It’s become a cliché of philosophical commentary on the ethics of voting that voting is prima facie irrational. If you vote as an individual, your vote represents a mere 1/n of the totality of the election, where n represents the total number of voters in the election. If we consider, say, national elections in the United States, then n will be a very large number. Since it is, 1/n represents a tiny number. The tiny number represents a given individual’s contribution to the overall electoral result, the implication being that each individual’s contribution is, qua individual, de minimis. It’s so small that it seems irrational to bother with it. The effort you put in is not worth the result that comes out.
Indeed, if you drive to the polls, the risk you run of an accident on your way to the polls (it might be thought) nullifies whatever good you as an individual could do by voting, regardless of who you vote for. The only exception is the case in which your vote is the tie-breaker in some momentous election. But the chances of that happening are close to zero, like the chances of winning a lottery. If it’s irrational to play the lottery (and it is), it’s equally irrational to vote in order to break the tie that might take place.
All of that is true as long as you grant all of the “ifs” embedded in the last two paragraphs: if you vote as an individual; if you vote in a large-scale election; if you drive to the polls; and so on. But if you replace those ifs with their negations, the “voting is prima facie irrational” cliché is far less obvious than it might have seemed. In fact, it seems false. Suppose you vote by mail as part of a bloc in a small, local election. Then what?
Then it seems to me that the “vote doesn’t matter” problem fails to arise. Let’s concretize this a bit. There are, in the United States, states with very small municipalities where the total voting population numbers in the low thousands, or even the hundreds or tens. These municipalities hold elections for small-scale but very hotly contested offices, often connected with the K-12 education system (e.g., positions on the local school board), or with local law enforcement (e.g., sheriff, district attorney), or with the judiciary (e.g., municipal judge), or with important administrative positions within local government (e.g., registrar, town council, positions on specific boards or committees within local government, often having to do with environmental or land-use planning decisions).
Imagine that the total number of expected votes in a given election is 2,000, and that a simple majority rules. Then, whoever commands 1,001 votes wins. One conceptually straightforward way of “commanding” these votes is to create a voting bloc of at least 1,001 voters. I abstract here from the political difficulties of creating such a bloc, and the logistical difficulties of assembling and organizing it so as to forge a consensus within it. Suffice it to say that since it has been done (or at least approximated), we know that it can be done.*
Suppose that such a bloc can be formed. Suppose that each member of the bloc has enough of a stake in the outcome of the vote as to agree to pledge their vote for a certain option on the ballot. Suppose that every member of the bloc agrees. Then every member of the bloc is under a promissory agreement to deliver his or her vote.
If the vote is tight enough, then every vote will count because the bloc needs literally every available vote to get the bare majority that defeats their opponent. In this case, the probability of my side’s losing the election because I failed to vote (or voted the wrong way) is not that small. But even if that isn’t so–even if there’s more numerical leeway involved–every vote will count in the way that every promise does: if you promise to deliver X, then prima facie, you’re under an obligation to deliver X. Satisfaction of the promise is not contingent on the size or scope of the causal contribution of delivery. A promise to deliver on 1/x of a contribution to some outcome is still a promise to deliver, even if 1/x is, all things considered, a small contribution.
Put it this way: I have an obligation to show up at work tomorrow. When I do, I’ll be making a marginal contribution to a company with 11,000 worldwide employees, a revenue stream in the hundreds of millions of dollars, and total assets in the billions of dollars. The marginal contribution I make by showing up and doing my work will be minimal (believe me). Even so, I still have to go, and there’s a sense in which it matters whether or not I do go. At a bare minimum, the promise obliges me, and the promise matters simply as a binding moral obligation.
