Rumor has it that there’s a presidential election scheduled in the U.S. this fall, which raises the perennial ethical question: Is there a duty to vote? Harry Brighouse provides some excellent arguments for there not being such a duty, but here I’ll lay out a few pros and cons and invite people to weigh in on whether there is such a duty.
I gather it’s uncontroversial that, except in very unusual circumstances, it’s always morally permissible to vote; no moral duties are violated by voting. So what are some arguments that there is such a duty?
If you support democracy, there’s a duty to participate in the democratic process by voting. This looks like an appeal to integrity. If you believe (as probably most sensible people do) that democracy is the most defensible form of government, then you have a duty to ensure that the democracy you live in functions properly, which can only take place if you vote. So voting is a vote for democracy.
Yet voting is not only not the only way to participate in the democratic process. Running for office, holding rallies and demonstrations, writing letters to the editor, educating others about the issues, rallying others to vote: These are just a few ways in which we can participate without voting — and arguably, these might have more impact on the process than one’s vote. Very few elections are actually decided by margins small enough to make one person’s vote meaningful.
A close cousin to this argument is ….
Failing to vote is an indication of apathy or indifference. This one has a virtue-theoretical ring to it. If you don’t vote, you are signaling that you’re indifferent to the state of your community. But people should care about the state of their communities and show that concern by voting.
But again, failing to vote doesn’t tell us that the non-voter is indifferent to her community. She may show her concern in other ways and/or she might conclude that her voting is not likely to make a substantive impact on her community. Furthermore not voting could be taken as the non-voters’ signal about the health of a democracy. Perhaps the candidates are so poor, the process so corrupted, etc. that by not voting, the non-voter expresses her opposition to the pseudo-democratic status quo.
Non-voters have no basis for subsequent complaint. This seems like an appeal to cooperative burdens. Voting is a burden, so by not voting, non-voters show themselves unwilling to bear these burdens. But only those who bear the burdens of a cooperative scheme should enjoy its benefits. Although it’s effectively impossible to keep non-voters from enjoying the benefits of democracy, there must be something that non-voters forego as a consequence of not bearing their share of the burden. What they forego is the right to hold one’s government accountable for its failures and to complain about these failures.
In response, voting doesn’t seem to be all that burdensome. Indeed, most jurisdictions are trying to make it increasingly convenient (by establishing vote by mail, etc.). More directly, however, it’s not obvious to me that a person who doesn’t bear the modest burdens of voting thereby relinquishes her right to legitimately complain about the cooperative scheme of which voting is a part. For one thing, the non-voter may well bear other burdens imposed by this scheme (paying her taxes, obeying the law, etc.). Moreover, this arguments seems weak when, by not bearing the burden, the voter is not materially undermining the cooperative scheme, and as I mentioned above, not voting rarely has much impact on the health of one’s democracy or the justness of one’s society. Not voting looks like a harmless instance of non-contribution to a cooperative democratic scheme.
Good old Kantian universalization. This is an old standby: You cannot coherently will that a maxim of not voting be universalized, for then no one would vote and democracy would collapse.
Obviously, the way around this argument is to build into the maxim various conditions: if one’s vote will not materially affect the outcome, etc. A more narrowly tailored maxim probably could be universalized. In a more substantive Kantian vein, I think many Kantians would agree that there is a duty to establish and maintain just democratic institutions. But here I suspect this operates more like an imperfect duty rather than a perfect duty: The duty prescribes an end which we may realize in a variety of ways, only one of which is voting.
So, in short, I don’t see any knockdown argument for a general moral duty to vote (some of these arguments might demonstrate a restricted or conditional duty). But perhaps these arguments are better than I’m giving credit for or there are other stronger arguments that I have failed to consider.
25 Replies to “Is there a duty to vote?”
Perhaps this goes without saying (whatever, I’ll say it anyway), it depends upon who you’re going to vote for. We have no duty to vote for the bad guy/girl.
