Voting as a trust

Mill thinks of voting as a trust or ‘public duty’ and claims that

The idea of voting as a trust, therefore, is meant to be different from the idea that one can have duties to vote, or to vote one way rather than another. It seems to imply that the duties one has are public rather than private, a consequence of one's citizenship, rather than of one's personal situation, beliefs and interests.Footnote ⁵ It also seems to imply that we have a duty to benefit those for whom we are trustees, to act on their behalf. Again, this is more specific, and potentially more demanding, than the idea that we have to consider the interests of other people when voting.

There are, though, several difficulties with the idea of voting as a trust, at least as Mill develops the notion. I will briefly describe the problems, and then examine them each more fully. The first is that the idea of voting as a trust seems to make non-voting the norm, or baseline, against which political rights and duties are to be judged. By contrast, democratic theory assumes that all competent adult citizens are entitled to vote, and to do so without special justification or duties. The second problem is that Mill's conception of voting erases the very significant differences in power and authority between citizens and legislators, and wrongly assumes that standards of accountability and publicity that are appropriate for the latter are justified for the former. Finally, it is hard to make sense of the idea of voting as a trust to other people who are also voters – as Mill appears to do. So, while we may have a variety of duties to others – to those who preceded and will succeed us, as well as to our contemporaries and those who are not compatriots – the idea of voting as a trust seems confused and confusing when applied to fellow voters. Taken together, these difficulties with Mill's idea of voting, I believe, suggest that we should replace the idea of voting as trusteeship with the idea of voting as a right, albeit one that may be hedged by a variety of duties to others.

Voting as a baseline for citizenship

Mill believes that ordinary citizens, no less than legislators, are carrying out a public duty or trust when they vote. Hence, in all but exceptional circumstances, citizens and legislators should vote in public, just as publicity should normally be expected when they carry out any other public acts or duties. Without publicity, it can be hard, if not impossible, to secure accountability. So, Mill supposes, citizens ought not to find it any more peculiar to vote publicly than to deliberate and vote openly with each other as members of a jury.

There is something appealing about the idea that voting, like jury service, is the exercise of a public duty, and that exclusion from a share in these duties can be a badge of shame, or of stigma.Footnote ⁶ But while we have no right to be part of a jury – although one may well have a right against unfair exclusion from it – it is hard not to think of voting as a right to which all adult citizens are entitled. As Mill says, ‘it is a personal injustice to withhold from any one, unless for the prevention of greater evils, the ordinary privilege of having his voice reckoned in the disposal of affairs in which he has the same interest as other people’ (302).Footnote ⁷ We can imagine legitimate forms of justice that do not involve jury systems – even if we believe that the former are preferable to the latter. By contrast, legitimate governments that lack universal suffrage are likely to be exceptional – at least from a democratic perspective – and are presumptively illegitimate until proven otherwise.

On this view of what constitutes legitimate government, voting is one of the rights that define what it is to be a citizen, just as military service, and jury duty, in countries with juries, are defining examples of the duties of citizenship. People can have a duty to obey the law and to pay taxes without being a citizen. They can hold considerable socio-economic power and political influence without being a citizen. However, one of the rights that illustrate the differences between citizens and non-citizens in democratic societies – perhaps, the defining example of such a right – is the right to vote in elections to the legislature.

The difficulty with Mill's idea of voting as a public trust, then, is that it seems to make non-voting the baseline against which the rights and duties of citizens are judged, so that the demand for publicity in voting appears a natural concomitant of holding a special position or responsibility. By contrast, once we expect all adult citizens to have a vote, the idea that voting implies a special position, with special duties and constraints on its exercise, appears far more problematic. I do not want to say that one could not come up with some democratic reconstruction of Mill's idea of voting as a public trust.Footnote ⁸ The point, for now, is simply this: that the idea of voting as a trust is hard to square with a democratic notion of citizenship, in part because it seems to make non-voting the baseline against which regulations, and expectations, of the right to vote are judged.

Voting and the differences between citizens and legislators

The second difficulty with Mill's conception of voting as a trust is that it erases the real differences in power between citizens and legislators, and implies that the same standards of publicity and accountability are appropriate to each. Thus, Mill claims, ‘Exactly in proportion as the vote of the elector is determined by his own will, and not by that of somebody who is his master, his position is similar to that of a Member of Parliament, and publicity is indispensable’ (329).Footnote ⁹ Mill insists throughout chapter 10 on the idea that voting confers power over others, but he gives no clear sense of what that power consists in.

