Bryan Caplan  

Jason Brennan's The Ethics of Voting

Garett Jones on IQ and Economi... The Depression as a Positive P...
Jason Brennan is my favorite philosopher under the age of 40.  Now he's published a marvelous book, The Ethics of Voting, with Princeton University Press.  In 210 short pages, he raises and resolves a series of ethical dilemmas every potential voter faces:  Is there a duty to vote?  What is the connection between voting and civic virtue?  How should you vote if you do vote?  When is buying and selling votes wrong?  He concludes by doing something very unusual for a philosopher: He combines his moral theory with social science to pass moral judgment on the citizens of the world's democracies.

I hate to spoil the suspense, but Brennan's main answers are:

1. There is no duty to vote.  In fact, if you're going to vote the wrong way (see below), you have a duty not to vote.  People might have duties of beneficence and reciprocity, but voting is only one way to discharge them.

2. There are many extrapolitical ways to exercise civic virtue and contribute to the common good:
[M]any activities stereotypically considered private, such as being a conscientious employee, making art, running a for-profit business, or pursuing scientific discoveries, can also be exercises of civic virtue.  For many people, in fact, these are better ways to exercise civic virtue.
While we're at it, let's add "having children" and "being a good parent" to the list.

3. There's no duty to vote, but if you do vote, you have a duty to vote "only for things [you] justifiably believe would promote the common good."  He later elaborates:
As a citizen, you do not owe it to others to provide them with the best possible governance.  But if you take on the office of voter, you acquire additional moral responsibilities, just as you would were you to become the Federal Reserve chairperson, a physician, or a congressperson.  The electorate decides who governs.  Sometimes they decide policy directly.  They owe it to the governed to provide what they justifiably believe or ought to believe is the best governance, just as others with political power owe it to the governed to do the same. [emphasis mine]
4. Paying people to vote for things you justifiably believe promote the common good is morally permissible.  So is accepting money to vote for things you justifiably believe promote the common good.  His evidence includes multiple examples involving the Godfather trilogy and Tetlock's work on sacred values.  Read it for yourself.

5. After surveying the literature on voter ignorance and voter bias, Brennan concludes that violations of ethical voting are widespread.  While voters usually focus on the common good, they are epistemically irresponsible and deeply in error on many questions of great policy relevance.

The book ends on a note both high and depressing:
My goal has been to defend certain normative claims.  If this book induces better behavior among voters, great, but I do not expect that.  My purpose has been proof, not persuasion or behavior modification.  If voters behave badly, we will need more than one philosophy book to fix that.
P.S. My favorite aside in the book:
Philosophers often use state-of-nature thought experiments to help illustrate how politics contributes to the common good... [W]e can also imagine an "inverse state of nature" - a political society that lacks private, nonpolitical activity.  In the inverse state of nature, people try to gather together for public deliberation, voting, and law creation, but no one engages in private actions.  In the inverse state of nature, life would also be nasty, poor, brutish, and short, because there would be no food, music, science, shelter, or art. 

COMMENTS (13 to date)
joecushing writes:

I'd.say that voting with ignorance is fine. It incourages politicians to speek to your wishes more than not voting at all. If ignorant voters just voted at random till a political leader had something to say that they were interested in, their votes would cancle out till then. For example: maybe the reason 18 year olds don't vote is because they don't care about the issues. Maybe they don't care about the issues because the issues are slanted towards older people because they vote. If millions of teens started filling the poles with votes at random, we would start hearing political speaches to them.

John writes:

joecushing articulates the view that any democratic outcome is acceptable in what he suggest. This is pretty much the opposite of Brennan's core thesis: democracy has to aim at the general good/common good rather than special interests.

Brennan, as do I, rejects the claim that some invisible hand process is operating within our democratic institutions. Without such a feature some other mechanism is required to keep democracy from devolving into some form or crony totalitarianism as history has shown occurred in a number of democratic societies. One might think a written constitutions serves this purpose; that seems debatable.

One line of thinking that is not well developed in the book is what ramifications this has for our political representatives when they vote on specific policies, legislations, regulations.... We can and certainly should ask more from our representatives that they have given us in terms of concern for the general good of the society over all the special interests that currently dominate the process.

Randy writes:

Not a believer in the idea of a "common good", so not buying the idea that this can serve as a justification for voting or to determine what constitutes a good vote. But the thought experiment about a society which consists of only political behavior is outstanding!

Nick Bradley writes:

Charging a nominal fee to vote, like $15 or $20 dollars, would eliminate most ignorant voters.

The informed voter realizes that more than $15 or $20 dollars out of his pocket are at stake.

Eric writes:

I can see "Being a good parent", but I have a hard time understanding the civic virtue of having children in a world that is, arguably, beyond a sustainable rate of growth. Seeing as how having children is an imperative driven by genetically programmed behavior, not having children is the civic virtue. Having children should be viewed as a privilege earned by performing other virtues.

Mavaddat Javid writes:

I love it when moral philosophers include paradoxical prescriptions (such as, "If you are not informed and you know it, you should abstain from voting") in their enumeration of public duties.

There is no argument for why we have no duty to vote in the above summary. Neither do I find one forthcoming on Brennan's website ( As far as I can see, this conclusion is just blindly asserted.

The fact that voting is only one way to discharge one's duty to public beneficence is not an argument for why it's not an obligatory way to do so.

