I’ve been thinking a lot lately about voting. I have Jason Brennan to thank for having stimulated me to sustained thought on the subject, via his much-acclaimed book, The Ethics of Voting. As I’ve said before, I agree with Brennan’s thesis in a general way, but the more I think about the details of his argument, the less plausible I find them. (I find his arguments for voter disenfranchisement downright hopeless.) Here’s a link to the 2013 Reason Papers symposium on Brennan’s book, and here’s a link to an earlier critique at PoT of Brennan’s account of character-based voting.

I’ll have more to say about Brennan’s arguments as I find the time to write about them. Meanwhile, here are four quick thoughts on voting, three of them relevant to American elections, the fourth to Israeli elections. In each case, it seems to me that the wrong issues are being discussed–when they’re being discussed at all–and that we ought to change the terms of debate. Only the last of the four topics is relevant to Brennan’s work.

(I) Felon Disenfranchisement
There’s been a lot of talk in the past few years about felon disenfranchisement: felons in the U.S. (perhaps elsewhere, but I don’t know) are deprived of the right to vote. Here’s a fairly typical piece from The New York Times criticizing felon disenfranchisement as racist.

I find discussion of this topic confused. There are at least three different issues involved here; each needs to be distinguished from the others and discussed on its own terms.

(1) A first issue is: given an ideal definition of “felony,” and a well-functioning criminal justice system, should felons be permitted to vote, or should they be deprived of that right as an inherent part of their punishment?

My answer is, “they should be deprived of the right to vote.” I endorse a debt-based conception of punishment according to which, when we interact with someone, we owe him or her (at a minimum) respect for their rights. When someone violates those rights, he incurs a debt to the victim–a debt consisting of compensation for the lost value of the exercise of the victim’s right, among many other things. Punishment, in my view, ought to consist of repayment of that debt. If the debt can’t be paid in full–and for a variety of reasons, it may be impossible to do so–offenders can permissibly be deprived of those goods that would count as ill-gotten gains from crime.

Some simple examples: If you rob me, your voting to dispose of my income without having compensated me for the commission of the crime counts as an ill-gotten gain. Since you’re not entitled to such a gain, you can be disenfranchised. If you kidnap me, what you’ve done is illicitly to try to “govern” my actions by brute force. If I survive, you owe me compensation for your trying to rule me in this way. But voting is a case of ruling me, as well. So ruling me by the ballot counts as an ill-gotten gain (or would) until you’ve paid off the debt you incurred by kidnapping me. And so, once again, you can be disenfranchised until you do.

Suppose that the repayment-requirements on such debts are prohibitively high–high enough that they can’t typically be paid in full by anyone, regardless of how wealthy they are. On my view, government should in such cases have the authority to deprive offenders indefinitely of the right to vote. If you (the offender) can’t compensate me (the victim) for what you’ve done to me, you don’t have the right indirectly (i.e., by voting) to decide the disposition of goods that belong to me. And that, in effect, justifies the disenfranchisement policy we currently have. (For a somewhat similar view of punishment, see the work of Daniel McDermott, who defends what he calls a debt-based conception of retributivism. I’m not sure where McDermott stands on felony disenfranchisement, however.)

Suppose now that we think of government, on Lockean grounds, as a kind of mutual-defense pact for the protection of rights. In that case, any attack on the rights of any member of the pact is an attack on the rights of every party to the pact. By implication, a debt owed to the victim is simultaneously a debt owed to every party to the pact. If the debt in question cannot be discharged in full–and if the crime is serious enough, it probably can’t be–then the parties to the pact can permissibly deny the offender access to ill-gotten goods in lieu of full payment of the debt.

This resolves the old problem of the “missing beneficiary.” For example: if you murder me, you incur a debt to me for having done so. Of course, being dead, I’m not around to collect the debt. In that case, you owe a debt to the rights-respecting members of my society–via their agent, government. Now suppose that you can’t pay the debt in full. In that case, they can deprive you of certain categories of goods on my behalf as well as theirs. One good they can deprive you of is the right to vote: after all, your having that right would give you the right to dispose of the income that they have earned while you still owe them compensation for the right you’ve violated. And you’re not entitled to that. (Incidentally, even if I’m physically unavailable to collect a debt for having been murdered or wrongfully killed, I could during my lifetime have set up an escrow account as an insurance policy in the event of my murder/wrongful death. In that case, an offender might still be obliged to compensate me posthumously, with the proceeds going to my heirs or to the state, as my will or lack of one implies.)