Such power as electors have is largely limited to the time between one election and another. It is, therefore, temporally very limited – as Rousseau sawFootnote ¹⁰ – and legally limited only to the ability to accept or reject a candidate as one's representative. It does not extend to power over other aspects of candidates’ lives – whom they marry, where they live, what car they use, what occupation they otherwise hold – although, of course, the desire to be elected puts pressure on candidates who face competition for positions to make themselves agreeable, inspiring and electable.

Above all, the power that electors have is contingent on the way that other electors vote, and this is likely to prove a very serious limitation on their power, whether as regards candidates or as regards fellow voters.Footnote ¹¹ Mill's belief that voters should act as though they are the sole voter in any election illustrates the problem (325–6). Doubtless intended to encourage serious and sincere voting, it merely highlights what, by now, is evident to many voters, as well as to political scientists: that most of the time our votes are so inconsequential that it is a good question why any of us bother to vote at all!

By contrast, legislators have very considerable, though still limited, powers over electors. The position of legislator, in representative governments, requires the legislator to represent the interests of those people who elected him or her, and to make binding decisions on their behalf. Not only are legislators more powerful than electors, while in office, but they are also answerable to the latter for how they have used their powers. So, while legislators may not be able to exercise much power unless acting in conjunction with other legislators, the powers that they are able to exercise in this way are very considerable.

Moreover, it is easier for legislators to organize themselves, and to act collectively, than it is for most citizens, simply because they are likely to be relatively few in number, easy to identify and to contact, familiar with the tools of organizing, and aware of the advantages of doing so. Indeed, simply being elected as a legislator gives one access to information, salaries, public and private buildings that would otherwise be unavailable; and it is likely to confer various types of respect and honour that one would otherwise lack. So, the idea that voters should meet the same standards of publicity and accountability as legislators seems to be based on a serious overestimation of the power and responsibilities of the former, and a striking indifference to the special powers and responsibilities of the latter.Footnote ¹²

Finally, it is hard to know on whose behalf we should be voting, or for whom we are trustees, on Mill's picture of voting. At no point does Mill mention duties to foreigners, when arguing against the secret ballot, or clarify how far the public interest that we are to serve extends beyond our compatriots, and a generation or so of these. On the contrary, Mill gives the impression that ‘the public’, for whom the voter is responsible, is primarily composed of fellow electors – at least under universal suffrage (325–6). However, we normally hold trusts for people who cannot act on their own behalf – so the idea that we are trustees for our fellow electors seems perverse. Mill is sensitive to the ways in which different electoral systems can give voters different – and, in his view, arbitrary and unjustified – powers over each other. But the solution to that problem, he implies, is to adopt some form of proportional representation (ch. 7). At all events, Mill's genuine concern about the different weight of formally equal votes does not figure in his arguments against the secret ballot. Thus, there is something puzzling about the importance that Mill attaches to the idea that voting is a trust because ‘the exercise of any political function, either as an elector or as a representative, is power over others’ (324).

Rejecting Mill's claim that voting is a trust

There are, then, several difficulties with Mill's arguments against the secret ballot, because his idea of voting as a trust is difficult to square with democratic conceptions of citizenship. However, it would be premature to reject Mill's case against the secret ballot on these grounds, because we may be able to capture at least some of Mill's worries about the secret ballot while accepting the more familiar notion of voting as a right. Although Mill was generally careful not to refer to voting as a right,Footnote ¹³ because of what he thought the term implied, I think we can safely reject his assumptions about rights, without fundamentally altering his arguments against the secret ballot. As I will show, voting can be a right without it following that we are entitled to vote in whatever way we wish. Consequently, I will suggest, the question whether or not the secret ballot is justified cannot be resolved solely by deciding that voting is a right, rather than a trust.

Voting as a right

Mill avoided talk of voting as a right, even though he clearly thought that people could be entitled to vote, and that they could be wrongly, or unjustly, denied the vote (315, 324). In part, this was because he thought that voting gives voters power over others, and denied that people could have a right to such power. But he also seems to have believed that if voting is a right, people are entitled to vote however they please. So, he urged, if voting is a right, ‘if it belongs to the voter for his own sake, on what ground can we blame him for selling it, or using it to recommend himself to any one whom it is his interest to please?’ (324).

