Last week, in an article entitled “Elephants in the Room,” Professor Thomas Dumm accused the Amherst College Special Oversight Committee on Sexual Misconduct of having deliberately overlooked scientific evidence that students athletes were disproportionately likely to commit sexual assault. Professor Dumm speculated that a possible reason for the failure of the committee to focus on athletes as perpetrators might be that “contributions to the College are correlated with the relative success of our teams, especially, though not exclusively, the football team.” What he was implying was that the committee was perfectly prepared to allow a rape-prone subculture (i.e. male athletes) to operate unimpeded just as long as the teams kept winning and the large checks kept pouring in.
As chair of the Special Oversight Committee, I will leave it to others to assess the credibility of Professor Dumm’s charge that we entered into a devil’s bargain in return for money, influence and favorable Division III rankings. I would like instead to examine the sweeping claims Professor Dumm makes about those students at Amherst who engage seriously in athletics, and in particularly to look critically at the “scientific evidence” he calls upon to support his position.
History teaches us that if one wants to demonize a particular group there are few better ways to go about it than to claim that its members are prone to predatory sexual violence. Over the centuries this charge was quite often made, on flimsy or no evidence, against Jews, African-Americans and gays, among other groups, and its purpose was always to drum up fear and justify extreme acts of “retribution.” Often it was a prelude to efforts to exclude or exile these groups entirely. Depressingly, the historical record also shows how quickly and easily people abandon conventional standards of proof when it comes to alleged sexual deviance among groups they consider different from themselves. Irregular legal procedures, shoddy use of evidence and sloppy or misconstrued science tend to accompany sexual scapegoating as maggots accompany rotting meat.
So has science, as Professor Dumm asserts, demonstrated a “clear relationships between single-sex teams and sexual violence?” As the Special Oversight Committee discovered when it surveyed the literature preparatory to writing its report, science has demonstrated nothing of the kind. A number of studies were done in the 1980s and 1990s that sought to measure college students’ and others’ tendency to believe “rape myths,” such as the myth that a good proportion of rape-charges were actually false or the myth that women routinely said “no” to sex when they really meant “yes.” Rape myth acceptance research is much bigger than the study of athletes, but at that time — now 20 and in some cases 30 years ago — some of the evidence from some schools suggested that male athletes were getting slightly higher “rape myth acceptance” scores than their non-athlete peers (Koss and Gaines, 1993). However, as a number of experts have pointed out, many of these early studies were flawed and inconsistent in terms of the definitions they used, how they chose and delimited their test subjects, how they presented their statistical findings and how they assessed the degree to which “rape myth acceptance” was predictive of a propensity to commit sexual assault (Hinck and Thomas, 1999; Buhi, 2005; Sawyer, Thompson and Chicorelli, 2002). In addition there were real questions about how representative they were, since almost all the “athlete/rape myth” studies relied heavily on large Division I and II universities and tended to lump all male athletes together into an undifferentiated mass. It is telling that it has proved impossible to replicate these older studies in the present day.
In the 21st century the monolithic view of “athlete” adopted by most of the earlier studies completely unraveled, largely thanks to a 2002 study led by Robin Sawyer of the Univ. of Maryland School of Public Health. That study did not set out to compare athletes with non-athletes (indeed there was no non-athlete control group), but instead focused on the often significant differences between athletes. So for example, the study showed that Division I athletes, both male and female, scored higher for rape myth acceptance than Division II athletes did (the study did not examine Division III athletes at all). It also showed differences between sports and demonstrated — perhaps surprisingly for those who consider sports teams training grounds for sexual assault — that first-year and sophomore male athletes scored higher for rape myth acceptance than junior and senior athletes did (in other words, the longer a man was on a team the less likely he was to accept rape myths). The study concluded that “like any group or community, student athletes should not be lumped together as a single entity with presumed identical attitudes and behaviors” (Sawyer, Thompson and Chicorelli, 2002).
