Why I Refuse to Refuse to Vote for Obama

My friend and Atlantic colleague Conor Friedersdorf has struck a chord with his piece "Why I Refuse to Vote for Barack Obama," in which he explains why he won't support the candidate he supported in 2008 even though he doesn't want his Republican opponent to win. (If he votes, he says, it will be for Gary Johnson.) At last check Conor's piece had more than 160,000 Facebook recommends -- a number that's up there in Romney 47 percent video territory.

I can see the appeal of Conor's argument. The Obama policies he finds unacceptable -- such as drone strikes that kill innocents, the assassination of American citizens abroad without due process of law, and other assaults on civil liberties -- are policies I've been criticizing for a long time. And Conor's principled stand on these "dealbreakers," as he calls them, is inspiring. To say that you'd rather vote for someone who can't win than for a candidate with odious values is one of those stirring, consequences-be-damned pronouncements that usually win me over when I hear them in movies. But this isn't a movie, so I have a hard time ignoring the consequences of (implicitly) encouraging would-be Obama supporters to nullify their votes and thereby increase the chances that Mitt Romney will be our next president.

"Consequences" is ultimately the word that divides me and Friedersdorf. I'm what philosophers call a "consequentialist," someone who judges the rightness and wrongness of behaviors by their consequences. Roughly speaking, you could say I'm in that subset of consequentialists known as utilitarians -- i.e., people who think that what's good is what maximizes overall human welfare. So if not voting for Obama only increases the chances of victory for the candidate I consider worse for America and the world than Obama, I'm not going to do it. (Unless you can show me that there are counterbalancing long-term consequences of a protest vote -- e.g., that this will strike the fear of God into the 2016 Democratic candidate.)

Friedersdorf says he respects the argument of utilitarians who take the position I've just outlined -- supporting Obama because he's the "lesser of two evils." But, in a follow-up piece, he says he wishes more of these utilitarians would at least confront a thought experiment that might get them to question whether they're really thoroughgoing utilitarians -- whether they don't in fact believe that some values are so important that they should be honored regardless of consequence; or, as Friedersdorf puts it, whether these professed utilitarians don't in fact have "deal breakers." Here's Friedersdorf's thought experiment:

Suppose that President Obama was surreptitiously videotaped in private, and found to be "repeatedly using anti-Hispanic slurs to refer to Mexican Americans, musing that his personal dislike of Mexicans motivated the record number that he deported, and noting that while he'd never transgress against the law by unlawfully targeting Mexican Americans, he sure does hate them." Wouldn't that be a dealbreaker for you, asked Friedersdorf? In other words, wouldn't some of us professed consequentialists, if pressed far enough, admit that we're not really consequentialists but actually hold some values so dear that their violation would trump consequentialist considerations?

I promise to answer Friedersdorf's thought-experiment question in a few paragraphs, but first I'll ask him to answer mine. Here it is:

Suppose that President Obama was what he in fact is: a drone-striking, civil-liberties disregarding president. Suppose you could be pretty sure (as I think you can, though Friedersdorf disagrees) that Mitt Romney's policies on drones and civil liberties wouldn't be any better. And suppose that -- through the magical powers that are permitted in thought experiments -- you knew that if Romney were elected he would start a needless war that would kill 100,000 people, and would also inflame the international arena in ways that led America (through the irrationality that has become its hallmark) to deploy more drone strikes, and disregard civil liberties on an even larger scale.

Of course, for purposes of the thought experiment, I could crank the stakes up even higher: Suppose Romney would start a war that killed a million people. Or 10 million. Or suppose he would go nuts and nuke half the world or the whole world. Is there any point at which you'd concede that casting a vote that increases the chances of a Romney victory is the wrong thing to do? If you'd rather see half the human species extinguished than vote for someone with a low regard for civil liberties and a high regard for drone strikes, just say so. But if you wouldn't, then it seems to me you're admitting that, actually, you've got a bit of consequentialist in you -- that your "dealbreakers" aren't really absolute, unconditional dealbreakers.

By the way, the reason I started out with relatively small stakes -- only 100,000 dead -- before moving up to a million and beyond is that when the number is 100,000, this isn't a mere thought experiment. In 2000, a bunch of voters on the left decided that Al Gore's likely policies included some "dealbreakers," so they voted for Ralph Nader. That's why George Bush became president. Bush then started a war that Gore probably wouldn't have started, and as a result at least 100,000 people died, and the international arena was inflamed in a way that gave his successor a rationale to (unwisely, but fairly predictably) conduct lots of drone strikes and disregard civil liberties. So my "thought experiment" is very much a real-world scenario -- way more plausible than the average philosophical thought experiment.

Obviously, there is no way of knowing for sure what the consequences of a Romney or Obama presidency will be. But I'm convinced (for reasons I will spell out in a later piece) that Romney is more likely to get us involved in a war with Iran than Obama is. And I don't think the fact that I'm just talking about a likelihood, not a certainty, invalidates my argument. (Obviously, we're never sure what any president will do, so if probabilistic assessment isn't a valid basis for voting behavior, I guess I shouldn't vote at all.)

So here's what I'd ask of Friedersdorf: Either (1) say that you'd rather see half the human population die than cast an unprincipled vote for a drone striking civil liberties disregarder; or (2) say that, actually, yes, the consequences of our choices are part of the moral calculus that should inform them. If my thought experiment is the inescapable trap that I hope it is, you have to do one or the other.

Now back to Friedersdorf's thought experiment: Would I vote for a closet racist if the alternative was to vote for someone who, in practical terms, would be even worse? Well, I don't live in a swing state, and, anyway, when was the last time a swing state's electoral votes were decided by a single vote? So if I were faced with that choice I'd probably stay home on election day and use the guaranteed insignificance of my vote as an excuse.

But if we assume -- as I think we should for purposes of these thought experiments -- that my vote would actually make a difference, then, yes, I'd vote for a closet racist rather than vote for someone who, in practical terms, was even worse (certainly including someone who wasn't racist but who out of political expediency would support policies more racist than the closet racist's policies).

Now over to Friedersdorf.