I think that whether we have this duty depends upon the facts. If candidate x is going to be very bad, and candidate y is going to be okay, then we should vote for y. Typically, x and y are relevantly similar (like Obama and Hillary and McCain), and when that’s true, we have no duty. When they are very different (JFK and Hitler, say), we do.
Basically, when the consequences are significant even the small probability of being the vote that makes a difference generates a duty. When these consequences are not significant, the voting hassle can outweigh them.
Is it really “uncontroversial” that voting is usually permissible? I thought Harry’s point was precisely that it’s immoral (“thuggery”) to cast a selfish or prejudiced vote.
Another quote: “When you vote, you have a very stringent obligation to deliberate responsibly about the effects of your vote, and about whether those effects are morally justifiable or not.” Though perhaps we should interpret this as saying not that the vote was immoral, but that the lack of responsible deliberation (given that one voted) was.
Judging by the outcome of previous U.S. elections, it sure looks like a lot of people did the wrong thing in the voting booth.
A more interesting question is if an individual ought to publicly decree that “one ought to vote”. The answer (to me) seems to be yes. I think this really underlays the ‘you should vote’ bit – an individual not voting is one thing, but once not voting becomes a norm, a Democracy WILL collapse.
ie – it is morally permissible for an individual to not vote, but morally impermissible for a population to not vote, and I think if you mirror the arguments you presented in terms of populations, the outcome will be such.
Moreover, if you accept that a population ought to vote and that an individual’s acts influence the ‘mood’ of a population and a population’s ‘mood’ influences other individuals, then it seems amoral to not vote.
I would imagine that the clearest way to think about this to say that there is an obligation to be politically active, and that in some circumstances but not all, this implies an obligation to vote. In which circumstances it has this implication will depend on certain empirical facts that philosophers qua philosophers probably have little expertise on.
That seems to capture most of the points above.
wouldn’t it be enough if almost everyone had almost always good reasons to vote? That would be more than mere permissibility. I would actually be quite worried about a universal duty to vote. As you say, that might imply a duty vote even in elections that were run in an objectionable way or in ones where all the parties have immoral policies.
I wasn’t sure about this though:
“Very few elections are actually decided by margins small enough to make one person’s vote meaningful.”
I don’t think that vote is a deciding one should be seen as a criterion for its meaningfulness. There are a variety of meaningful things a vote can express even if
it’s not a deciding one. It can express one’s commitment as a citizen to the state and its democratic system, it can express one’s disapproval of the policies and aspirations of some parties and approval and commitment to values and goals of other parties. It seems to me that these are the kind of things one has good reason to express even if it failing to do so might not be morally wrong.
Also, coming to think of it, you might think that there is an obligation to vote based on carrying out the role of a citizen. This corresponds to other role-based obligations such as a doctor might have to save lives. Taking the citizen-role on gives one many rights and privileges but you might think that it comes also with obligations such as the one to vote.
I’d like to lay to rest the idea that an individual voter would not influence an election. Individual voters can have a huge impact on local elections, where the voting pool is small. It does follow through to larger situations, as well. I give you 2 nation-wide Presidential elections. First, in 1960, the national popular vote was so close that, if Nixon had one more voter in each precinct across the country, he would have won the popular vote. In 2000 we see an example that is a bit more relevant to our electoral college system. When it came down to it, after all the lawsuits, the Miami Herald got a chance to carefully count all the controversial votes in Florida. They were able to conclude that Bush actually won the state by a mere 600 votes.
600 votes over a state as large as Florida is a small enough margin that I would say individual votes certainly mattered. The idea that individual votes don’t matter is anathema to the entire democratic process, whether it is moral or not.
Cormac, I don’t understand what you mean by saying that each individual vote mattered in Florida in 2000. Every person who voted could sincerely and truly say, “Had I gone to the movies instead it would have made no difference.” And everyone who didn’t vote can truly comfort himself by saying “It wouldn’t have made any difference if I had voted.” So no individual vote mattered, in that sense.