Beyond that, my de minimis contribution does play some role, however minimal, in the overall economy of the firm. My failure to show up would in some sense (on some days, by some people) be missed, and not simply because my co-workers miss my charming, handsome presence (granted, a real loss), but because, despite my bad attitude and lax work ethic, I do make some kind of half-assed contribution to the place. If I told my boss that I didn’t have to show up at work because I was only making 1/11,000th of a contribution to the total product of the company (if that), I’d be fired on the spot. A large part of why I’d be fired is not just that I promised to show up and reneged (full stop), but also because it would free ride on my co-workers if I didn’t.
At some least some of this applies, mutatis mutandis, to voting in small-scale elections. Failing to vote after agreeing to vote breaks a promise, and breaks faith in a venture aimed at the common good. It’s unjust, irresponsible, and disrespectful even in the case where each individual is making only 1/2,000th of the contribution to the overall outcome. Failure to make a small contribution when the cost is low says something about one’s commitment to the value of the intended outcome–and says something problematic if one professes a commitment to that supposedly intended outcome. It says that you’re full of crap and can’t be relied on.
Mail-in voting makes all of this even easier than it might be. Yes, you have to make a special request for a mail-in ballot. Once you do, however, the ballot comes straight to your mailbox. You make a few marks on it. You seal it. Then you’re done. If you’re particularly averse to effort, you put the ballot back in your mailbox, and presumably the postal worker takes it away. If you’re willing and/or able to travel a bit, you find a mailbox and drop it in. It doesn’t require a stamp. I grant that some effort is involved here, effort that would make no sense if the outcome had no value. But the effort makes some sense if the outcome has some value. In that case, at least, you can no longer hide behind the “vote doesn’t matter” cliché.
None of this is meant as an exhortation to vote in the 2024 US election, or any election. I mostly don’t care how you vote, or whether you do. I already voted (today, by mail-in, for Jill Stein), so I’m done, and that’s that. Do whatever you want, as long as you don’t try to sell me on the wonders of your candidate. My aim here is simply to offer up a counter-example to the “vote doesn’t matter” cliché, and my point is that it only holds under some conditions, not all. Change the parameters, and your vote can matter.
It’s worth noting, incidentally, that voting is a nearly ubiquitous decision-procedure, not limited to political contexts except when written about by contemporary Anglophone political philosophers. We vote at home or among friends on what to eat and where; we vote at work (or other similar institutions) on courses of action for or against a given project or policy; we vote within juries; we vote as alumni of various institutions, whether for representation on a committee or options on a referendum; we vote on the leadership of the corporations in which we’re invested; we vote in polls on our favorite books, stores, restaurants, or bands; we’re constantly polled or surveyed on our rankings, priorities, and preferences when it comes to products and occasionally policies; and so on. It’s a long list.
Just this weekend, PIAD, an activist group from my undergraduate university, asked me (among many others within its constituency) to write a comment in favor of its proposal that Princeton disclose its investments in, and divest its investments from, the State of Israel. The due date is this Friday the 11th, and the task is to write something that convinces CPUC, the relevant university committee, to recommend divestment to the university’s executive leadership and Board of Trustees. That’s a vote–a vote of historic significance–in which the best-written vote will, if read in good faith, surely matter to the overall outcome. The effort might fail in its intended outcome, but no one can deny that every written statement submitted to CPUC is a vote of some kind, and no one can tell me that my statement or vote doesn’t matter. It can’t fail to matter if I use the opportunity wisely. The act of writing out the reasons is an exercise in rationality and justice that has value in itself regardless of whether or not it persuades the Board of Trustees.
If you insist on thinking of “voting” as “voting in US presidential elections,” you can persist in thinking that your vote never matters. But if you widen the lens, it becomes hard to claim that voting plays no significant role in your life. You’d have to be a very unusual person for that to be the case. I’m not saying you shouldn’t be that unusual person. I’m just saying that I doubt that most of you are. Given that, it is, I think, past time either to retire “voting doesn’t matter” cliché or at least to qualify it. Your vote can matter, as long as you to choose to vote when it actually does. You just have to stop complaining for long enough to figure out when that is.
