As national attention to sexual assaults on college campuses has intensified in recent years, so too has the debate about the constitutional due process protections to which students are entitled in campus disciplinary hearings. In 2011, the U.S. Department of Education, Office for Civil Rights, promulgated a “Dear Colleague” letter1×1. See Dear Colleague Letter: Sexual Violence, Russlyn Ali, Office for Civil Rights, U.S. Dep’t of Educ. (Apr. 4, 2011) [hereinafter Dear Colleague Letter], https://www2.ed.gov/about/offices/list/ocr/letters/colleague-201104.pdf [https://perma.cc/BE3F-238T]. The Trump Administration has indicated that it will rescind the Dear Colleague Letter and enact new rules governing campus sexual assault adjudications. Stephanie Saul & Kate Taylor, Betsy DeVos Reverses Obama-Era Policy on Campus Sexual Assault Investigations, N.Y. Times (Sept. 22, 2017), https://www.nytimes.com/2017/09/22/us/devos-colleges-sex-assault.html [https://perma.cc/LG7H-EY85]. offering informal administrative guidance and numerous recommendations for changes to disciplinary hearing systems of universities that receive federal funding. In response to the letter and ensuing investigations by the Office, schools significantly restructured their sexual assault disciplinary procedures.2×2. Aaron Nisenson, Constitutional Due Process and Title IX Investigation and Appeal Procedures at Colleges and Universities, 120 Penn St. L. Rev. 963, 969–70 (2016). One of the most significant recommendations was that universities use a preponderance of the evidence standard. See Dear Colleague Letter, supra note 1, at 10–11. For a discussion of the contentious due process debates that have arisen regarding the revised policies, see Emily Yoffe, The Uncomfortable Truth About Campus Rape Policy, The Atlantic (Sept. 6, 2017), https://www.theatlantic.com/education/archive/2017/09/the-uncomfortable-truth-about-campus-rape-policy/538974/[https://perma.cc/M9WP-7NRQ]. Students subject to sanctions under the revised regimes have responded by suing the universities in federal court, alleging violations of Title IX3×3. Title IX of the Education Amendments of 1972, 20 U.S.C. §§ 1681–1688 (2012). Title IX is a federal statute that prohibits discrimination on the basis of sex in any educational programs or activities that receive federal funding. Id. § 1681(a). and due process.4×4. See, e.g., Doe v. Ohio State Univ., 239 F. Supp. 3d 1048, 1055 (S.D. Ohio 2017). Federal and state governments may not deprive any person of “life, liberty, or property, without due process of law.” U.S. Const. amend. V; U.S. Const. amend. XIV, § 1. Procedural due process requires that a person subject to deprivation by the government receive notice and an opportunity to be heard, both of which have been defined in detail by the Supreme Court. See, e.g., Mathews v. Eldridge, 424 U.S. 319, 333 (1976) (discussing requirements for constitutionally sufficient opportunity to be heard); Mullane v. Cent. Hanover Bank & Tr. Co., 339 U.S. 306, 314 (1950) (discussing requirements for constitutionally sufficient notice). Recently, in Plummer v. University of Houston,5×5. 860 F.3d 767 (5th Cir. 2017). the Fifth Circuit held that the due process rights of two students subject to a university’s sexual assault disciplinary proceedings were adequately protected.6×6. Id. at 772. Plummer is the first precedential circuit court opinion to address the merits of a student’s due process claim arising out of a sexual assault disciplinary proceeding at a public university since the Dear Colleague Letter,7×7. See Quade v. Ariz. Bd. of Regents, No. 15-16975, 2017 WL 2814446, at*2 (9th Cir. June 28, 2017) (due process claim was barred); Faparusi v. Case W. Reserve Univ., 690 F. App’x 396, 397 (6th Cir. 2017) (appeal dismissed as moot); Doe v. Univ. of Ky., 860 F.3d 365, 371 (6th Cir. 2017) (did not reach the merits of the due process claim); Doe v. Cummins, 662 F. App’x 437 (6th Cir. 2016) (unpublished opinion). but, worryingly, the court suggested that the due process afforded to the students was constitutionally acceptable because visual evidence established the guilt of the accused.8×8. See Plummer, 860 F.3d at 774. By presuming that photos and videos can objectively depict guilt, the court, in its due process analysis, ignored a growing body of scholarly literature questioning the “objectivity” of such evidence and opened the door to lower requirements of due process in campus sexual assault cases.