For example, imagine someone observing a drowning child and reasoning to themselves, Well, there are many ways I could aid that child's health, therefore I don't have any duty to aid their health by saving them from drowning. But this is clearly absurd. This is because the argument, "There are many ways to do X, therefore I am not obligated to do X in any particular instance," is just logically invalid. Just because there are many ways to fulfil a category of duties (e.g., contributing to public health), that doesn't mean one isn't obligated to execute that duty in any particular instance (e.g., vote to prevent a bill that compromises public health).

If Brennan argues that we have a duty to abstain from voting when we will vote poorly, it follows that we have a duty to vote if we will vote well. This is because, in general, it's difficult to discern a morally relevant difference between failing to prevent harm to others (when one is in a position to do) and causing it oneself. Where we are obligated to prevent harm by not acting, we are also obligated to prevent harm by acting (where it is possible).

All this is assuming we only have negative duties (obligations to prevent harm). However, it cannot be the case that we have no positive duties whatsoever (i.e., obligations beyond non-maleficence). This consideration opens up a whole host of other considerations that Brennan seems to ignore.

Perhaps most devastatingly, Brennan takes it for granted that we have epistemic duties to be justified in our beliefs. But if this the case, then we have a duty to be justified in our political beliefs. But where we are justified in our political beliefs about potential harms to others, we must be obligated to vote to prevent those harms. More than anything else, this point seems most fundamentally to contravene his conclusion that we don't have a duty to vote (as though voting were purely supererogatory).

Philo writes:

On 1.: Voting is a way of expressing support for one's democratic form of government. The value of this expression by any one voter is small, but the cost to that voter of voting is usually also small; the former will often exceed the latter. These small quantities are hard to measure; especially, the value of the existing form of governance, as compared with the likely alternatives, cannot be determined with any precision. But in a large country with a pretty good government the positive expressive value of voting probably outweighs the negative expected value of an ill-informed (or even an ill-intentioned) vote in determining the outcome of the election or referendum. So in many circumstances someone who is eligible to vote had better do so *even if he is completely ignorant about the issue on which he is voting*. Of course, saying this falls short of saying that he has a *duty* to vote.

On 3.: the alleged duty to vote "only for things you justifiably believe would promote the common good" is very implausible. In a national election in the U.S. it is almost certain that my vote will not decide the issue, so the question, "What would happen if my vote *did* determine the outcome," is of almost no practical importance.

Jason Brennan writes:


On 1: I have a rather interesting argument claiming that expressive voting is impermissible.

On 3: The reason philosophers find my argument interesting is that I find a way to defend an obligation not to vote badly despite the fact that individual votes don't make any difference.

Jason Brennan writes:


I spend two chapters arguing against there being a duty to vote. I don't blindly assert it.

Panu Pelkonen writes:
if you do vote, you have a duty to vote "only for things [you] justifiably believe would promote the common good."

Isn't it the key strength of democracy that it 'averages out' the different interests of people? Voting from your personal interests only can therefore hardly be very harmful. In the worst case it leads to 51% robbing the other 49%.

If a policy for 'the best common good' does exist, doesn't it follow that someone has better understanding of this policy than others, and (s)he should be made a dictator?

Secondly, it is still possible to know which policies promote your personal interests. It's quite difficult to guess which policies would benefit say ones co-workers or relatives, and impossible to guess it for people one doesn't even know.

Surely promoting common good in the elections is important, but isn't it actually so difficult, that people should be given advice first to make sure they understand their own interests and defend them, and only secondly, promote public good (if they think they can understand it).

John T. Kennedy writes:

Brennan says:

Citizens have a right that any political power held over them should be exercised by competent people in a competent way.

Considering the nature of government, this is akin to saying that citizens have a right that any robbery inflicted upon them be done by competent people in a competent way.

No, people have a right to not be robbed, which is not the same thing at all as a right to be robbed competently. Likewise people have a right to not be governed.

John T. Kennedy writes:
But if you take on the office of voter, you acquire additional moral responsibilities, just as you would were you to become the Federal Reserve chairperson, a physician, or a congressperson. The electorate decides who governs. Sometimes they decide policy directly. They owe it to the governed to provide what they justifiably believe or ought to believe is the best governance, just as others with political power owe it to the governed to do the same.

Suppose there are two candidates on the ballot: A and B. A promises to rob Peter to pay off Paul, B promises to rob Paul to pay off Peter. Suppose that we can reasonably judge that A will do somewhat less harm than B to the governed in general, but considerably more harm to me since my name is Peter.

In this case Brennan's argument would mean that every voter is morally obliged to collaborate in robbing me, and that I am morally obliged to cooperate in harming myself if I "take on the office of voter".

Faced with a government bent on robbery I would in good conscience vote for B to protect myself from robbery if I thought the expected return on my vote was sufficient. I can't fathom why I (someone who wants no one robbed) owe anything to the governed, especially considering that many of the governed are enthusiastic participants in robbing me.

Vince Skolny writes:

Point 2 intrigues me. In my unscientific empirical (anecdotal, I suppose)observations the citizens least likely to engage in civic virtues as "being a conscientious employee, making art, running a for-profit business, or pursuing scientific discoveries," are likely voters that vote incorrectly and ignore their duty to abstain.

Also, until recently, Bryan Caplan was my favorite economist under 40. ;)

Comments for this entry have been closed
Return to top