Something similar would apply to rape, to assault and battery, to drunk driving, and to plenty of other recognizable felonies. In short, I don’t see why, as long as we define “felony” properly, felons should be allowed to vote. The debts they’ve incurred to the rest of us are sufficiently high that we needn’t worry so much about whether they have the right to govern us, or dispose of our income. They don’t. I don’t mean to suggest that we have no obligations toward them. I just means that access to the ballot isn’t one of them.

(2) Second issue: is the operative definition of “felony” in the U.S. a good one? Does it, on moral grounds, include and exclude the appropriate items?

I’d say: “no” and “no.” This issue is the one that, in my view, actually gives rise to the felon disenfranchisement controversy. The real problem, it seems to me, is that we’ve made felons of people who shouldn’t be felons, and in consequence of that, have deprived people of the right to vote who should have it. If doing so has adverse racial consequences, my suggestion is: redefine “felon” more narrowly, so as to exclude certain categories of crimes from the list of felonies. If we do, I suspect that the “felon disenfranchisement” problem (insofar as it is a problem) either disappears or is greatly reduced in scope.

(3) Third issue: regardless of the definition of “felony,” is the U.S. criminal justice system systematically and unjustifiably biased against certain populations or sub-populations?

My answer: “probably.” No matter how we define “felon,” there will probably be residual problems in our criminal justice system, some of them with adverse racial consequences–some of them just plain old unjust–and those problems need to be addressed. But the resolution of those problems is not facilitated by the enfranchisement of felons. Convicted murderers, rapists, batterers, and drunk drivers have no distinctive insight into the rights and wrongs of criminal procedure. Nor does it make much sense to bank on the possibility that some small fraction of those convicted felons might be innocent (I’m sure some are), and might impart the wisdom of innocence to us via the ballot. The probabilities of that happening are tiny enough to render the venture as a whole quixotic.

The bottom line is that instead of crusading for voting rights for murderers, kidnappers, rapists, robbers, etc., we ought to be redefining “felony” and actively reforming the defects of our criminal justice system. Felon enfranchisement is just a distraction from those far more important tasks.

(II) Voter ID laws
Now consider voter ID laws. Here’s a usefully balanced article, also from the Times, suggesting that voter ID laws, while problematic, do not have the large-scale effects that some have alleged of them.

The standard argument for voter ID laws is that they pre-empt or minimize voter fraud. The standard argument against them asserts that there is little evidence of voter fraud in the U.S., that voter ID laws have racist effects, and that contrary to their proponents’ rhetoric, voter ID laws are covertly there to produce racist effects.

Once again, however, all this seems to me a distraction from the real issue. To see why, consider the tacit implication of the arguments against voter ID laws. Why, according to those arguments, are voter ID laws unfair? Spelled out, the answer is that large numbers of Americans lack the means to obtain photo IDs for themselves. Lacking access to photo IDs, they can’t meet the requirements of voter ID laws, and are de facto disenfranchised by them.

Suppose ex hypothesi that that’s true, and pause on it for a moment. Voting aside, isn’t that precisely the problem in need of discussion and rectification? How is it that large numbers of people in a first world country do not have access to the means of self-identification? Even if we do away with voter ID laws, the underlying problem remains in place. In other words, even if you don’t need an ID to vote, you need it for other things. How are people without IDs expected to open bank accounts, visit the doctor, or travel by plane–or get driver’s licenses, library cards, discount cards, or government benefits, etc.? Either they’re to do without these things because they lack ID, or they need access to these things, and must therefore obtain access to IDs. I would opt for the latter option, but no matter how you slice it, the issue is not voter IDs, but access to IDs as such. 

The scarce-access-to-IDs situation seems to me a good argument for having some equivalent of a national identity card in just the way and for just the same reasons that so many other countries have them. Here’s a case for them, from the Washington Post.

