Mandate Reasons and the Ethics of Representation

Research

Mandate Reasons and the Ethics of Representation

Author:

Abstract

Democratic theorists have developed a rich literature exploring why winning democratic elections gives representatives the moral right to occupy public office. However, they have largely ignored the question of whether receiving greater or lesser voter support ever makes a moral difference to how representatives ought to govern. In this essay, I defend the existence of mandate reasons for representatives to govern in specific ways that are derived solely from voter support. I argue that the relationship between votes and mandate reasons is characterized by the dependence thesis, which says that votes strengthen representatives’ mandate reasons to govern in specific ways when and only when voters express supportive attitudes for governing in those ways. Because not every vote for a representative expresses the relevant kind of support for their proposals, not every vote strengthens their mandate reasons. This has important implications for the ethics of representation. 

Keywords: Democracy, Vote, Mandate, Representation, Equality, Accountability, Humility

How to Cite:

Bajaj, S., (2025) “Mandate Reasons and the Ethics of Representation”, Political Philosophy 2(1). doi: https://doi.org/10.16995/pp.19736

Downloads
Download PDF
Download XML

Mandate Reasons and the Ethics of Representation

Sameer Bajaj

Philosophy, University of Warwick, UK

Democratic theorists have developed a rich literature exploring whether and why winning elections gives representatives the moral right to occupy public office and make authoritative laws.1 However, they have largely ignored the question of whether winning elections with greater or lesser voter support ever makes a moral difference to how representatives ought to govern once in office. This is an important gap in normative democratic theory that impedes a reasonably complete understanding of the moral power of the vote and the ethics of representation.

In this article, I defend the existence of mandate reasons for representatives to govern in specific ways that are derived solely from voter support. I argue that the relationship between votes and mandate reasons is characterised by the dependence thesis, which says that votes strengthen representatives’ mandate reasons to govern in specific ways when and only when voters express supportive attitudes for governing in those ways. Generally, the ‘specific ways of governing’ that mandate reasons justify are determined by the proposals that representatives advance during their electoral campaigns. Importantly, because not every vote for a representative expresses the relevant kind of support for their proposals, not every vote strengthens their mandate reasons to implement their proposals.

The dependence thesis makes a sufficiency claim (‘when’) and a necessity claim (‘only when’). I defend both claims by distinguishing three morally significant things votes can do: they can issue judgments on political disputes, convey uptake of campaign promises, and express beliefs about the wisdom of political choices. I then argue that each of these functions of voting is associated with core democratic values—equal respect, accountability, and humility—that convergently support the dependence thesis.

The dependence thesis opposes the assumption, widespread in real-world democratic politics, that mandate reasons track margins of electoral victory. When representatives claim ‘powerful’,2 ‘resounding’,3 ‘clear’,4 or even ‘huge great stonking’5 mandates after electoral victories, they imply that their margin of victory has given them special moral justification to implement their platform. I argue that because only a fraction of those who vote for a representative will support any given proposal they advance, the dependence thesis entails that only a fraction of voters will strengthen their mandate reasons to implement any given proposal. At the same time, the dependence thesis opposes the kind of scepticism about mandate reasons that seems to prevail among political theorists.6 I argue that voter support for proposals grounds genuine moral reasons that compete with other kinds of reasons in determining how representatives ought to govern, all-things-considered.

This article proceeds as follows. In Section I, I clarify the moral question I address and lay out key assumptions. In Section II, I examine Alex Guerrero’s view that votes for a representative strengthen their moral justification to act as a ‘trustee’ who governs according to their own independent judgment rather than a ‘delegate’ who defers to voters.7 I argue that it is difficult to see why votes always make this moral difference. In Section III, I develop the account of mandate reasons based on the dependence thesis outlined above. In Section IV, I discuss how to measure the strength of mandate reasons. In Section V, I explore when mandate reasons can make an all-things-considered difference to how representatives ought to govern. In Section VI, I address the worry that, if the dependence thesis is true, there are significant epistemic barriers to determining the content and strength of mandate reasons. I respond that these barriers can be overcome to some degree through institutional reform or polling analysis, and that the moral importance of doing so imposes a distinctive set of demands on representatives and other political actors. Because my aim in this article is to identify and address foundational questions about mandate reasons that have been largely ignored in the literature, the argument leaves important questions unanswered. In Section VII, I identify several remaining questions.