In November 2011, Ryan McConnell and Natalie Plummer were undergraduate students at the University of Houston.9×9. Id. at 770. In November 2011, they were dating (they are now married). Id. McConnell met “Female UH Student” at a bar and they went to his dorm room, where they engaged in sexual activity.10×10. Id. Later, Plummer entered the dorm room to find McConnell and Female UH Student both nude and unconscious on the floor.11×11. Id. Plummer took a photo of McConnell and Female UH Student.12×12. Id. She also recorded two brief videos. The first depicts McConnell fondling the unresponsive Female UH Student.13×13. Id. The second shows Plummer leading the (still nude) Female UH Student into the dorm elevator.14×14. Id. A subsequent exam by a “Sexual Assault Nurse” found that Female UH Student suffered injuries consistent with sexual assault.15×15. Id. After the incident, Plummer posted the photo to Facebook and showed the videos to her friends.16×16. Id.
When Female UH Student submitted a complaint to the University alleging that she was a victim of sexual assault, Richard Baker, the Vice President of the University’s Office of Equal Opportunity Services, investigated the incident.17×17. Id. The University provided McConnell and Plummer with written declarations of the allegations against them, and each student retained counsel.18×18. Id. at 771. Baker authored a report concluding that McConnell and Plummer violated the University’s sexual assault policies.19×19. Id. The report stated that McConnell violated the policies “when he engaged in sexual activity with [Female UH Student]” and that Plummer did so when she “‘facilitated/encouraged the sexual assault of another [UH] student[,]’ ‘electronically recorded the sexual activity of another [UH] student and then shared that video . . . without that student’s permission[,]’ and ‘made lewd, lecherous and humiliating comments of a sexual nature against another [UH] student.’” Id. (alterations and omissions in original). Both students appealed under the University’s disciplinary procedures to a panel of University personnel, which applied a preponderance of the evidence standard in determining whether the report’s findings should be upheld.20×20. Id. At their separate hearings, “McConnell and Plummer each made opening and closing arguments, testified, presented witnesses, cross-examined witnesses, and raised legal and factual objections to the panel.”21×21. Id. at 772. Additionally, McConnell’s and Plummer’s attorneys participated in the hearings by “examining and cross-examining witnesses and making statements to the panel.”22×22. Id. However, as the dissent pointed out, the attorneys’ roles remained limited: McConnell and Plummer bore most of the responsibility for representing themselves in the hearings. Id. at 780–81 (Jones, J., dissenting). The students were notified of the evidence before the hearings, although the identities of some witnesses were redacted.23×23. Id. at 781. Female UH Student did not appear before the panel and was not deposed for the proceedings.24×24. Id. Ultimately, both panels upheld the investigation’s findings.25×25. Id. at 772 (majority opinion). The two students then appealed to Richard Walker, the University’s Vice President and Vice Chancellor for Student Affairs and Enrollment Services, and those further appeals were denied.26×26. Id. McConnell and Plummer were expelled from the University.27×27. Id.
McConnell and Plummer filed a lawsuit against the University, Baker, and Walker, complaining that the defendants violated their constitutional due process rights under 42 U.S.C. § 1983 as well as their rights under Title IX.28×28. Plummer v. Univ. of Hous., No. 4:14-CV-2959, 2015 WL 12734039, at *1 (S.D. Tex. May 28, 2015). Section 1983 creates a private right of action for “individuals who are deprived of ‘any rights, privileges, or immunities’ protected by the Constitution or federal law by any ‘person’ acting under the color of state law.” Id. at *9 (quoting Stotter v. Univ. of Tex. at San Antonio, 508 F.3d 812, 821 (5th Cir. 2007)). The district court granted the defendants’ motions for summary judgment on both the due process and Title IX claims.29×29. Id. (granting University’s motion for summary judgment on the basis of Eleventh Amendment immunity); id. at *14 (granting Baker and Walker’s motion for summary judgment on the basis of qualified immunity). In conducting its analysis, the district court found that the University “afforded Plaintiffs a level of due process adequate under Fifth Circuit case law.”30×30. Id. at *13. The court also granted the defendants’ motion to dismiss the plaintiffs’ Title IX claims of gender bias, noting that “the undisputed facts are that Plummer [(a woman)] and McConnell [(a man)] received identical process in UH’s investigations and identical disciplinary sanctions under the Policy.” Id. at *15.