I agree with the reasons the Post gives for having them, but I’d give one more. It’s been argued by critics of social contract theory since Hume* that express consent theories of consent to government do not or cannot work because we never in fact consent expressly to government. I suppose that that’s partly true, at least for natural-born citizens; we don’t consent to government in the way that we consent, say, to the terms of a credit card. But I see national ID laws as a chance to respond to that problem. Why not structure the task of getting a national ID so that the act of getting one either requires express consent to the government issuing the card, or requires explicit non-consent? If you consent, you get an ID card, and with it, the benefits and burdens of “membership” in the polity. If you refuse consent, you don’t get a card, and can be denied the benefits of membership while being spared the burdens.

There are, to be sure, lots of complications here, many of them entangled in debates about immigration and immigration policy. I can’t settle those here. I would just say that it seems to me that the mechanism I describe is possible, and that its existence would rebut Hume-type arguments against consent, and solve some other practical problems as well. At the very least, focusing on our ID problem–which has significant adverse effects on people’s living their lives–beats focusing on a voter ID problem that seems not to have any significant effects on voting.

(III) Low voter turnout
Now consider the low voter turnout issue. The problem here is supposed to be that relatively few voters show up to vote. In partisan terms, that means that Democrats fare badly in the elections (which, of course, matters more to Democrats than to others). In more general terms, it means that our democracy is not as “robust” as it could be. Personally, I happen to think it means that the ballot choices we’re typically offered aren’t worth voting for, whether for or against. Here’s a website, FairVote.org, devoted to discussion of the issue. Once again, however, it seems to me that much of the discussion there and elsewhere is focused on the wrong things.

Suppose that we want to increase voter participation. (There are reasons not to want to, having to do with wrongful voting and voter incompetence, but set them aside.)  In that case, I’d offer two proposals:

(1) Put a “None of the Above” option on the ballot, so that voters can vote against all the (other) options on the ballot. As things currently stand, you can write “NOTA” as a “write in” on the ballot (I regularly do), but few people realize this, and most people surmise, correctly, that write-ins are meaningless. (I’ve encountered poll workers unaware of the fact that NOTA is a write-in possibility.) But if “NOTA” were on the ballot, it would be at least as significant as any other option on the ballot, and all those disgruntled voters who don’t vote because they dislike all the options might now vote in order to express that view.

(2) Move Election Day from Tuesday to the weekend. Yes, a small minority of mostly religious voters might be inconvenienced by that move (if so, they can use absentee ballots), but as it stands, huge numbers of working people are inconvenienced by Election Day’s having to compete with the workday. Change the day, and I suspect you’d increase voter turnout.

(IV) Voting and the right to complain
 Let me move now from American to Israeli elections, or more precisely, elections in Jerusalem. When I visited Israel/Palestine in 2013, I was both surprised and dismayed to discover that while East Jerusalemite Palestinians have the right to vote in Jerusalem’s municipal elections (though not in Israeli national elections), they almost unanimously refuse to exercise that right, even though their exercising it would substantially change the political landscape of Jerusalem, and benefit them. The argument I heard from Palestinians was that voting would legitimize Israel, which they refuse to do. Sadly, the few Palestinians who offered to run for municipal office, or to vote for pro-Palestinian candidates or causes, were widely regarded by other Palestinians as traitors to the Palestinian cause.

I find that a self-defeating and incoherent set of attitudes. East Jerusalemite Palestinians widely accept–and demand–government benefits from Israel, so it makes no sense for them to refuse to exercise political rights that are on offer from Israel, especially if the refusal to exercise those rights merely disempowers those who refuse to exercise them. The fact is, the budget for government services in East Jerusalem is in the hands of non-Palestinian Israelis, as are decisions bearing on the protection of Palestinian rights. As things currently stand, decisions on both sets of issues are made in ways that ignore or violate Palestinian rights. I would argue that respect for one’s rights is essential to one’s well-being. As it happens, the only efficacious way of ensuring respect for Palestinian rights in Jerusalem is to make changes to the budget and policies of the Jerusalem municipal authority. And the only efficacious way of changing the budget and policies of that authority is to vote to change them. So the options are: vote to defend your rights, or acquiesce in their violation and the consequent diminution of your well-being.