However, the idea that voting is a right need not have the implications that Mill supposes. Legally, of course, it may not be possible to prevent people voting in ways that they should not but, morally, there is nothing in the idea of a right to vote which prevents us from condemning people who sell their votes, or use them with reckless indifference to the lives and interests of others. We are familiar with this idea when it comes to free speech. People often disagree about the extent to which legal limits on speech are justified by moral considerations, just as they disagree about the nature and weight of the latter. In general, though, no one thinks that the reasons why you are entitled to speak your mind also entitle you to deceive or demean others; to incite people to crimes of various sorts; or to speak so imprudently and recklessly that you put other people's lives in danger – the proverbial ‘crying fire in a crowded theatre’.Footnote ¹⁴

Familiar objections to Nozick's claims about property rights help to illustrate and explain the difficulties of Mill's view.Footnote ¹⁵ Without belabouring the point too much, most rights – and, certainly, all important ones – are really a bundle of rights, rather than a single claim to do something. Quite how we define that bundle – in terms of the differences between claims, liberties, immunities, and privileges, for example, or in terms of the extent and weight of that right compared to other rights, or to various duties – is generally a matter of interpretation, and not something one can deduce from conceptual analysis of the term ‘right’.Footnote ¹⁶ So, one might adopt the idea that if voting is a right one is entitled to sell one's vote, just as one might adopt Nozick's idea that if one owns one's talents, one should be free to sell their use (or, even, oneself) to the highest bidder. However, the idea of a right to vote, a right to property or, indeed, a right tout court, does not require us to adopt these ideas. On the contrary, there are good reasons to reject them, on the ground that they undermine the plausible and attractive idea that rights – moral or legal – are largely justified by the (legitimate) interests that they protect.Footnote ¹⁷

The idea that there are inalienable rights is a familiar one. An example of such a right might be the right against torture. Not everyone agrees that there are inalienable rights, or that the right against torture is one of them. Nonetheless, it makes perfect sense to suppose that our interests in not being tortured are so strong and of such moral (or political) importance that people ought not to be able to renounce them. That being so, the idea that people may not sell their vote is consistent with the idea that voting is a right, albeit a right that people should be able to renounce, though not sell.

If these arguments are correct, we can reject Mill's reasons for thinking that voting is a trust, rather than a right. In fact, while Mill seems to have believed that the arguments against secret voting turn on the idea that voting is a trust, I think we can reject that idea, as well. It is likely that we have duties to others as voters, including duties to promote the public good. However, those duties help to explain why secrecy in voting is generally justified for citizens, though not for legislators. Thus, I will argue, whether we look at voting as a right, or at the duties which plausibly constrain it, we should reject Mill's conclusion that citizens have a duty to disclose, and to defend, their electoral choices to their fellow citizens.Footnote ¹⁸

Self-protection and the right to vote

According to Mill, people are entitled to vote because they are entitled to protect themselves, ‘but only against treatment from which he is equally bound, so far as depends on his vote, to protect every one of his fellow-citizens’ (324). However, it seems likely that people are entitled to protect themselves from threats that other people are incapable of suffering. So, this conception of our right to self-protection and, therefore, the vote, seems too narrow. Moreover, it also seems likely that people are entitled to protect themselves even if they are unwilling to protect other people – as long as they are not actively threatening them. As third parties, we may lack a duty to help those who, for no good reason, are unwilling to help others. But to suppose that an unwillingness to help other people – reprehensible though this may be – deprives one of a right to protect oneself, looks far too strong.