I’m surprised nobody has yet suggested fairness (in the Hart/Rawls sense) as a reason to vote. My chance of causing Obama to win in the primaries is very, very small, but if I don’t do it the other Obama supporters who did vote have a legitimate complaint against me: I’m being a free rider.
Here’s an argument for staying home and rewatching episodes from the underrated (even for Wire fans) Season 2 of the Wire rather than voting in a presidential election, no matter how politically active you are otherwise.
(1) The probability that your vote will make a difference in the election–and that it will be a positive difference–is either vanishingly small or zero (in states where the winner is a foregone conclusion).
(2) The probability that you will accidentally hit a child (or anyone) with your car on your way to the voting booth is very slight, but far greater than the chances that your vote will make a positive difference even in a swing state. (If it’s Hillary vs. McCain, the chances of hitting someone are much greater because you’ll probably be drunk.)
(3) Therefore it’s permissible and maybe obligatory to stay home and watch Season 2 of The Wire.
That seems to work for me, as long as you keep this line of reasoning to yourself (so that enough people aren’t convinced to stay home for the outcome to be affected).
I think it’s Mostly an issue of Kantian universalization as a solution to free riding particularly within group (ie Obama voters or McCain voters) free riding.
However having a very narrowly tailored maxim might mean that a lot of people would still percieved others as free riding.
Every person who voted could sincerely and truly say, “Had I gone to the movies instead it would have made no difference.” But, in Guam, few if any could have said that knowingly. So, they shouldn’t have said it. And maybe they shouldn’t have reasoned from it.
They could have said it truly (which means that ‘no individual vote mattered’). They could have said it with a credence extremely (and justifiedly) close to 1.
I’m not sure if Tamler Sommers is joking. His argument looks like a non sequitur. How does (3) follow from (1) and (2)?
I was joking about the ‘obligatory’ part but not the permissibility. I’ll make the argument valid (more or less):
(1) The probability that your vote will make a positive difference in the election is either effectively zero or zero (in states where the winner is a foregone conclusion. (This is even true in Guam, I’d argue, since the nomination will not stand or fall by those results).
(2) The probability that you will accidentally kill a child (or anyone) with your car on your way to the polling station is very slight, but it is not effective zero or zero. It is greater by many orders of magnitude than the chances that your vote will make a positive difference in an election.
(3) It’s permissible to refrain from doing something that has the slight chance of killing a child in order to perform an act that has effectively no chance, or no chance at all, of making a positive difference in an election.
(4) Thereore, it’s permissible to stay home and refrain from voting.
Of course, the argument assumes that you live far enough from the polling station that you need to drive.
As an aside, I do think that in general it’s morally obligatory to watch the seasons of the Wire more than once (although not necessarily on election day). To paraphrase Proust, one does not watch The Wire. One rewatches it.
The post was written half in jest. I thought it was funny that as this discussion was going on, the vote in Guam came down to 7. Given the knowledge account of assertion, few Guamanians could be warranted in saying ‘Had I gone to the movies, it wouldn’t have made a difference’ since they so easily could have been mistaken. If you think knowledge is necessary for treating something as a premise in practical deliberation, few would have been right to reason from this. And, if anyone was driving a van with seven other Obama supporters, _they_ would have been wrong to think ‘Had I gone to the movies, it wouldn’t have made a difference’.
First of all, Tamler is more right than he may realize that ‘even in Guam’ the chance of a given voter making an appreciable difference to the nomination is very near zero. If (say) thirty more people had voted for Obama, that would have made no difference at all in the distribution of delegates: the delegates were in fact split 50-50, and would have been the same for small changes in the voting.
I don’t understand what “effectively zero” means in the argument, though. The chance of one’s vote making a difference is incredibly small. But if it were zero, then (it seems to me) there would plainly be no obligation to vote, whereas if it is very small there is an interesting open issue.
Maybe I can make my point clearer like this: the chance of my vote making a difference to the outcome of the election is tiny. But the effect I would have were I to have an effect is enormous. It is much, much, much more significant (I say) than accidentally killing someone on my drive to the polling place.