Suppose that we each  have a self-regarding moral obligation to promote our well-being (insofar as doing so is open to us). If so, give the preceding facts, Palestinians ought to vote. If “ought-hood” is sufficient for “duty” or “obligation,” then eligible Palestinian voters have a moral obligation to vote. Contrary to a recent argument of Jason Brennan’s, then, the case of East Jerusalemite Palestinians seems a picture-perfect example of the old saw that if you don’t vote, you shouldn’t complain–or more precisely, if you don’t vote, you shouldn’t complain about the things that voting would have improved, and that only voting can improve, at least for the foreseeable future. If you’re going to be taxed, and you’re going to be regulated, it makes no sense to stand by as your tax money is spent by everyone but you on everything but what matters to you. It likewise makes no sense to stand by as you are regulated to death by the people who are spending your money, as your rights go violated or ignored. Voting is in effect an act of self-defense, and self-defense is a moral obligation.

The obstacle here is supposed to be that it is not instrumentally rational for individual voters to vote, because individual votes cannot change the outcome of an election (or more precisely, cannot change the outcome of a sufficiently large election–a qualification that is sometimes relevant but often ignored in discussions of “voting,” as though all voting were large-scale voting). But if you know anything about Palestinian political culture, I think you’ll see that this objection is spurious. There is no need to worry about the efficacy or utility of individual votes qua individual if the voters in question don’t conceive of their votes in those terms in the first place. If voters naturally conceive of themselves as members of a solidaristic group, and can coordinate their efforts in a given direction as a group–and have a strong reason to do so, and might well be inclined to do so–then the unit of concern is not the utility of individual votes, but the the votes of voting blocs qua blocs whose members self-consciously act in concert.

I realize I’m describing an idealized case, but my point is, it’s a possible case. In fact, it’s more possible and plausible than half of the thought-experiments that clog the philosophical literature. (By the way, there is no contradiction between seeing yourself as an individual with an individual obligation to promote your well-being, and seeing yourself, qua voter, as part of a voting bloc. Membership in the bloc could precisely be what promotes your individual well-being, so that your individual well-being is what dictates membership and a solidaristic self-conception in the first place.)

Now suppose that Palestinians** get their act together, ditching the nationalist and Islamist rhetoric that has retarded their progress for decades. They come to see voting as an act of both collective and individual self-defense. They also see the defense of their rights as a contribution to the common good (which includes Israelis). Suppose (perhaps improbably but not impossibly) that the Israelis do not interfere significantly with Palestinians’ voting en masse.

Suppose further that Palestinians think of voting by analogy with having an intifada. In other words, as with the first intifada in the 1980s, they organize their efforts to vote strategically*** as a single unified voting bloc: they caucus, organize, and promise one another to vote for pro-Palestinian policies. Suppose that it is relatively obvious what these policies should be, and what the votes for these policies should be. Suppose, further, that voters are well-informed. Now suppose that a large number of Palestinians enter these caucuses voluntarily, and through caucusing, manage to ascertain (by mechanisms internal to the caucuses) that there are enough Palestinian votes among them to tip the scales of a given Jerusalem election. If so, each Palestinian voter could regard himself or herself as part of an assurance contract with all other Palestinian voters. And if so, each voter would have an obligation (to the others and to him or herself) to vote in the way he or she had promised in the contract.

My argument here is essentially that if you can organize a mass uprising–an intifada–you can organize a mass voter campaign. Further, if an intifada involves the implicit equivalent of an assurance contract (as it does), you can in principle model  a mass voter campaign on an intifada, and turn the campaign into an activity that involves an actual assurance contract. But if contracts bind, an electoral assurance contract yields a duty to vote. So under certain nomologically possible conditions, there can be a duty to vote, and given this duty, it can be irrational to complain about unfair or harmful political policies if you don’t vote.

I can’t work through all the details here, but take a look at Brennan’s argument in light of the preceding. Either my East Jerusalem case is a counter-example to his thesis, or it’s a defeater for it. In the first case, it refutes the thesis as stated. In the second case, it suggests that the thesis is highly misleading as stated. Given that, my argument requires that Brennan qualify his claims about the ethics of voting in ways that take more explicit stock of cases like the East Jerusalem one–something that would substantially change the “flavor” of his theory.