Mill, here, is trying to assimilate a right to vote to a duty to vote, by linking one's claims to self-protection to a duty to protect other people. But this is unpersuasive. My claims to protect myself, quite plausibly, entail like claims for others. What is unclear is why my claims to protect myself entail a duty to help others to protect themselves – and Mill provides no argument in support of this claim. This matters to Mill's arguments against the secret ballot: for these turn, centrally, on the idea that I am fulfilling a public duty when I vote – and it is implausible that I am doing any such thing when I vote to protect myself. The plausibility of Mill's objections to the secret ballot, therefore, turns on his ability to assimilate justified self-interest in voting to a duty to vote on behalf of others.Footnote ¹⁹

If we are entitled to vote, Mill rightly assumes that we generally have the necessary abilities to use the vote to protect ourselves. This precludes paternalist, or authoritarian, arguments for public voting on the grounds that we do not know our own interests, and so need help in voting correctly. We need public voting, Mill assumes, because we are less motivated and knowledgeable about the public interest than our own. By linking our entitlement to protect ourselves to a duty to protect others, Mill seems to have thought that he could treat all voting as though it were voting for the public good and, therefore, on his arguments, properly conducted in public. However, if we are entitled to vote to protect ourselves, it is hard to see how voting can be public without falling into the sorts of paternalist and authoritarian assumptions that would undermine our claims to vote to begin with.

Mill's contrast between secret voting in private clubs, which he endorses, and in general elections, which he condemns, is illuminating. Mill claims that one reason why secret voting in a club is unobjectionable

There are at least two problems here. First, while it is true that some people presume that they are entitled to dictate how others vote, there is no need for us to condone their presumption by endorsing public voting over the secret ballot. Second, it is distinctly peculiar to suppose that the evil of lying, when faced by such presumption, is worse than the presumption itself. Mill's contrast between private clubs and public elections, then, helps to bring out the deeply inegalitarian assumptions of the case for public voting. That case turns on mistrust of ordinary voters, and confidence in their self-appointed tutors.

The latter is sufficiently objectionable to undermine the case for public voting, I think, even if one is pretty sceptical about the knowledge and behaviour of most voters. However, it is unclear how much, or what, information voters ought to know in order to make an informed choice among candidates. A good deal, presumably, depends on who the candidates are, and in what ways they differ from each other. But as long as candidates are free to publicize their own merits, and the supposed demerits of their opponents, and as long as voters have access to a variety of sources and types of information about candidates, and matters of public policy, it is unclear what could justify public voting.Footnote ²¹

Mill's case for public voting implies that we may force people to participate in a public tutorial, in which they have to state and defend their political beliefs, and to listen and respond to countervailing arguments. It is very hard to see how this can be reconciled with a commitment to the freedom and equality of citizens, or with Mill's own justification of representative government.

Self-development and the right to vote

As Mill recognizes, people want to be able to vote not merely to prevent misgovernment, but so that they, too, might govern. He supposes that the vote is important primarily for the former reason, but recognizes that the latter, too, has a place in explaining why people who are capable of self-government should be able to govern themselves. The question, therefore, is what attention to these interests implies about the justification of the secret ballot.

People are entitled to vote, in part, because of the challenge and satisfaction that comes from self-government. The interest in voting, here, is decidedly personal, or self-regarding, to use the language of On Liberty. This interest in self-development is consistent with the argument that, in voting, we should try to identify and pursue the public interest, because the pleasures and challenges of representative government are likely to depend on our ability to identify and to pursue ends that are not our own.Footnote ²² However, it does not therefore follow that voting should be public, rather than private.

To see the problem, we need simply to consider what would justify forcing people to tell us how they have voted, or intend to vote. We might say that we doubt that they have really understood the public good, or that they have chosen the right candidate, given their notion of the public good. We might be right. But unfortunately for Mill's argument, that only shows that some people have failed to promote, or fully to realize, their interests in voting. This no more entitles us to know how they have voted, than it entitles us to force them to discuss their religious beliefs with us. People's religious, as well as political, beliefs may be erroneous in ways that harm, rather than promote, their interests. But absent special reasons to suppose that they are incapable of looking after themselves, we are no more entitled to know how they have voted – even if it affects us – than we would be to prevent them from voting in the first place.

When we consider voting as a right, then – whether in terms of self-protection, or more broadly – it looks as though Mill's case for public voting fails. It fails because if voting is justified primarily by the interests of the voter in voting, there are good reasons why people should be able to keep their votes secret, if they so wish. These reasons are, essentially, anti-paternalistic and anti-authoritarian. Against a background of freedom of expression and political competition, civil liberties and protections for personal privacy, people can always ask each other for advice, if they want it, and seek information to improve their voting, should they so desire. These civil, political and personal liberties mean that there are a variety of ways to promote considered and informed voting, without forcing people to defend their political choices to anyone who is interested in them. Should people choose to disclose their votes, and the reasons behind them, they are free to do so. They may even broadcast this information from the treetops – at least metaphorically – and will be free to urge their own example as an example to others. In those circumstances, it is hard to see why ordinary voters should be forced to report and defend their political decisions. So, the presumption must be that citizens are entitled to keep their vote secret, whether or not they are concerned about coercion, corruption and intimidation.

