A sight quibble with my esteemed colleague Brian on the logic of abstention from voting. (I should note that I’m staying home as well, though because D.C. is so absurdly safe for Kerry, more than out of any general objection to voting.) After running through the familiar argument that the chances of your vote actually affecting the outcome are infinitessimal, and amid many other excellent points, he says:

No American is responsible for the voting behavior of our countrymen; so don’t worry for a moment about what would happen “if everyone thought that way”? (If you did control thousands of votes, the math might make it worth voting. But you don’t.) We each have only one vote, and only one November 2, 2004, in our precious lives.

This I’m not so sure about. This seems like one of a class of cases that, after the most memorable example, I call “harmless torturers cases.” The name comes from one of Derek Parfit’s thought experiments. He describes:

Scenario One: Ten thousand torutrers and ten thousand victims, with each torturer turning a dial to deliver an electric shock to his one victim, causing terrible agony.

Each torturer obviously acts very wrongly. But what about:

Scenario Two: Ten thousand torturers, each of whom delivers a tiny, imperceptible shock (one ten-thousanth the shock each delivers in Scenario One) to all ten thousand victims. The cumulative effect is that the victims suffer terrible agony, just as in Scenario One. Yet in this case, any one (or, indeed, any small group) of torturers refusing to turn their dials would have no effect on the felt experience of any victim.

It seems intuitive that each torturer in Scenario Two, while in some sense “harmless”, nevertheless still acts extremely wrongly. If that’s right, as Parfit suggests, our actions (or abstentions) from participation in some collective action can be subject to moral criticism whether or not they make a difference, in isolation. A less dramatic example would be littering, where your gum wrapper or cigarette butt aren’t likely to make a perceptible difference in the aesthetic quality of a particular place, but everyone’s decisions together do determine how nice it looks.

The notorious problem, of course, is identifying the relevant cases and classes. One well known attack on (an the most simplistic version of) the Categorical Imperative runs as follows: If everyone were a doctor (or a philosopher, or a journalist) then there would be nobody left to grow food, and we’d all die. So, QED, if the moral law requires us to act only on the maxim we could will universally, it’s immoral to be a doctor, philosopher, or journalist. But, of course, we don’t think division of labor is immoral—if there’s a universalization test for the morality of actions, then at least sometime the maxim has to be specified in terms of participation in systems in variable roles, rather than concrete rules everyone follows in the same way.

This, of course, is where it gets tricky. I think it’s OK to stay home in D.C. where Kerry’s safe, even though he wouldn’t be if the maxim of my action were universalized. I think I would be under an obligation to vote in a tighter state, even acknowledging that my lone vote isn’t going to be decisive. And I think that’s a reasonable enough position. As the doctor example shows, it can’t be morally wrong to take some account of how the likely actions of others are apt to influence an outcome, regardless of what you do. But because a large part of morality consists precisely in overcoming coordination problems, sometimes you’ve got to imagine you’re deciding, if not for everyone, then at least for a relatively similar class—maybe libertarians, in this instance. But I can’t think of a way that isn’t highly ad-hoc of deciding which is which. It’s clearly vain to hope that abstract theory can eliminate the need for that vague creature, the judgement call, in moral deliberation. Still, it’d be nice to have a more fleshed-out way to think about issues like this. I guess I’ll have to go back digging through my Kant stacks…