Just to interject a historical note, here is Mill on the subject:
“If your vote could affect only yourself . . . it would still be a question whether unless those others govern you with perfect justice, you are morally entitled to forego [sic] the right and power which a vote would give you to force them to do justice, and thereby become themselves better moral creatures. But it is not the fact that the possession of a vote would enable you only to protect yourself. Every citizen possessed of a vote is possessed of a means of protecting those who cannot vote, such as infants, the sick, idiots &c. as well as of a means of helping others who can vote to do good in every conceivable way in which just and provident legislation can affect human happiness. I am deeply persuaded that nothing but a most regrettable absence of thought on this subject can account for or even partially excuse, for wholly excuse it cannot, the very common neglect of the power of voting which prevails among gentlemen and educated persons. I am certain that a time will come when it will be felt that a man, and I need not add a woman too, because any rational creature, is committing a most gross dereliction of duty when he habitually neglects to make use of this power conscientiously and at any cost of labour to himself.”
Mill doesn’t take the argument back to first principles here, but I’ve argued that he takes the obligation to vote to be part of a more general obligation to participate actively and disinterestedly in public affairs, an obligation for which he takes there to be what we would call a rule-utilitarian basis. So I would add a rule-utilitarian argument to Michael’s list of possibilities: the set of moral rules with the highest acceptance utility would include a rule that would require citizens of democratic nations to vote, at least under the circumstances that exist in most Western democracies now.
When you say that if [the chances of deciding the election] were zero, then there would plainly be no obligation to vote, you’re conceding that only people in contested swing states are even candidates for having the obligation, right? I’m not sure everyone would agree with that (although I do).
As for your second point about the difference in the magnitude of the effects, I guess I’d say two things. First, you have to compare the difference in the effects with the difference in probabilities. If you’re one billion times more likely to hit a small child than you are to decide the election with your vote, then it still seems permissible to stay home. (I have no clue what the actual number is, but it wouldn’t surprise me if it were larger than that. Does anyone have any idea?)
Add to that that given the uncertainty involving the candidates and what they’ll be facing, none of us likely has better than a 55-60% chance of knowing which candidate will be the better one. This means that even in the extremely unlikely event that you did decide the election with your vote, there’s still a 40% chance (at least) that you’d have a harmful effect on the world rather than a good one.
Oh and I did realize how right I was about Guam but thanks for the additional support!
Jamie’s final point is more or less the crux of Derek Parfit’s argument in favor of voting in the 26th section of Reasons and Persons. Even if the chance of swinging the presidential election is 1 in a hundred million, the difference between the candidates is often worth a trillion dollars. So a vote for the better candidate has an expected value of $10,000. This is greater than any of the costs of voting. If you don’t live in a Presidential swing state, you can reapply the argument on the level of lower offices. The payoffs may fall, but the odds get better.
I don’t know if this generates an obligation to vote — I don’t know how a consequentialist should determine where the obligatory ends and the supererogatory begins. But it’s certainly a good reason to vote.
Just days ago in that very same election campaign mentioned Obama won by 7 votes. So, yes, quite few elections are decided by a marginal ammount of votes, but I for one I am not willing to take the chance that it will be some very-very important election.
Secondly: in countries where you have the legal duty to vote you also have the option of not choosing any of the candidates/proposals. So if you are indifferent towards the choices presented in an election but you have a actual real opportunity to express that indifference, then I see no reason not to do it. So in the case of institutionalized impartialism there is no reason for not voting.
OK, let’s try this. I think democracy is a far better form of government that oligarchy. You can debate that, if you like, but I think I’m on reasonably safe ground here.