Private duties and the secret ballot

People have a right to vote because they have legitimate interests in voting, and these cannot be reduced to duties to others, as we have seen. Still, we all also have a variety of duties that should affect the way we vote and, even, whether or not we vote. For example, we may have duties of solidarity and concern for the helpless, impoverished and poor in our countries, and abroad, and these duties may not simply arise because of our duties as citizens, but because of supranational and subnational ties to others. That is, we may have special duties to the poor of our religious or ethnic communities, in addition to duties based on our citizenship and concern for human rights. We may have duties to protect the environment, promote justice, and more specific duties of care to family members, loved ones and friends. Again, some of these may be duties based on ties of citizenship, but others may follow from both broader and more specific sources of duty. We generally accept that these sorts of duty can, and ought, to constrain our self-interested behaviour, including self-interested voting. The two questions we need to consider, however, are: (1) Can such duties properly determine our votes, and (2) If voting based on these duties is justified, should voting be public, rather than private?

Mill is happy to recognize that people have a variety of duties, which cannot be reduced to a duty to pursue the public interest. However, he assumes that it would be wrong for us to base our vote on such duties. Mill never says so explicitly, but it is perfectly clear that he assumes that we should only vote on duties that we share as citizens, and those duties, he believes, are duties to pursue the public good. We will shortly consider whether or not it is true that the only duties we share as citizens are duties to pursue the public good. For now, however, the question is what justification, if any, there might be for voting on what Mill would class as ‘private’ rather than ‘public’ duties.

Private duties, on Mill's view, are duties that we are genuinely entitled to pursue, within the normal constraints that dictate moral action more generally. However, he assumes that it would be wrong to determine our vote on this basis, because such duties do not arise from, or necessarily reflect, our citizenship and the duties that follow from it. When it comes to voting, as to other public activities such as serving on a jury, fulfilling military duties, paying taxes and so on, Mill assumes that we must set aside private concerns – whether self-interested or not – in order to act out of concern for our country, and our fellow citizens.

It is possible that Mill would have agreed that because some religious and moral beliefs require pacifism, conscientious objection should be permissible in the case of war. But whether or not he would have allowed such qualifications to the duties of citizenship, he would have had little sympathy for the idea that our religious beliefs should affect our willingness to pay tax, to serve on juries, or to vote. This is partly because his admiration for religion was highly qualified but, more importantly, because he thought that public duties are important, and properly determine our conduct when the common good is at stake.

Many of us will share Mill's assumptions, to some degree. Few people suppose that our duty to pay taxes should depend on the religion we affirm; and while, perhaps, there is general support for tolerating those whose conscientious convictions preclude jury service, or voting – or, indeed, anything other than the most minimal contact with the modern state – such cases are generally understood to be rare and, potentially, controversial exceptions to general rules.Footnote ²³

Should we agree, then, with Mill that it would be wrong to vote on these private duties, rather than on the public interest? The answer, I think, is ‘no’. Some, though not all, of these duties are likely to be of great, perhaps supreme, importance in the lives of those who hold them; and it can matter enormously to their beneficiaries whether or not they are fulfilled. Religious duties, for example, are not all of a piece, and do not always have an earthly beneficiary; however, fulfilling them can be of utmost importance in people's lives. Likewise, it can be of the utmost importance, to oneself and to others, to stop massacres, wars, famine and torture. Such duties can, quite rightly, prove weightier than our self-interest, the interest of our loved ones, and other duties that we owe. It would be arbitrary and dogmatic to suppose that these otherwise overriding reasons for action could never be more important than public duties. Indeed, there are at least two circumstances in which it would be reasonable for our personal duties to determine our vote.