*The National Black Political Convention of 1972 provides a suggestive example. For commentary, see Robert Greene II, “The Legacy of the Gary Convention,” Black Perspectives (March 25, 2022); Paul Prescod, “The Gary Convention Elided Black America’s Internal Disunity,” Jacobin (March 10, 2024); Nicole Poletika, “‘Tired of Going to Funerals’: The 1972 National Black Political Convention in Gary,” Untold Indiana (Feb. 21, 2018); and Junius Williams, Unfinished Agenda: Urban Politics in the Era of Black Power (2014).
Policy of Truth 
I think we have an imperfect duty to contribute to public goods. Since the duty is imperfect, we can pick the occasions as we prefer. So there’s no duty to vote, but voting is one possible way of satisfying a duty, and so is not per se irrational. (Of course refusing to vote can also be a way of satisfying such a duty, by contributing to a [different] public good.)
LikeLike
I agree with your take on voting. I take myself to be doing something consistent with your view, but also different from it.
Your view is: we have an imperfect duty to contribute to public goods, and this imperfect duty gives rise to a disjunctive duty to either vote or not vote, as appropriate. That entails that voting is not per se irrational (but can be irrational in context).
I agree, but I’m addressing a slightly different problem. Take a case in which there is prima facie reason to vote. Suppose you do. In what sense does your individual vote “matter”? Why not think of an individual vote as in some sense inconsequential even in the cases where an imperfect duty gives you license to do it?
Well, take something that’s a sufficient condition of making an instrumental difference to an electoral outcome. If you promise to make a contribution to that sufficient condition, then you’re promising to become part of the sufficient condition that makes an (expected) instrumental difference to an election. That’s the sense in which your vote can, under certain conditions, matter.
“Choosing to become part of the sufficient condition” will tend to happen in cases where…
(a) the issues are (or an issue is) weighty enough to motivate you to action,
(b) the issues matter not just to you but to a bloc of like minded voters,
(c) members of the bloc can in some sense organize themselves as a bloc,
(d) it is possible to calculate what this bloc needs to tip the scales of the vote, and
(e) it is possible jointly to act effectively on the preceding calculation.
When (a)-(e) obtain, and you-the-individual-member-of-the-bloc promise to contribute to the efficacy of the bloc in a given election, the conjunction of those two things is a sufficient condition of: your vote’s mattering.
Part of my point here is that abstracted from this context, it becomes harder to explain why an individual vote matters. If you simply think of your vote as a single, detached contribution to a remote electoral outcome, then while it’s not overtly irrational, it lacks any clear instrumental connection to the outcome. It’s just a shot in the dark. But if you can identify the sufficient condition of which your vote is a part, and you agree to become part of it, there is a sense in which your morally dutiful action connects in a determinate, specifiable way with the teleology by which the outcome happens. You can identify the causal process that makes the outcome happen in a more explanatory way than the merely additive, i.e., “I added one more vote to the n-1 votes already added.”
Ideally, that’s how–or at least part of how–an imperfect duty should be specified. It makes sense to act on an imperfect duty when doing so matters, not just arbitrarily, i.e., just because the duty is “there,” and you know that you somehow have to satisfy it at some point. Ideally, when you specify an imperfect duty, you identify the case in which it would matter to act on it, in the sense of bringing about some ideal outcome by some determinate, rationally expected process. You then, for those occasions, self-consciously turn what was previously an imperfect duty into a perfect duty: you promise to become part of the larger cause that brings about the preferred outcome. When the time is right for the role you expect to play, you act.
Since that doesn’t happen all that often when it comes to voting, I’m inclined to think that voting is, of all the ways of contributing to public goods, among the less important. But there are cases in which it can be important, particularly in local politics.
More importantly, perhaps, the larger pattern is one that helps put activism in proper perspective. The division of labor in an activist movement has to identify some long-range outcome, organize a bloc around it, assign roles in bringing it about, and then elicit a promise from each member to act according to their chosen role, where acting makes the agent part of the sufficient cause of the action–the more determinate, the better.
