20/20 mistakes

[Megan] (Enters, stage centre-right, looks around)  Wow, it sure is big in here, isn't it?

In my normal life, I'm an economics blogger.  And since it seems to be obligatory to post on the tragedy at Virginia Tech, I thought I'd start by talking about the decision errors I see in the commentary.  Daniel Drezner, one of my favourite bloggers, has been blogging on the coverage a bit:

According to the Washington Post, there were some warning signs from Cho Seung Hui before he killed more than 30 people at Virginia Tech: "Cho was an English major whose creative writing was so disturbing that he was referred to the school's counseling service, the Associated Press reported."

This fact prompted an e-mail from a colleague that raises a disturbing question: 

In 8 years, I've taught hundreds of students. 2 of them so alarmed me by their behavior, I contacted the Dean of Students office to see what could be done. The answer: nothing. The best I got was a half-baked assurance that voluntary counseling would be suggested to one of them (he was an undergraduate who had insisted on taking my graduate seminar, showed up and refused to leave on the first day of class, and then sent me increasingly enraged emails filled with expletives and threats to bring charges against me to the Dean of Students). I ended up having to have a staff member escort me to class in case the student showed up again. He didn't, fortunately. But I didn't follow up and I bet nobody else did, either.

When a faculty or staff member reports disturbing student activity, what is the appropriate response? Can any actions be mandatory? What feedback loops should be regularly instituted? I don't have any answers, but I do have an acute sense of vulnerability -- universities, esp. public ones, are wide open.

All professors have encountered or will encounter this problem in their careers -- the student who seems way too intense for their own good.

That said, I'm also concerned about overreaction. What happened at Blacksburg is a rare event, and red-flagging students just for being intense and weird can create problems as well.

This strikes me as a classic example of hindsight bias.  The teacher flagged two students; one of them happened to turn out to be a mass murderer.  But how many other college students have written things so creepy their teachers were worried about their sanity?  Say it's two per sizeable college in the United States; that's thousands of students who wrote really disturbing stuff, and didn't shoot anyone.  I recall my college creative writing classes, in which my most notable short story prompted several students and a very well-meaning teacher to approach me with offers of sympathy and help for my tortured issues about class, my father, and the Catholic Church.  I was almost unable to bring myself to tell all those lovely, helpful people, that I had no such issues; in the grand traditions of fiction writers everywhere, I had made it all up. 

All right, I confess; I did allow one particularly cute specimen to console me with a few drinks at the New Deck Tavern.  But that's a story for another day, and probably, another blog.

The point is that even if all mass-murderers did write scary prose, or make sweeping apocalyptic statements, or otherwise give some signal of their impending meltdown, the signal wouldn't do us any good, because mass murderers are really, really rare.  You'll have a thousand false positives for one false negative.  In hindsight, we can always pick out some clue to what was about to happen.  That doesn't mean that we can, or should, see those things beforehand.

Related is the criticism of administrators for sending students to class after the first murder, or of police for not locking the campus down immediately.  This is a classic problem with recriminations:  we tend to assume that the fact we had a bad outcome means we made a bad decision.  But in an uncertain world, this is ludicrous.  Good decision making concentrates on the most likely events, not the wild outliers.

The overwhelming majority of murders that take place on campus (or anywhere else) are not a prelude to a mass killing.  Should we really act as if they were, because it might prevent the 0.001% that are?  Shutting down campus is not free; if nothing else, it absorbs a huge number of police resources that could otherwise be used to track down the killer in the vast majority of cases where the killer is still at loose, armed, dangerous, and not planning to kill himself.  In this particular case, shutting down campus would have been the right answer.  But in 99.999% of cases, it would have been the wrong answer, and would have placed the public at greater risk, as well as producing mass hysteria on campus.  Castigating the administrators for getting it wrong, or rushing to enact legislation that ensures administrators do the wrong thing in most cases, is bad decision-making.  Not that this will prevent us from doing just that.