Many recent studies of rape myth acceptance have followed this dictum and now focus on behaviors and personality traits rather than on whether or not one happens to be an athlete, or, indeed, a member of any other student group. Those recent or relatively recent studies that do consider participation in athletics as a variable have been unable to show any difference at all between athletes and non-athletes, either in rape myth acceptance or in sexually aggressive behaviors. Locke and Mahalik (2005) tested the hypothesis that athletes believed more strongly in rape myths than other groups and committed more sexual assault. Simultaneously they tested the relationship between rape myth acceptance, sexual aggression and a series of multidimensional markers of masculinity. They succeeded in showing that men who drank to excess, avoided emotional commitments, engaged in high-risk behaviors and disliked homosexuals were more likely than other men to report behavior on surveys that sounded like sexual assault to researchers. However, in and of itself being an athlete made no difference at all. These results are pretty typical of what one finds today. In addition, they match the findings of Meegan Mercurio whose BA thesis from Providence College Professor that Dumm, somewhat unwisely, cites in his article as “proof” that athletes as a group are especially prone to sexual violence. Mercurio did indeed try to prove that hypothesis, though only for Division I athletes. But if Professor Dumm had taken the time to read further than the first few pages he would have seen that she does not succeed. I quote from her conclusion: “This research addressed the problem: Does playing as a Division I student athlete affect dating violence attitudes and are student athletes more likely to be perpetrators of violence? The hypothesis inferred that athletic participation may have an effect on dating violence. Instead the null hypothesis was proved and the findings theoretically proved that student athletes are not more likely to be perpetrators of violence, but in fact are less likely to be perpetrators of violence.” (Mercurio 2010, p. 25).
Along with charging that our committee neglected the science, Professor Dumm asserts that we did not take seriously the accounts of rape survivors at Amherst. The real story, clearly stated in the report, is that we did interview rape survivors and we heard via email, letter and phone from many others, including a number who are now graduated. We also surveyed every sexual assault complaint made in the last 10 years, the vast majority of which never advanced to the Hearing Board. In those accounts we did find assaults perpetrated by athletes. We also found evidence that teams had sought to silence victims. What we did not find was evidence that either athletes or particular teams were standouts in these respects. Athletes were not disproportionately more likely to be perpetrators than were non-athletes. Most important of all we found absolutely identical patterns of silencing and, at times, retaliation against victims from student organizations that had no connection whatsoever to athletics — and we found them repeatedly.
A lot of people like the idea of fobbing off the problem of sexual assault onto some other group that is not them. One can see why. If the problem can be traced to, say, the football team, why should anyone who is not on that team bother to endorse and work for sexual respect among their own groups or networks? Or encourage more comprehensive reporting? Or try to adopt or get others to adopt less risky drinking habits? Our committee believed and still believes that eliminating sexual assault at Amherst College demands a community response. The notion that the problem is all or mostly to be laid at the door of athletes is at variance with the evidence our committee painstakingly gathered about the past history of assault, and it is not supported by scientific research. If one really wants to eliminate sexual violence and enlist the entire community in doing so, blaming athletes is a truly counterproductive way to start.
And that brings me to opportunism. What is Professor Dumm’s purpose in endorsing such a patently futile strategy for eliminating sexual assault, not to mention one that unfairly singles out a particular subgroup? It is no secret that Professor Dumm is a critic of varsity athletics at Amherst College, especially football. It appears that he sees in the sexual assault issue an opportunity to discredit the athletic program, and he hopes that if enough people can be persuaded to believe athletes are natural rapists they will rise up in outrage and get rid of, or at least severely limit the program. He has such an animus against athletes and is so intent on his goal that he doesn’t even see the injustice of tarring an entire demographic with the name of sexual predator. Furthermore, he also has no compunctions about enlisting science that is out of date, anything but clear and no longer able to be replicated in order to push his views. It is anyone’s right to oppose varsity athletics. It is not, however, all right to do so in such an irresponsible way.
The full citations for the scientific studies mentioned in this article can be found at http://www3.amherst.edu/~mrhunt/