Sexual harassment at Berkeley, cont.

Berkeley’s miserable sequence of sexual harassment cases continues: this week it embarrassed our men’s basketball team, as we fired an assistant coach for abusing a reporter in an episode that included a kidnapping and threats to damage her career — the day we were accepted into the national tournament with a very flattering #4 seed.  So in a calendar year, we’ve lost a distinguished chemist’s service as Vice Chancellor for Research, a top astronomer has left the campus entirely, a law school dean has resigned, and an otherwise (apparently) effective coach was fired.  At last report, we have 26 more sexual harassment cases in the pipeline.

The president of the university and our chancellor and provost are all over this, of course, and in an email to everyone, the two campus officials “acknowledge that some recent decisions in cases of sexual misconduct have exacerbated these concerns, and…profoundly regret any and all errors of judgment on our part.” The president is setting up a panel to review all sanctions imposed on senior university leaders, presumably to make them tougher in cases like Choudhry’s; the chancellor is standing a similar body up for Berkeley.  Our state-mandated biennial two-hour on-line training, (which didn’t prevent the 30 current cases), is being revised, and spread out to cover all employees, including student employees. As the expanded reach does not include all students, who were the victims of the egregious Marcy case and many others, it is presumably directed at potential perpetrators and not victims;  a big mistake.

Unfortunately, I think the leadership response to date-mainly asserting concern, sitting us down for more training, and cranking up the punishment for perps-is poorly-designed, even though the last part, on the evidence of the Marcy and Choudhry cases, is overdue.
It’s a principle of good design to have a ‘model of the user’. A saxophone has keys and linkages because players’ hands aren’t big enough to reach across the distance physics requires to separate the holes; airport signs use a lot of graphic symbols because there’s no language all travelers can read.  What’s the implicit model of the user behind this institutional response?
I infer:

1. The ‘user’ is someone tempted at some time, now or in the future, to sexually abuse a student or coworker.
2. In at least some cases, he (female abusers are not unknown but rare) will offend because he doesn’t know that the contemplated behavior is abusive and forbidden. Hence the additional training.
3. In others, he compares the expected sanction for offense, combining the likelihood of being caught with the severity of the punishment, to the ‘pleasure’ he will get from the act; if the net expected utility is negative, he doesn’t offend. Hence the panel to inspect sanctions for defects, with power to scrap or rework them if they are too mild. It’s all about incentives; just tune them up and we can get whatever behavior we want.
4. Sexual abuse is distinctive and largely unrelated to other abuses of power that may afflict the campus, such as disrespecting or mistreating people of color, or of particular gender identities, or students and subordinates in general, in non-sexual ways.
If #4 is wrong, and the chatter on faculty listservs here indicates it may well be, a lot of synergistic gains will be left on the table, but I will leave those issues to another time.

Element 3 is a classic economic behavioral control model. What could go wrong? Taking the issue as we’re confronting it, and channelling the insights of Prof. Kleiman (the book, When Brute Force Fails, is right over there in the list on the right):  we have a naïve model of the relationship between threat of sanctions and behavior, and critically, we’re not recognizing that both sexual harassment and punishing harassers are costly. We don’t just want every sleazy act to be punished a lot: we want less sleazy behavior, and there’s some reason to believe we are stuck at an equilibrium of many costly offenses and a lot of costly punishment. It’s not a trivial matter to lose the value created by people like Choudhry, Fleming, Marcy, and yes, the coach, and it does nothing to alleviate the injury to their victims.

Element 3 has a long history, from public hanging of pickpockets (crowd events at which pickpockets thrived), to Admiral Byng “il est bon de temps en temps de tuer un amiral pour encourager les autres“,  to our own “three strikes” and ” throw away the key” incarceration policies.  As Kleiman explains, the likelihood and promptness of punishment are as important as the severity, and a variety of evidence indicates that a lot of sexual harassment is going on that is not reported formally in the first place, or once reported is covered up or dismissed by administrators.  Missing from the new program is a model of its other ‘users’,  victims and witnesses. How many marginal or mild cases aren’t handled at all, some because being a complainant in cases like these is itself a really awful experience, others because victims don’t want to blow up the lives and careers of offenders? How many offenders then go on to really bad behavior from which they could have been diverted (like Marcy, whose colleagues completely pooched their duty to straighten him out when he started on his scuzzy path)?  Atom-bomb sanctions have indirect as well as direct costs. At the least, we need some serious social science research on these matters (anyone at a great R1 university competent to do that stuff?).

I know that when my students (and peers) don’t cheat , and when my colleagues don’t steal lab supplies or plagiarize, it is almost always not because of fear and the calculation in element 3, but because they know it’s just wrong, and don’t want to think of themselves that way. But culture and social influence is almost completely missing from our new, improved program of browbeating and cranked-up sanctions.

How can we learn to behave ourselves better?  Formal rules and sanctions are important signals, but far from everything. From our parents, of course, and in school, but we can’t reverse the arrow of time.  Otherwise, I think popular arts and culture matter a lot, especially the cinema, fiction, and music (the last has a lot to answer lately for degrading women and legitimating a thuggish misogyny). Channels like these are carrying messages whether we want them to or not, but don’t figure at all in the harassment reduction program.

Making the offense less gratifying/tempting in the first place doesn’t figure either. Our required training is a two-hour on-line text lecture, with unedifying cheap clipart pictures and boxes to click on next to uninteresting questions; the whole thing is universally despised as a lawyers’ exercise in ass-covering.  The authors’ model of their users (us) is deeply ignorant.

The training also ignores the real lived experience of victims (all the consequences in that training are to harassers), offering instead a stupefying sequence of legalistic definitions of what “counts”, what the law says, and who must tell whom about it when it happens. It may have prevented zillions of abuses (is there any research behind it?), but it certainly didn’t work in our current thirty episodes.

We have no managerial apparatus to signal (not just assert; teach) that harassment is diagnostic of cowardice, misuse of power, and deficiency of character, nor that it obstructs and undermines exactly the research and learning purposes of the whole institution. The tone of the whole program is that harassment is bad because it’s illegal or against the rules, or could trigger civil litigation, but this is exactly backwards.   (There is an alternate training in the form of theater pieces with discussion afterwards that seems much more promising, and I will try to get to it the next time it comes around.)

Building a healthy culture at the intersection of power and gender sounds to me more like a quality assurance project than a criminal justice enterprise. We need to greatly change our relative investments in the two blades of this scissors, and pay much more attention to (i) undermining the ability of offenders to think of themselves as kindly mentors advancing their students’  and colleagues’  careers by pawing and shtupping them, (iii) strengthening and diffusing the view that sexual engagement with someone who doesn’t welcome it is not gratifying but repulsive, and (iii) enabling all of us to see people over whom we have power as, well, people. Of course offenders need to be seriously, and promptly, sanctioned, but remember those pickpockets…

I do not have a program of action to recommend.  That program must have a lot of moving parts and needs to be developed broadly across the institution; formal authority is not enough to make a healthy, supportive workplace and it cannot be the chancellor’s and provost’s job alone. Nor will they figure out how to do their part by being railed at, even when criticism is richly deserved. No-one will figure this out if we wear blinders that block out everything except the crime and punishment part of the enterprise.