How Judges of Campus Sex Offenses Are ‘Trained’

I’ve written frequently about the unfair, guilt-presuming processes that colleges and universities from Harvard to Occidental use when deciding sexual assault cases. But a second trend has occurred largely outside the public eye. As they have “reformed” their sexual assault procedures, colleges and universities also have increasingly instituted training programs for members of these disciplinary panels—a practice not used for panelists that hear other forms of campus discipline. Because virtually no training material has been made public, it’s impossible to determine how many schools specifically train sexual assault panels. But the demand is a consistent one among anti-due process advocates.

It seems plausible to infer that these “training” materials increase the likelihood that panelists will come back with guilty findings, but for the most part, it must remain an inference: to the best of my knowledge, no university has publicly posted its training materials. One school’s—Stanford’s—was obtained by FIRE, and contained such astonishing guidance as an accused student acting “persuasive and logical” could be interpreted as a sign of guilt. Portions of a second school’s—Duke’s—appeared as a result of the McLeod litigation, and did nothing to reassure concerns about unfairness.

A more general set of model guidelines now has appeared. Developed by the University of Pennsylvania’s general counsel and director of office of student conduct (with “appreciated assistance” from the director of the school’s Women’s Center) and “intended for adaptation by individual academic institutions” as a “template” for the post-Dear Colleague letter university, the model guidelines confirm that the special training sexual assault panelists receive only compounds the due process problems on this issue.

Of the 17 tips in this self-described “fair and effective way to resolve disputes,” ten are (at best) neutral. Six would seem to make a panelist more likely to vote to find guilt. None would seem to make a panelist more likely to vote against guilt. One—regarding alcohol—is ambivalent, and seems likely to disappear when the next round of model guidelines revisions occur.

The neutral tips are common-sense observations that few, if any, people at a school like Penn would find surprising. For instance, panelists are told that sexual assaults might not feature “serious” physical injuries. Or that victims suffer “profound, long-lasting psychological injury.” Or that the brain stores traumatic memories differently than non-traumatic ones. The Penn document contains a stray comment about the accused student being presumed innocent, but that guidance doesn’t reach the level of one of the 17 “tips.” Neutral “tips,” it seems, are only those that apply to the accuser’s situation.

Then there are the “tips” that seem designed to make panelists more likely to return a guilty finding. For instance, Penn’s panelists are informed about the numerous “myths and stereotypes” (or “victim-blaming”) surrounding sexual assault. The sole source? A list compiled from a single judge, in New York, more than 20 years ago. That sort of sourcing would generate an F in most History papers at Penn.

And the panelists also encounter the oft-cited statistic that one in five women will be victims of sexual assault or attempted sexual assault while in college. The apparent message: a typical college campus has a higher rate of violent crime than inner-city Detroit, so don’t look too closely at due process. (That message, it seems, was the focus of Richard Pérez-Peña’s latest “scoop,” a revelation that 17 percent of women who responded to an MIT survey claim they had been sexually assaulted, even as most also admitted they didn’t report the offense to anyone, often because they didn’t see the episode as a clearcut crime—a finding that, if true, would suggest that the Cambridge Police Force should assign a special violent crimes unit solely to the MIT campus.)

Perhaps the most troubling of Penn’s “tips” is item 14: “False allegations of rape are not common.” The explanation: “One myth is that, out of revenge or other motives, women make numerous false allegations of sexual assault. Recent research at a major northeastern university suggests that, over a ten-year period, 5.9% of its cases involved false allegations. Other reputable research places the rate in the general population between 2% and 10%.”

The sole citation for this “tip” is an article co-authored by David Lisak (a favorite of campus due process opponents). As the word “false” would suggest, Lisak, and the other studies hinted at by Penn’s model guidelines, define the concept tightly: “The determination that a report of sexual assault is false can be made only if the evidence establishes that no crime was committed or attempted.” Or, in other words, if there’s a zero percent change that the accused student is guilty.

But a disciplinary panel is supposed to acquit if there’s anything less than a 49.99 percent chance that the student committed a rape. Assume for the sake of argument that Lisak is correct, and 6 percent of sexual assault cases are outright false. What percentage of claims are likely false? Are probably false? As with outright false claims, likely-false claims and probably-false claims should yield an acquittal, even under a preponderance-of-evidence threshold. Penn’s model guidelines don’t appear interested in sharing with panelists what percentage of cases are, in the aggregate, supposed to yield not-guilty findings. No explanation for this exclusion is given, but the result leaves the impression (at the very least) that a smaller percentage of students are not-culpable than actually are.

Yet a statistical sleight-of-hand isn’t the main problem with this “tip.” Imagine the (appropriate) outrage from civil libertarians if jurors in all murder trials received specialized training, including a “tip” that only a small percentage of murder claims are false. It seems likely, of course, that most people indicted for murder are, in fact, guilty. But such a statistic should (and must) be irrelevant to the individual juror, who is supposed to evaluate based on the specific evidence in a specific case. Including such “training” would send a message that prospective jurors should conclude that the individual defendant is guilty.

So it is with Penn’s training “tip” that false accusations are allegedly rare. The only purpose for such a “tip” would be to prompt panelists that they should presume the accuser is telling the truth, and return with a guilty finding.

Though Penn touts its guidelines as a model for other universities to follow, it’s not clear how many actually have done so. But the Penn training material, on top of the Stanford and Duke items, raises profound questions of what universities are actually doing when they claim a need to “train” panelists for sexual assault cases.