If democracy is a better form of government than oligarchy, I have an obligation to vote because my vote may keep my democracy from becoming an oligarchy. I present a sort of Theseus’ Ship problem here. I see our democracy with 100 million voters, or so, making decisions for 300 million citizens, or so. Remove one voter and you still have a democracy. Remove a million voters and you still have a democracy. Remove 99,999,000 voters and I would say you definitely have an oligarchy. Where is that line? Of course, you can’t determine that line, unless there is some mathematical definition of an oligarchy out there that I’ve never seen. I believe I have a duty to NOT be the guy who tips us over toward oligarchy and away from the superior form of government, democracy.
You might say that there are plenty of other people out there who will vote and that will keep us away from anything that could be called an oligarchy. In so saying, you are taking this duty that we as a group have to be a democracy rather than an oligarchy and shirking your part onto whatever part of the rest will take it. Try this, a cardiologist is sitting in a lecture hall filled with other cardiologists at a cardiology convention. His colleague next to him develops chest pains indicative of a cardiac event. He stands idly by and does nothing, assuming that the cardiologist on the other side of the victim, or the one to the front or behind will step in and help. He says, “There are plenty of others to get the job done satisfactorily, and so I do not need to participate in the solution.” Meanwhile, 2 of the other 3 have the same attitude and the last is left trying to do the job on his own when help would almost certainly lead to a better outcome. ALL FOUR of those doctors, right, left, front and back have a duty to pitch in and offer their expertise. No one is particularly special in the group, and maybe 3 would be enough, but the fourth still has the duty to not sit there with his thumb up his rear and ignore the problem.
I don’t see a notable difference between the fourth doctor and the 100 millionth voter. We all have a duty to preserve the superior form of government and not voting shirks that duty.
I think that theirs a bigger purpose for voting other than just to change the ultimate decision of an election. Voting’s purpose is also to maintain the current democratic system. Like it was said before, if people don’t vote, you might have a bit of a problem maintaining a democracy.
Yes, maybe the chances of you running over a child is greater than the entire collapse of the US democracy. However, basically like Cormac said if you say that it is permissible for an individual not to vote, than you would also have to say that it is permissible for a society as a whole not to vote, because your conclusion that is okay for someone not to vote, would apply to every individual with in that society.
A few points:
a) I want to emphasize Tamler’s point that we may not know who the best candidate is. Assume there are two options, and one of them (call it “the right choice”) is significantly better than the other (which we can call “the wrong choice”). Most of the foregoing discussion seems to assume that, if you vote, you will vote for the right choice. But for most people, I think it is about equally likely that they would vote for the wrong choice.
b) The same argument from Parfit that shows that voting for the right choice can have significant expected benefit also shows that voting for the wrong choice has significant expected harms. Voting for the wrong choice may also be a rights-violation, because (as the Brighouse article concedes) voting is an attempt to bring about coercion. If the wrong choice would be a rights-violation, but mere failure to support the right choice would not, then that gives the reason against voting greater weight than the reason in favor of voting. I take it that a x% chance of imposing a rights-violating harm on others outweighs an x% chance of creating a benefit of comparable size.
c) Because of (a) and (b), I think it is wrong to vote, in most elections, *unless* one has very strong justification for believing that one is voting for the right choice. But I think almost no one has such justification.
d) Perhaps individuals have an obligation to acquire such justification first, and then vote. But I think that the costs of acquiring such justifcation are so great as to make this an unreasonable demand. The obligation probably would not even be justified according to utilitarianism. In my view, figuring out the answer to a single typical political issue would require several days to several months of intensive research. The costs of figuring out *all* the issues in play in a typical election are prohibitive.
One would disagree with point (d) if one thinks that in most elections, it is pretty obvious what the right choice is. Many people seem to think that. But I don’t know how they acquire their confidence.
More evidence for the consequentialist case against voting.
I am rethinking one of my earlier claims, however:
“None of us likely has better than a 55-60% chance of knowing which candidate will be the better one.”
After this past couple of weeks, I might want to bump that up to around 65%…
Just found this. Here’s a related thread on a paper of mine where I argue that citizens often have a moral obligation not to vote badly, but should abstain instead, though they have no obligation to vote.
Comments are closed.