The first is where the candidates for our vote – be they parties or individuals – are pretty similar in their likely consequences for the public good, so that considerations of the public interest give us no reason to vote one way rather than another.Footnote ²⁴ In a system with the sort of proportional representation and national lists that Mill espoused, such circumstances might be less common than they would be under a two-party system with a single non-transferable vote. Mill clearly favoured voting systems that maximized the number of ways in which the public good could be defined and pursued – although, the difficulty of such systems is that voters may have little idea how their representatives will vote once they are legislators (ch. 7). Still, it is clear that voting systems can affect how much choice one has, as a voter, over how to think of the public good. It seems fair to suppose, then, that where considerations of the public good are indeterminate, private duties should be decisive. However, the extent of the resulting permission to vote on private, rather than public, duties will depend on the electoral system of each country.

A second justification for putting private duty ahead of public interest is if we believe the latter adequately protected by all candidates, so that we are free to vote based on pressing personal duties. For example, we may believe it of the utmost moral urgency to stop a particular war or massacre, to prevent the spread or deployment of nuclear weapons, to do something to stop AIDS, poverty, starvation, global warming and so on. Even though voting to prevent those evils means that we can only secure an adequate, rather than the best, candidate from the perspective of the common good, such voting may be justified, given our conscientious convictions. In short, we may feel bound to subordinate the national interest, or common good, to our other duties, when deciding how to vote. Provided that the national interest is adequately protected, it is hard to see why our fellow citizens should be entitled to complain of such behaviour, or to consider such voting immoral.

We generally have a variety of ways, other than our votes, in which we might promote the causes we believe in, or fulfil our duties, as we see them. So, it is important to recognize that it can be wrong for people to vote on the basis of personal, rather than public, duties. It is, doubtless, controversial when, if ever, concern for the latter should take precedence over the former. However, it seems likely that even if we are sometimes entitled to put private duties above public ones, we are not always entitled to do so.

What, all things considered, we have a duty to do depends not only on the relative importance of our different duties, but on the types of action we are contemplating. The differences between voting, supporting a charity, demonstrating, going ourselves to help, and so on, are thus likely to have some role in determining the relative weight of our different duties at election time. So, we may conclude that, as a general rule, preventing nuclear war, helping the starving in Africa or pursuing the national interest should determine our vote. Nonetheless, we may rightly believe that there are circumstances in which the differences between voting and other types of action requires us to alter this rule, and to vote on a different conception of what we are morally bound to do.

Mill, it seems, would deny that this is possible.Footnote ²⁵ But it is not obviously immoral to suppose that the national interest is sometimes less morally urgent than other things we have a duty to pursue, and that our vote should be given in ways that reflect these moral priorities. At all events, as I have tried to show, you do not have to be indifferent to the claims of public duty to hold such a position. On the contrary, you may believe that, most of the time, national interest properly determines your vote, because the different candidates are about the same on the matters of utmost importance.

If these arguments are convincing, private duty, as well as self-preservation and self-development, can properly determine our votes, on occasion. But, even if Mill is wrong about private duty, he may still be right to believe that voting should be public. After all, we are here considering duty-based voting, and even if these are not the same types of duty that Mill had in mind, as long as we are concerned with duties, rather than rights to vote, the arguments for public voting may be persuasive. The question, then, is: ‘is publicity the price we should pay in order to vote on our private duties?’

Again, I think, the answer must be ‘no’. If people are entitled to vote on private duty, in preference to public interest, they are entitled to do so whether or not other people agree with them. There is, therefore, no justification for conditioning this entitlement on a willingness to submit to the judgement of others. This argument, of course, does not apply to those who wrongly sacrifice public good to personal duty. As a practical matter, it is impossible to have a rule whereby only those about to vote wrongly are forced to reveal their voting intentions. Nonetheless, we might think, were such a rule possible, it would be justified, and would provide some evidence for the view that voting should be public.

The arguments for public voting are, clearly, stronger in the case in which someone would wrongly vote on private duty, rather than public interest. Even here, however, the case for mandatory publicity is weak. The problem, in part, is that publicity in this case unavoidably exposes people to public humiliation, shame and censure for mistaking their moral duty. As a general matter, democracies are chary of exposing people to humiliation and shame, even when they have been exposed as criminals, and carefully circumscribe the forms of public censure to which people are exposed. Public shame and humiliation threaten our ability to see and treat each other as equals.Footnote ²⁶ Though some citizens, having broken the law, may justly be subject to punishment, whereas others are due public gratitude, recognition and reward, these moral differences are not so great that we should deny the former citizenship, or grant the latter special political powers.