The Fifth Circuit affirmed. Writing for the panel, Judge Higginson31×31. Judge Higginson was joined by Judge Wiener. determined that the process provided to McConnell and Plummer by the University was constitutionally sufficient.32×32. Plummer, 860 F.3d at 772–76. The panel relied on an application of the Mathews v. Eldridge33×33. 424 U.S. 319 (1976). test, which weighs three considerations to determine if the constitutional requirements for due process are met: (1) the private interest affected by government action; (2) the “risk of an erroneous deprivation” of that interest through the applicable procedures and the probable value of further procedural safeguards; and (3) the government’s interest, including the burdens that further procedural requirements would entail.34×34. Id. at 335. The Mathews three-factor test “was first conceived to address due process claims arising in the context of administrative law,” but has since been considered “a general approach for testing challenged state procedures under a due process claim.” Medina v. California, 505 U.S. 437, 444 (1992) (quoting Parham v. J.R., 442 U.S. 584, 599 (1979)). The court found that the first and third were both “easily identified,” and implied that they were of approximately equal weight.35×35. See Plummer, 860 F.3d at 773. The court then focused its Mathews analysis on the “risk of an erroneous deprivation” and determined that “[i]n light of the graphic conduct depicted in the videos and photo” and the “multiple, meaningful opportunities to challenge the University’s allegations, evidence, and findings” provided to the students, further procedural safeguards would not have changed the outcome of the case.36×36. Id. at 774. The panel firmly declined to articulate a general “constitutional ‘floor’ for state university disciplinary procedures”37×37. Id. at 777. and repeatedly emphasized the “unique facts” of the case.38×38. Id. at 774–75. The court also rejected McConnell and Plummer’s Title IX claims, determining that “[t]here [was] no sound basis for an inference of gender bias” in the University’s proceedings. Id. at 778.
Judge Jones dissented. She excoriated the majority for not “step[ping] in to protect students’ procedural due process right,” describing the panel opinion as “impl[ying] that because [McConnell and Plummer] were guilty, they got enough due process.”39×39. Id. (Jones, J., dissenting). Judge Jones argued that the majority’s reasoning was contrary to Carey v. Piphus,40×40. 435 U.S. 247 (1978). which held that “the right to procedural due process is ‘absolute’ . . . it does not depend upon the merits of a claimant’s substantive assertions.”41×41. Plummer, 860 F.3d at 779 (Jones, J., dissenting) (quoting Carey, 435 U.S. at 266). She discussed several aspects of the University disciplinary process that she would have held violated the students’ due process rights.42×42. Id. at 779 (emphasizing “most prominently the intermingled and inherently conflicting duties of UH Title IX Coordinator Baker”). The dissent argued that “Baker’s official Title IX position placed him in the multiple, and inherently conflicting, roles of advocating for the female student, investigating the events, prosecuting McConnell and Plummer, testifying as a witness at their hearings, and training and advising the disciplinary hearing panels.” Id. at 780. Judge Jones also offered a different accounting under the Mathews three-factor test.43×43. Id. at 782–83 (arguing that the students had compelling interests in preserving their education and reputations, that the risk of erroneous deprivation was exacerbated by central features of the disciplinary process, that additional procedural safeguards would have enhanced due process, and that the University had an interest in impartially adjudicating allegations of campus sexual assault). Finally, she argued that the majority’s citations to Flaim v. Medical College of Ohio44×44. 418 F.3d 629 (6th Cir. 2005). and Doe v. Cummins45×45. 662 F. App’x 437 (6th Cir. 2016). were inappropriate and did not adequately support the majority’s Mathews analysis.46×46. Plummer, 860 F.3d at 783–84 (Jones, J., dissenting). She noted that the Flaim court described its decision as “‘quite different from the ordinary’ student discipline matter,” thereby limiting its precedential value. Id. at 784 (quoting Flaim, 418 F.3d at 643). She also emphasized that Doe is an unpublished decision that should not be cited as precedent. Id.
