Occasionally I will suggest to friends and acquaintances (from somewhat puckish motives I have to admit) that they shouldn’t vote, or at the very least that it doesn’t matter whether they vote or not, or who they vote for, especially in large (national or state-wide) elections.
My reasoning is fairly simple: whether you vote or not will make no difference to the outcome of the election. Suppose you vote for candidate X and she receives 5,497,385 votes while candidate Y receives 5,326,792 votes. Now suppose you hadn’t voted at all. Candidate X would’ve won by 170,592 votes instead of 170,593 votes. Wow! Your vote really counts! (For simplicity’s sake, I’ll ignore the fact that in the U.S. there are no truly national elections. In the case of presidential races we all vote in statewide elections, which then determine which candidate the state’s electors vote for. But the point stands.)
Even in those (extremely rare) cases such as in the presidential election in 2000 where the margin is relatively small, it still never comes within spitting distance of a single vote. And, as we saw, as the margin of victory approaches a small number of votes, the chance that the election will be decided in a courtroom rather than the voting booth increases significantly.
But wait, some will say (if there was anyone reading this), what if everyone acted like that? Then no one would vote and where would we be??
The intuition underlying this protest is that we ought to act in such a way (at least, in so far as our moral duties are concerned) that we would want anyone (or everyone) to act in circumstances that were relevantly similar. Philosophers call this the “universalizability principle” (hereafter, UP). As Kant put it, we should act in such a way that the maxim guiding our action can consistently be willed as a universal law.
If this is right, then I should refrain from voting only if I am willing to say that everyone else should do the same. Since I’m not willing to say this (because it would spell the end of democratic governance, which I regard as a good thing), I should not refrain from voting, nor should I advocate that others refrain from voting. Q.E.D., right?
Well, I’m hesitant to embrace this conclusion, because it seems to me that we can use similar reasoning to proscribe other kinds of voting behavior – kinds of behavior that most people would, intuitively, regard as perfectly acceptable, even commendable.
Take, for instance, third-party voting. Many people vote for third-party candidates, not because they actually want their candidate to win, but because they want to “send a message” of dissatisfaction with one or both of the major parties. Maybe it’s because the major parties are not addressing some issue they deem important, or because they want one of the major parties to move further to the right or left, or whatever. I imagine there are people who voted for Ralph Nader or Harry Browne in 2000, not because they wanted either of those esteemed gentlemen to actually become president, but because they wanted to send a message of protest to the Democrats and/or Republicans (this is not to deny that there were people who voted for Nader or Browne who did want them to become president; I’m just stipulating that there was some sub-set of third party voters who were voting primarily to send a message, not because they wanted their guy to win).
But, if we apply UP, it looks like these third-party voters are behaving irrationally or immorally, just like someone who refrains from voting altogether. After all, by hypothesis, these folks didn’t want the candidate they voted for to win. But, according to UP, one should always act in such a way that one would will everyone to act. So, in voting for a candidate they didn’t actually want to win, they are running afoul of this principle, since if everyone voted as they did, their candidate would have, in fact, won.
Now, that might not seem so bad, since you can always say that, yeah, you should only vote for the guy you really want to win. But it gets worse. It looks like we can get an even more counterintuitive result when we apply UP to voting for someone whom we actually want to win. Suppose I really and truly in my heart of hearts want John Kerry to win the upcoming election and so I vote accordingly. Now, according to UP, I should be willing to say that everyone should vote for John Kerry, since that’s what I think is best all things considered.
But I’m not so sure I’m willing to say this. Even if I want John Kerry to win, would I want everyone to vote for him? Wouldn’t that create a situation in which the winner had an overwhelming (unanimous!) mandate, and couldn’t that lead to him running roughshod over all opposition? The problem with this is that I think a vigorous opposition is essential to a healthy democracy, so I want there to be a significant amount of support for the opposition party in any given election. But this means that I can’t will that everyone act as I do in voting for John Kerry, even if I want Kerry to win. So that means that I shouldn’t vote for Kerry, even if I want him to win!*
The upshot of this is that it looks like the consistent application of UP rules out not only abstaining from voting, but also voting for any candidate unless you want him or her to win with 100% of the vote. This suggests, to me anyway, that UP may not be the self-evident principle it has often been taken to be, since it seems to lead to some extremely counter-intuitive conclusions. But, if we reject UP, then at least one argument against non-voting appears to be without foundation.
———————-
*Now, there may be fanatical party loyalists who think that a 100% vote tally for their guy would be just dandy, but I take it as nearly self-evident that, other things being equal, a one-party state is a bad thing.