Some modern democracies, it is true, believe that those who have committed serious crimes should be denied the vote more or less permanently – and Mill agrees with these restrictions. Nonetheless, in modern, as opposed to classical, democracies, people no longer believe that heinous crimes justify stripping their perpetrators of their citizenship, or forcing them into exile. Indeed, modern democracies no longer feel the need to give exceptionally virtuous people special political powers.Footnote ²⁷ There are a variety of reasons for these changes, including differences between modern and ancient ideas about punishment, entitlements to citizenship and, importantly, to political justification.

Given these changes, it is hard to justify mandatory public voting in order to pre-empt morally wrongful voting. The risk of unjustified humiliation, even in cases where the voting was, or would have been, wrongful, makes publicity very hard to justify. When, in addition, we consider that publicity is not necessary to secure informed voting, and is not guaranteed to secure morally desirable voting, either, the case for publicity is further weakened. It is dealt a fatal blow once we consider how unlikely it is that any of us are actually harmed if people mistakenly vote on their private duties, rather than the public interest, and how small any harm is likely to be. Most people's votes are not critical to an election. Despite Mill's injunction, there is no reason why they – or we – should deceive ourselves about the likely effect that any one individual can have by voting.

This means that there is no justification for mandatory publicity in voting, even when people are likely to mistake their duties. It is wrong to force people to disclose their moral failings when there is almost no chance that their behaviour seriously harmed anyone, and when such disclosure likely entails public shame and humiliation. It punishes confusion about the relative weight of our duties to vote with great severity, although it is doubtful that moral failings in such cases are especially serious, particularly harmful, or even very common. To punish such moral errors so severely is, therefore, inconsistent with the equality of individuals. So, even if we set aside concerns that mandatory public voting might undermine people's freedom, because of its chilling effects on voting, we would have to reject mandatory publicity on the grounds that it is inconsistent with equality amongst voters, and between voters and other people.

So, it seems, the case for public voting fails because there is no justification for public voting if people vote as they ought; and public voting is unjustified as a form of deterrence or punishment for those who fail to do so. As we will see, mandatory public voting is no more justified if we look at voting based on public, rather than private, duties. One of the difficulties with Mill's case against the secret ballot is its assumption that we can draw a neat line between the public and the private, and safely locate all voting in the former. But, as we have seen, ordinary citizens lack the power and responsibilities of legislators. It is quite possible that it is legislators, rather than ordinary citizens, who are likely to be bribed and intimidated into voting one way, rather than another. Nonetheless, it is citizens, and not legislators, who are entitled to keep their political convictions and decisions to themselves.

Public duty and the secret ballot

Mill believed that voters should have the greatest choice possible amongst candidates and, therefore, conceptions of the public good. His case for public voting assumes that people will be able to reach agreement on a conception of the public good that is freestanding, or independent of people's private moral beliefs. Publicity, vibrant political competition, and direct challenges to the beliefs of candidates and voters alike are essential to discover the public interest, according to Mill, and to counter the corrosive and blinkering effects of self-interest.

Political competition and freedom of expression can, indeed, have these beneficial effects. However, political debate and competition can generate more heat than light, intensify mutual suspicions and solidify rifts, especially when people believe that their failure to reach agreement can only be explained by the ignorance, selfishness and obstinacy of those involved. Indeed, this is all too likely to be a problem if, with Mill, you assume that intelligent, sincere and well-motivated voters ought to agree on what the common good requires.

‘Reasonable pluralism’ is the term Joshua Cohen coined for the many ways in which informed, well-meaning and conscientious people might think about matters of ultimate moral and political importance.Footnote ²⁸ It means that there is almost certainly no single conception of the public good on which voters can be expected to agree if, as is likely in a society that protects their freedom and equality, they hold different and mutually incompatible views about what is of ultimate importance in life.