The panel’s assertion that visual evidence depicted McConnell’s and Plummer’s guilt such that “further procedural safeguards would not have lessened the risk of an erroneous deprivation of [the students’] interests or otherwise altered the outcome”47×47. Id. at 774 (majority opinion). runs afoul of Supreme Court case law48×48. The Plummer court’s use of specific evidence allegedly establishing a party’s guilt as a critical part of its “risk of an erroneous deprivation” analysis has no grounding in Supreme Court precedent, which has never before suggested that the weight of the evidence against the accused may be considered as part of a due process analysis. Instead, the Court asks whether the procedures in place pose a general risk of erroneous deprivation of the rights of parties similarly situated to the accused. See, e.g., Hamdi v. Rumsfeld, 542 U.S. 507, 530–31 (2004) (plurality opinion); Lassiter v. Dep’t of Soc. Servs., 452 U.S. 18, 28–29 (1981); Parham v. J.R., 442 U.S. 584, 606–07 (1979); Smith v. Org. of Foster Families for Equal. & Reform, 431 U.S. 816, 850 (1977). and ignores a growing body of literature documenting how subconscious biases affect viewers’ perceptions of photos and videos. Data demonstrate that human perception of video imagery is often subject to various cognitive biases and distortions, which can have a number of deleterious effects on judicial proceedings and outcomes. By incorporating visual evidence into its Mathews due process analysis as objective proof of guilt, the Plummer court threatened the rights of both victims and accused students in campus sexual assault cases by inviting lower courts to disregard the dangers of viewer bias.
Significant scholarly attention to judicial overconfidence in the contents of visual evidence arose after the Supreme Court’s treatment of a video in Scott v. Harris.49×49. 550 U.S. 372 (2007). For subsequent scholarly criticism of the decision, see Adam Benforado, Frames of Injustice: The Bias We Overlook, 85 Ind. L.J. 1333 (2010); Peter Brooks, Scott v. Harris: The Supreme Court’s Reality Effect, 29 L. & Literature 143 (2017); and Dan M. Kahan et al., Whose Eyes Are You Going to Believe? Scott v. Harris and the Perils of Cognitive Illiberalism, 122 Harv. L. Rev. 837 (2009). In Scott, the Supreme Court determined that video evidence of a car chase, provided by the dashboard camera of the policeman Scott’s cruiser, “so utterly discredited [respondent’s version of events] that no reasonable jury could have believed him.”50×50. Scott, 550 U.S. at 380. The Plummer court cited Scott as support for its determination that “[i]n light of the graphic conduct depicted in the videos and photo,” McConnell and Plummer were not entitled to further procedural safeguards; it described Scott as “recognizing that the existence of undisputed video evidence . . . justified summary judgment.”51×51. Plummer, 860 F.3d at 774. But the video evidence in Scott was not “undisputed” — Justice Stevens disputed the majority’s characterization of the video in a vigorous dissent,52×52. Scott, 550 U.S. at 389–93 (Stevens, J., dissenting). Justice Stevens (then eighty-seven years old) suggested that, had the eight Justices in the majority “learned to drive when most high-speed driving took place on two-lane roads rather than on superhighways — when split-second judgments about the risk of passing a slowpoke in the face of oncoming traffic were routine — they might well have reacted to the videotape more dispassionately.” Id. at 390 n.1. and both the district court and the Eleventh Circuit had held that the case should proceed to trial.53×53. Id. at 376 (majority opinion). Similarly, the evidence in Plummer was not “undisputed,” as McConnell and Plummer denied that the video depicted a sexual assault of Female UH Student.54×54. Plummer v. Univ. of Hous., No. 4:14-CV-2959, 2015 WL 12734039, at *5 (S.D. Tex. May 28, 2015) (“Plaintiffs also allege that no direct evidence of misconduct by either Plaintiff was presented at the hearing . . . .”). Thus, the same concerns arise after Plummer as did in the wake of Scott: both majorities shaped legal outcomes around the perceived contents of contested photos or videos, disregarding the dangers of biased interpretations of visual evidence.