I realize that Brennan has an explicit discussion of strategic voting in his book (Ethics of Voting, pp. 129-33), and that the discussion includes a “strategic voting clause” (p. 131), but I think almost all of what he says talks past what I’m saying here. What he doesn’t discuss either in the book or in the article I’ve linked to, is the possibility that you could have a duty to vote in cases like the East Jerusalem one, that your vote would matter in those cases, and that you’d have no right to complain if you didn’t vote. (See the notes below for a comment on “strategic voting.”)

While you’re looking at Brennan’s arguments, read his discussion of “the moral disenfranchisement of poor minorities” in The Ethics of Voting, pp. 105-7. I find the discussion very inadequate even on its own terms, but for present purposes it’s worth noting how narrow it is. Like so many American writers, in writing about “minorities,” Brennan structures his discussion around black-white relations in the U.S., assuming somehow that what he says about that will generalize elsewhere–everywhere. It doesn’t. In particular, he assumes that “poor minorities [will] overwhelmingly qualify as bad voters” by his criteria, and offers some rather handwaving suggestions about how they’re to handle–or how he would think about handling–their disenfranchisement.

What he doesn’t consider is the possibility that the issues in contention in a given election may sometimes be entirely straightforward and require nothing in the way of the social scientific “credentials” he regards as necessary conditions for eligibility to vote. Putting aside the American case, I think this is patently obvious in non-American ones, like that of East Jerusalem. It takes no special social scientific wisdom to figure out that your interests, your rights, and the common good are better promoted by someone who stands for fairness than by someone who makes no secret of wanting to subvert your interests, violate your rights and exclude you from the common good. If Brennan’s epistemic elite hasn’t figured that out, frankly, they have a lot to learn.

I’m hoping to spend the summer of 2015 in East Jerusalem teaching at Al Quds University. While I’m there, I intend to make the case for what I call rights-based strategic voting by Palestinians in Israeli elections. Feel free to hit me with objections in the combox if you disagree with the sketch I’ve just given of it. I may well be hit with more than that while I’m there, and I’d like to start my preparations now.

*Actually, Hume concedes, almost parenthetically, that consent is a possible basis for political legitimacy: “I only pretend [aver] that it has very seldom had place in any degree, and never almost in its full extent” (paragraph 20). But that claim is entirely compatible with consent’s coming to be the basis of political legitimacy in the future by concerted effort aiming to bring it about. Considered as an argument against Locke on consent, what Hume says in “Of the Original Contract” strikes me as a series of ignoratios elenchi.

**For brevity, I use the word “Palestinian” throughout, but I don’t really mean to be restricting that to ethnic Palestinians. I’m using “Palestinian” as short-hand for those who would actively organize for and act on behalf of Palestinian rights in East Jerusalem. The bulk of those people would most likely be ethnic Palestinians, but not all of them would. It’s just too cumbersome to be explicit about this in every sentence.

***I’m using the term “strategic” in its colloquial, not its technical sense. In its technical sense, “strategic voting” is voting for candidates or policies that are contrary to one’s sincere preference, in the hopes that doing so will realize some preferred outcome. In the colloquial sense, “strategic” voting is simply voting to bring about some end by means of a collectively-adopted political strategy for bringing the end about. I happen to think that the technical concept of “strategic voting” is a confused and equivocal one, but that doesn’t matter. My scenario makes no reference to insincerity on voters’ part.

Postscript, Nov. 30, 2014 (relevant to proposal I, felon disenfranchisement): This blog post, at Slate Star Codex, is well worth reading on the race and criminal justice in the United States. It complicates the picture, but I don’t think it changes anything I said about felon disenfranchisement. Hat-tip: Kate Herrick.

Postscript, April 5, 2015 (relevant to proposal IV, voting and the right to complain): Useful background on the political situation in East Jerusalem, from the London Review of Books.

Postscript, December 25, 2015 (relevant to proposal II, voter ID laws): An interesting article in The New York Times about Mayor DeBlasio’s “New York ID” program and the obstacles to success it’s facing at area banks. All things considered, the program seems a step in the right direction.