If we left activism at the level of a merely imperfect duty, and said no more than that, we would turn it into a series of haphazard actions that never achieved any practical aim. It is an imperfect duty, but there has to be more to it than that.
LikeLike
I came up with an example that I think clarifies the point I’m making in the post. Suppose that we have an imperfect duty to contribute to public goods, and that public health is one of these goods. Now suppose that contributing to your local first aid squad is a way of promoting public health. Consider four ways of contributing:
(1) You send the first aid squad a check whenever you feel like it.
(2) You resolve to send the first aid squad a $20 check every payday.
(3) You learn that the first aid squad needs a new defibrillator, so you set up a fundraiser to buy one, and do.
(4) You learn first aid, and volunteer on the squad yourself.
None of the ways of helping the first aid squad is wrong, obviously. They’re all laudable. But other things being equal, they’re arranged in ascending order of praiseworthiness. I’m inclined to think that within an Aristotelian virtue ethic, if we speak of imperfect duties at all (I guess it’s harmless), a proper subset of those duties should rise from something like (1) to become more like (2), (3), or (4).
The examples are arranged in ascending order along several different dimensions. One is stringency. Example (1) is an imperfect duty that stays imperfect; it’s almost a matter of (benign) whim. But the other examples either voluntarily convert the imperfect duty into perfect duties, or add perfect duties to the imperfect one. All three involve promises, either to self or to others or both (promises being perfect).
A second dimension is causal efficacy at bringing about the beneficial effect. (1) has minimal effects, (4) the greatest, when iterated across time. That said, you could in principle do more good through modified versions of (2) and (3) than some version of (4). If you gave a greater amount in (2), or bought something of even greater value in (3), that might do more good than volunteering a few hours a month.
A third dimension is proximity to the benefit produced: how close are you to actually producing the target good yourself (however much good you do)? Here, (4) wins.
The most praiseworthy ways of bringing about public goods are the ones that involve the greatest stringency, bring about the greatest amount of good, and are most proximate in the way they bring it about. In these cases, there’s a sense in which your individual act matters far more than it would than if you were realizing the imperfect good by method (1). This is not to say that (1) is wrong, or that nothing in your life should be like (1). It’s just that (1) is axiologically low grade.
To bring this back to voting: precisely because voting is a form of representation, there’s really no form of voting that exemplifies (4). That leaves options (1), (2), and (3). My point is that an individual vote is most meaningful–it matters the most–when it exemplifies something like (3), and bloc voting does.
If your pattern of voting always exemplifies something like (1) and (2)–you either vote whimsically or on a rigid schedule–the problem is not so much that that’s irrational as that it seems like a purely rote way of checking off the “contribute to public goods somehow” box. Unless you’re doing something else of a public-good-promoting nature that has causal efficacy, your voting probably won’t. Granted, if you do have some such non-voting thing in your life, maybe it doesn’t matter that your voting doesn’t have causal efficacy. But consider the case in which voting is your main public-goods-promoting activity. In that case, unless you at least approximate (3), you’re not really doing anything worth doing. And special cases aside, (3) can only be done in local politics, where the scale of the electorate is amenable to bloc voting.
All of that suggests to me that unless you can approximate something like bloc voting, voting should really be conceived as an add-on to other public-goods-promoting activities. In other words, if all you can manage is whimsical voting (i.e., 1) or rigidly scheduled voting (i.e., 2), you either should turn that into (3), or else find a way to hitch (1) or (2) to something where your individual contribution matters. But someone who votes every now and then (full stop) is like someone who gives to charity now and then (full stop). Mediocre. And someone who votes in every election without fail is like someone who feels morally satisfied in setting up an automatic $1 payment to Oxfam every two weeks. You could do worse, but you could definitely do better.
I guess this is a long way of saying that there’s a tension between speaking of “imperfect duties” and speaking of hitting the mean. The first needs to be supplemented by the second, and maybe an account of the second is what I’m after.
LikeLike