The Plummer court’s characterization of the visual evidence as objective is unsurprising, but it overlooks the many ways humans misperceive photos and videos. Humans process images “differently than we do text — more quickly, with a heightened (perhaps exaggerated) confidence in our understanding, and with more emotion.”55×55. Elizabeth G. Porter, Taking Images Seriously, 114 Colum. L. Rev. 1687, 1752 (2014); cf. Kahan et al., supra note 49, at 839–40 (2009) (quoting multiple Justices expressing confidence in their interpretation of the Scott videotape at oral argument). People also interpret photo and video evidence as particularly authoritative, blunting a viewer’s ability to critically evaluate the ambiguities and limitations of the contents.56×56. Brooks, supra note 49, at 145–47. Indeed, the effects of judicial overconfidence in the perception of visual evidence have been studied in several doctrinal contexts.57×57. See Porter, supra note 55, at 1758–59 (summarizing several studies that discuss the effects of “naïve realism” on copyright law, First Amendment doctrine, and trials). While humans are inclined to believe that such evidence portrays an objective truth, countless subconscious factors shape people’s perceptions of visual depictions of alleged crimes. For example, the perspective from which a video is shot affects a viewer’s understanding of the events captured on tape.58×58. See Benforado, supra note 49, at 1348–49; G. Daniel Lassiter et al., Videotaped Interrogations and Confessions: A Simple Change in Camera Perspective Alters Verdicts in Simulated Trials, 87 J. Applied Psychol. 867, 868 (2002). Although certain cases might present video evidence that approaches objective indisputability, a substantial majority of cases will present interpretive difficulties, particularly when they depict emotionally charged subject matters like possible sexual assault. So, while both the Plummer majority and dissent described the photo and video as “undisputed,”59×59. Plummer, 860 F.3d at 778; id. (Jones, J., dissenting). the judges’ perceptions of the depicted events were likely shaped by forces beyond their control.
By inviting lower courts to consider “undisputed” visual evidence when evaluating due process claims arising from campus sexual assault disciplinary hearings, Plummer encourages the pernicious influence of cognitive imperfections in the courtroom. Demographic factors — including race, gender, and age — influence viewer perception of video depictions of alleged crimes.60×60. A study that showed the Scott video to 1350 participants revealed clear demographic divergences in which participants believed Scott’s actions violated Harris’s rights. See Kahan et al., supra note 49, at 854, 886. To experience viewer bias firsthand, see Timothy Williams et al., Police Body Cameras: What Do You See?, N.Y. Times (Apr. 1, 2016), https://www.nytimes.com/interactive/2016/04/01/us/police-bodycam-video.html [https://perma.cc/G4AG-7YAK]. The risks of characterizing visual evidence as objective (thereby permitting it to shape legal outcomes) are especially stark in cases, like Plummer, that confront an alleged sexual assault.61×61. See Dan M. Kahan, Culture, Cognition, and Consent: Who Perceives What, and Why, in Acquaintance-Rape Cases, 158 U. Pa. L. Rev. 729, 753–93 (2010) (describing a study showing that perception of consent or nonconsent to sex significantly diverged along demographic lines). Accused members of minority populations long stereotyped as sexually dangerous may be particularly vulnerable to judges perceiving their actions as criminal.62×62. See Frank Rudy Cooper, Against Bipolar Black Masculinity: Intersectionality, Assimilation, Identity Performance, and Hierarchy, 39 U.C. Davis L. Rev. 853, 857 (2006) (discussing the social perception of African American men as “animalistic, sexually depraved, and crime-prone”); see also Jeannie Suk Gersen, Shutting Down Conversations About Rape at Harvard Law, New Yorker (Dec. 11, 2015), http://www.newyorker.com/news/news-desk/argument-sexual-assault-race-harvard-law-school [https://perma.cc/NK8H-SYLU] (presenting anecdotal evidence that the vast majority of sexual-misconduct complaints filed at Harvard Law School are against minorities). Conversely, implicit