In these circumstances, mandatory public voting is likely only to intensify and harden social divisions, recriminations and misunderstandings. There is no reason why people should be able to understand and sympathize with each other's moral views and predicaments when, apart from their citizenship, they may be perfect strangers, with very little in the way of shared experiences, language or morality. We see the problem every day. Whether the socio-political divides are based on class, geography, sex, race, religion or – as is commonly the case – some combination of these, a few thoughtless remarks, let alone deliberately incendiary ones, can expose seemingly unbridgeable gulfs between one group of citizens and another. It is one thing to require those in power, or those who seek political office, to learn to live with the difficulties created by our shared citizenship, and often irreconcilable values, interests and experiences. It is quite another to suppose that citizens, however timid, inarticulate and uncertain, ought to have to shoulder these burdens in order to vote.Footnote ²⁹

If people are entitled to vote only if they are willing to submit to each other's criticism, many people will be unjustifiably deterred from voting. Who votes, it seems likely, will depend on how likely we are to have other people who are interested in our political intentions; on whether or not our views are familiar, bland and conventional enough to pass unchallenged; and on whether or not we have the good fortune to have the right demeanour, appearance and words to convince others that we are trustworthy, right-thinking and safe. Needless to say, this is not a recipe for freedom and equality among citizens, or for the informed and self-critical political engagement that Mill desired.

The first problem, then, with Mill's arguments for mandatory voting is that there may simply be insufficient common ground to justify forcing people publicly to state and defend their views of the common good. The second problem is that, even if they can agree, this agreement provides no warrant for the view that mandatory voting is justified. Agreement on the public good is not ruled out by reasonable pluralism. However, the latter means that our conceptions of the public good – whether in harmony or not – are dependent on our conscientious convictions about matters that we are generally entitled to keep to ourselves.

As long as you are not responsible for coercing or intimidating your children, it is no business of anyone else whether you believe that arranged marriages are infinitely preferable to ones that children arrange for themselves. As long as you do not promote racial hatred, deny people jobs or otherwise mistreat them, it is nobody's business if, in your heart of hearts, you believe that you are a racially superior being. Which, if any, of our personal convictions, suspicions, interests and actions end up playing a role in national (or local) elections depends not only on us, in a democracy, but on other people. Hence, many of our most deeply held beliefs may have little role to play in determining what is for the common good – at least when it comes to voting. The point, however, is that whether or not they do so, people ought to be able to elect their government without having to defend themselves, their view of the world, or their conscientious convictions to anyone who asks.

By contrast, legislators, and others with special political powers, ought to provide the information that is necessary to establish whether or not they are doing their job. Mill is quite clear that representatives should not be bound to vote on the interests of their constituents, or of their party, no matter what. They must be free to develop their own considered conceptions of the public good, to urge these on other people, and to change them, if they prove mistaken. If one holds a less individualistic view of the duties of representatives and legislators, the need for transparency, publicity and accountability amongst legislators becomes even more acute. Party discipline, procedural rules, the complexity of government and the necessarily significant gaps between elections, all make it hard for voters to know what their representative has done, with what justification, and with what consequences. Of course, there can be justified forms of secret deliberations and voting amongst legislators, in part because secrecy may be required in order to treat some matters with the frankness and freedom they deserve. Nonetheless, publicity in voting should be the rule amongst legislators, in order to ensure that legislators will be unable to manipulate, coerce and bribe other people.

Finally, the duty of citizens is not exhausted by the duty to vote for the common good. On the contrary, as citizens we may all have duties that require us to curtail the pursuit of national advantage, and collective ends that we would otherwise be justified in pursuing. Some of those duties are owed to fellow citizens – to those who are poor, disabled, victims of prejudice and injustice, for example. As citizens we may owe these fellow citizens special consideration and a first priority on collective resources. Such duties are not best thought of as duties to advance the public good. If anything, they are constraints on how we can define the public good, and the ways in which we might pursue it. Likewise, we may have duties that, collectively, we owe to foreigners, such as duties to repair injustices that our country caused, and from which it benefited; duties of aid and refuge; duties of reciprocity; and so on.

People can legitimately disagree about the content and weight of these different duties. That much is evident from debates over what reparations, if any, are due to descendants of those who suffered from the slave trade, from imperialism, and from the Holocaust. But it is perfectly reasonable to claim that people owe duties to others as citizens, and that these duties are not reducible to a duty to pursue the common good of their compatriots. I rather doubt that Mill would have objected to the claim that such duties exist. However, the problem of evaluating the relative duties of citizens in this case, like the difficulty of determining what is for the common good, or deciding whether or not private duties should take precedence over the national interest, illustrates, once again, the problems of mandatory public voting.