I. PRELIMINARIES

I will assume that representatives who are duly elected according to the electoral rules of a given system (e.g. plurality or proportional rule) legitimately occupy their office. By ‘legitimately occupy’, I mean that they have the moral right to occupy the office and to exercise its constitutive legal powers at least some of the time. The question is whether receiving greater voter support during elections ever gives duly elected representatives stronger moral reasons to govern in specific ways rather than others. I refer to such reasons, if they exist, as mandate reasons.

By calling reasons given by voter support ‘mandate’ reasons, I mean to piggyback on conventional discourse surrounding electoral mandates, though I do not mean to adopt all the assumptions of this discourse. As I mentioned in the Introduction, I take it that representatives who claim especially strong ‘mandates’ imply that their margin of electoral victory has strengthened their justification to implement their platform. This is a view about the electoral phenomenon—margin of electoral victory—that grounds mandate reasons. I do not want to accept or dismiss this view at the outset. It might be right, but there are other views that deserve consideration. Another possible view is that mandate reasons are determined by the number of votes a representative receives, regardless of how many votes they win by. A third possibility—the one I will later defend— is that some votes for a representative but not others strengthen their mandate reasons. Any defence of this view must provide a principled criterion for distinguishing votes that strengthen mandate reasons from those that do not. So, there is a question about which empirical phenomenon grounds mandate reasons.

There is also the question of moral grounds: what moral values or principles ground mandate reasons? I assume that any answer must explain why mandate reasons are genuine pro tanto reasons that can, in certain contexts at least, make an all-things-considered difference to how representatives ought to govern. Moreover, an account of the moral grounds of mandate reasons must vindicate their content-independence—it must explain why they are derived solely from voter support rather than the goods achieved by governing in specific ways. And such an account must explain why mandate reasons aggregate in an egalitarian way, in the sense that greater voter support entails stronger mandate reasons.

Finally, a theory of mandate reasons faces the question of content—what actions do mandate reasons justify? In the next section, I try to make progress in answering this and the other questions by examining Alex Guerrero’s view that receiving more votes invariably gives representatives stronger reasons to use their independent judgment rather than defer to voters’ judgments. I argue that it is difficult to see why votes always make this moral difference, and this points the way towards the alternative theory of mandate reasons that I develop in the rest of the article.

II. GUERRERO ON THE ETHICS OF REPRESENTATION

During an innovative defence of the rationality of voting, Guerrero offers what is, as far as I know, the only account of mandate reasons in the literature. The core idea is that receiving more votes strengthens a representative’s ‘manifest normative mandate’—their level of measurable support within an electorate8—which strengthens their justification to govern as a ‘trustee’ who gives more weight to their independent judgments rather than a ‘delegate’ who defers to voters.9 When representatives impose policies in the face of voter disagreement, Guerrero argues, ‘concerns about domination loom large’. The threat to autonomy looms less large when a representative receives a large share of possible votes, since ‘so many people voted to authorise her to be in a position to make these decisions’. Moreover, representatives who receive fewer votes ought to display a greater degree of epistemic humility by deferring to voters’ judgments.10 They rightly show greater confidence in their own judgment when they receive more votes, since ‘one thing the vote usually implies is that one believes the candidate will do a good job acting as one’s representative’.11

The problem with this argument is that, in many cases, it seems that the values of autonomy and epistemic humility require representatives to defer to voters’ judgments. Suppose that everyone in a political community votes for a representative to implement a specific health care policy that they promised to implement on the campaign trail. If this increases their moral freedom to decide to not attempt to implement the policy, it implies a perversely self-undermining picture of democratic autonomy in which voting for political ends necessarily gives representatives the moral freedom to pursue different ends. Likewise, greater voter support plausibly gives representatives stronger reasons of humility to defer to voters rather than ignore them.

