towards my genitals,” a location he further described as “toward the inside of my thigh.” App. 615-16. The District Court organized its rulings around whether the testimony was “clear” or “equivocal” in order to determine whether, and with what specificity, a jury could conclude that the described acts constituted “sexual assault” under Rule 413. App. 46. It found that Haskins’ description of being touched on the buttocks was clear—and allowed that testimony to be admitted—but that the other testimony was all equivocal, describing touching “near” or “towards” the proscribed areas but not touching them directly. Finding that these ambiguities removed or weakened any presumption of admissibility under Johnson, the Court then deemed the equivocal testimony inadmissible based on a Rule 403 balancing analysis. We find no abuse of discretion in the District Court’s decision to exclude this evidence. The District Court followed the Johnson framework and, as part of that analysis, engaged in a close analysis of the relevant evidence. Given the importance of live testimony here, that analysis deserves deference. In determining whether the alleged acts—inappropriate and unwanted though they might have been—fell within the bounds of Rule 413, or with what specificity they were described, the District Court had the benefit of physical indications and demonstrations of where the witnesses said they were 18 touched.8 But regardless of the District Court’s analysis under Rules 413 and 415, it ultimately decided based on Rule 403 grounds. We give district courts “significant latitude” to engage in Rule 403 balancing, even when sexual assaults are implicated under Rules 413 and 415. Johnson, 283 F.3d at 156; see also United States v. Kellogg, 510 F.3d 188, 197 (3d Cir. 2007) (a Rule 403 analysis should be reversed only if the District Court’s decision was “arbitrary or irrational” because “if judicial self-restraint is ever desirable, it is when a Rule 403 analysis of a trial court is reviewed by an appellate tribunal”). We cannot say the District Court abused that substantial discretion here. Given the Court’s analysis of the factual content of the testimony, it could have perceived the acts described to be on the edge of admissibility under Rule 413 and Rule 415, diminishing the presumption of admissibility and allowing the risk of prejudice from past acts testimony to loom larger. Cf. Johnson, 283 F.3d at 153 n.8 (noting “higher risk of unfair prejudice” when past acts of sexual assault are presented to jury). On the other side of the scales, the District Court observed that four witnesses—the three plaintiffs and Haskins, whose past acts testimony was partially allowed—were able to testify at trial about Sanders’ sexual assaults, enough for plaintiffs to begin to establish a pattern of behavior. See id. at 156 (Rule 403 balancing should consider “the need for evidence 8 We lack the benefits of live testimony, one reason for our deference to district courts in this area. Appellants attempt to rectify the informational deficits of appellate courts by including in their briefs what appear to be photographs of reenactments of the testimony. These photographs are not part of the record on appeal, however. Moreover, they lack the most basic indicia of reliability. We have no reason to believe that counsel’s restaging of what occurred match the witnesses’ representations, much less reality. 19 beyond the testimony of the defendant and alleged victim”). Given these considerations, among others, it was within the Court’s discretion to deem the inclusion of this testimony more prejudicial than probative. VII. CONCLUSION For the above reasons, we conclude that each of Appellants’ arguments on appeal is unsuccessful. We therefore affirm. 20