biases shaping perceptions of credibility may lead judges to interpret a recorded encounter to be less criminal based on the demographics of the alleged victim.63×63. See Donna Coker et al., ACLU, Responses from the Field: Sexual Assault, Domestic Violence, and Policing 13–24 (2015), https://www.aclu.org/issues/womens-rights/violence-against-women/responses-field [https://perma.cc/U8D3-TQEU] (showing that police are more likely to be dismissive or hostile toward intimate partner violence and sexual assault claims by young, poor, LGBT, and minority women); Jeffrey W. Spears & Cassia C. Spohn, The Effect of Evidence Factors and Victim Characteristics on Prosecutors’ Charging Decisions in Sexual Assault Cases, 14 Just. Q. 501, 517 (1997) (finding that the only significant predictors of prosecutorial charging decisions in sexual assault cases were victim characteristics); Karl L. Wuensch et al., Racial Bias in Decisions Made by Mock Jurors Evaluating a Case of Sexual Harassment, 142 J. Soc. Psychol. 587, 588 (2002) (showing that jurors in a sexual harassment case favored litigants of their same race and same gender). Even without particular identity-based biases, misplaced judicial confidence in visual evidence could have devastating consequences. A judge recently dismissed sexual assault charges because a video showed the accuser enthusiastically leaving a bar with the accused, even though the video did not portray the circumstances immediately surrounding the alleged assault.64×64. Hannah Alani, Judge Drops Rape Case Against U.S.C. Student, Citing Video Evidence, N.Y. Times (Aug. 5, 2017), https://www.nytimes.com/2017/08/05/us/usc-rape-case-dropped-video-evidence.html [https://perma.cc/4YHD-5X3V]. In short, surface-level treatment by judges of visual evidence endangers both parties in sexual assault cases.
Plummer appears quite vulnerable to the prejudicial influences associated with visual evidence: the court relied on the photo and video evidence in rejecting all four due process claims.65×65. Plummer, 860 F.3d at 775–76 (rejecting due process claims regarding retroactive application of the standard of misconduct, inadequate opportunity to confront Female UH Student, inadequate notice of evidence, and Baker’s dual role). The precise effects of cognitive biases on the judges in Plummer are unknowable, but the general implications of the case are clear. Photo and video evidence appears in courtrooms with increasing frequency66×66. Precise statistics on the use of visual evidence in courtrooms are difficult to find. On the growing prevalence of images in legal advocacy, see generally Porter, supra note 55, at 1725–40. as social media and video recording become ubiquitous in American culture.67×67. See id. at 1718–19 (aggregating statistics on the explosive growth of images in American society); see also Howard M. Wasserman, Orwell’s Vision: Video and the Future of Civil Rights Enforcement, 68 Md. L. Rev. 600, 600–07 (2009) (discussing the growing prevalence of audio and video recording of interactions between law enforcement and citizens); Breaking News, Radiolab (July 27, 2017, 10:09 PM), http://www.radiolab.org/story/breaking-news/ [https://perma.cc/RGM7-WJUG] (discussing emerging technology capable of fabricating videos). So, the question of how visual evidence should be evaluated by legal decisionmakers seems quite likely to emerge again in the near future, especially in the university sexual assault disciplinary hearing context.68×68. See Franziska Roesner et al., Sex, Lies, or Kittens? Investigating the Use of Snapchat’s Self-Destructing Messages, in Financial Cryptography and Data Security 64, 66–67 (Nicolas Christin & Reihaneh Safavi-Naini eds., 2014) (showing that 14.2% of survey respondents reported having used the social media platform Snapchat to exchange sexual images). Moving forward, courts should leave the misplaced confidence of Scott and Plummer behind and acknowledge the risks that cognitive biases pose to judges’ perceptions of visual evidence.69×69. For innovative suggestions about how courts can systematize their approach to visual evidence, including developing “canons of visual interpretation,” see Porter, supra note 55, at 1775–81. The rights of both victims and the accused depend on it.