Here, then, is my objection to Guerrero’s account. If people vote for a representative to govern a specific way—to implement specific policies or perform other official actions—it is hard to see why this should give them the moral freedom to govern a different way. If people vote for a representative to exercise greater independence of judgment, this plausibly strengthens their justification to do so, but only because people have voted for it. This points towards the alternative account of mandate reasons that I defend below.

III. DEMOCRATIC VALUES AND THE DEPENDENCE THESIS

The rest of the article develops an account of mandate reasons based on the dependence thesis, which says that votes strengthen representatives’ mandate reasons to govern in specific ways when and only when voters express supportive attitudes for governing in those ways. Mandate reasons are generally reasons for representatives to implement political ends they proposed during their electoral campaigns. Of course, the extent to which representatives are successful in getting their proposals implemented will generally depend on the support and opposition of other representatives and political actors. Crucially, for any given proposal that a representative advances, many people who vote for the representative will not express the kinds of supportive attitudes for the proposal that ground mandate reasons to implement it.

I defend the dependence thesis by distinguishing three things that votes can do: they can issue judgments about political disputes, convey uptake of campaign promises, and express beliefs about the wisdom of options. I then argue that each of these functions of voting is associated with core democratic values—equal respect, accountability, and humility—that convergently support the sufficiency claim (‘when’) and necessity claim (‘only when’) advanced by the dependence thesis.

A. Equal Respect

When democratic citizens vote in elections, they issue judgments about how political disputes are to be settled. Voter judgment involves the expression of attitudes, but it is not the mere expression of attitudes. It alters the normative situation more directly by contributing to the authoritative process of democratic self-rule.12 Votes are analogous to judicial judgments in this way. Judicial judgments express beliefs about legal disputes, but they are not mere expressions of beliefs. Rather, they exercise a normative power that authoritatively settles legal disputes in a way that disputants are duty-bound to follow. Likewise, voter judgment involves the exercise of a normative power that contributes to the authoritative settlement of political disputes.13

What are the ‘political disputes’ that voters’ judgments help settle? In a representative democracy, the most obvious dispute in elections concerns who will occupy a given public office. The power to settle this kind of dispute implies that democratic citizens share what Niko Kolodny calls ‘occupancy control’ over the office. There is a further question of whether voters share ‘directive control’: can they bind representatives to implementing specific directives once in office?14

The ideal of egalitarian self-rule is more fully realised if they can. Representation is not a fundamental requirement of egalitarian self-rule. On the contrary, it introduces a significant inequality at a crucial stage of the decision-making process by concentrating power in the hands of the elected few.15 This is why a direct democracy that consists exclusively of decision-making by referenda is more procedurally egalitarian than a representative democracy. The institution of representation is a practical concession that is necessary to maintain a healthy division of labour in a complex society that depends on people specializing in different tasks, including the task of political decision-making. Of course, representatives need some latitude to determine how best to pursue the directives voters give them within the complex legislative process and in a way that coheres with a holistic agenda. But if representatives have total moral freedom to ignore voters’ judgments about how they ought to govern, they sit at the top of the decision-making hierarchy rather than voters.16

There is a puzzle about how votes for a candidate can express judgments about how they ought to govern. It is perhaps clear how votes can express the judgment that a candidate ought to occupy a given office. However, there is rarely17 space on the ballot for voters to explicitly articulate judgments about how candidates ought to act once in office. How, then, can votes express such judgments?

One promising answer is that the practice of electioneering makes it possible. Candidates propose to pursue political ends during their electoral campaigns, and this creates a deliberative context for voters to pass judgment on these ends. The proposed ‘ends’ generally include specific policies—concerning, for example, the economy and taxation, health care, public safety, immigration, international relations, criminal justice, the environment, liberal and political rights, and so on—bundled together in a platform that candidates publicly defend and reference in rallies, debates, and interviews. Sometimes, parties or candidates run on a single-issue platform or give clear priority to a single issue (e.g. the National Health Action Party in the UK). It is also possible for candidates to propose governing with independence of judgment or certain virtues of character or personal qualities (e.g. strength, compassion, patriotism). Because the ‘ends’ that candidates propose can be varied in these ways, citizens can pass judgment on a wide variety of ends. The more general point is that the content of representatives’ mandate reasons is determined by the content of their electoral proposals.

The egalitarian element of egalitarian self-rule implies that voters have a legitimate interest in having their judgments about political ends given equal respect in the process of self-rule.18 This cannot be done by implementing their judgments directly, since they disagree about which ends to pursue. It can be done only by giving their judgments equal weight in determining which ends to pursue. Since failing to give voters’ judgments equal weight wrongs them and, other things equal, it is worse to wrong more people in this way rather than less, greater voter support plausibly gives representatives stronger mandate reasons to implement their proposals. Importantly, however, mandate reasons can make a moral difference only within the limits fixed by the value of equality itself; they cannot all-things-considered justify the implementation of deeply inegalitarian proposals. In this way, mandate reasons grounded in the value of equality must be understood as pro tanto reasons.

A crucial question is: does voting for a candidate invariably express support for their proposals? It might be argued that because candidates publicly articulate and defend their proposals in advance of the election, a vote for a candidate just is a vote for their proposals. If this is right, then voters cannot issue different judgments about two different choices—first, the choice of a candidate and, second, the choice of their proposals. They can only issue a single judgment of support for the candidate and their proposals. This implies that voting for a representative invariably strengthens their mandate reasons to implement their proposals. Perhaps something like this idea undergirds the conventional view that mandate reasons track margins of electoral victory.

Voters’ interest in having their judgments given equal respect counsels against this view. This interest is best promoted if voters can, but do not invariably, express support for a representative’s proposals. Ordinary voters are detached from the internal workings of political parties. As such, they must choose between candidates whose proposals are shaped by political actors with much greater opportunity to influence what those proposals are, including candidates themselves. Despite this, some voters may genuinely support a representative’s entire platform. But many voters will actively oppose some, most, or all of the platform of the candidate they vote for. A clear case is the negative voter who votes for a representative whose platform they find deeply morally problematic, but still all-things-considered better than the platforms of rival candidates. If the representative interprets this person’s vote as expressing support for their entire platform, they give greater weight to the judgments of those with greater power to influence that platform. From the standpoint of egalitarian self-rule, then, votes should be interpreted as expressing a judgment of support for a proposal—and thus strengthening a representative’s mandate reasons to implement the proposal—only if voters intend their vote to express support for the proposal. If this is right, then representatives also ought not interpret votes made in ignorance of a given proposal as expressing support for it, since the ill-informed voter lacks the relevant intentions. The ideal of egalitarian self-rule thus vindicates the dependence thesis—votes strengthen a representative’s mandate reasons to implement their proposals when but only when voters express judgments of support for those proposals.19

B. Accountability

The idea of equal respect for judgment seems to miss something important about the nature of voting. Votes are not merely judgments of support for settling political disputes in certain ways. They are a valuable resource for promoting one’s legitimate civic interests and the interests of all citizens. When a candidate encourages me to spend this precious resource to further their bid to gain political power, I have the right to insist that they follow through on the proposals that I relied on in casting my vote. Campaign proposals are not mere statements of a representative’s beliefs or preferences. They are ‘solicitations of trust’.20 Representatives invite voters to rely on their proposals in exchange for votes. When voters accept this invitation, a relational structure of accountability is generated. This is akin to promissory accountability, or perhaps even an element of it. If representatives win office and pursue policies that are fundamentally at odds with the proposals that citizens have relied on in casting their votes, they take citizens’ votes for political self-gain without following through on their end of the bargain. This wrongs voters in a certain way. In cases where representatives are not merely overly idealistic or ignorant, but wilfully deceptive or misleading, they exploit voters for their own gain. This provides the moral basis for a conception of mandate reasons grounded in the value of democratic accountability. These reasons are content-independent: they arise from voter uptake of proposals rather than the content of the proposals. And because it is plausibly worse to wrong more people in this way rather than less, reasons of accountability plausibly aggregate in the way characteristic of mandate reasons.

Some might be sceptical that campaign proposals in the rough-and-tumble of real-world politics could generate accountability in this way. Everyone knows that representatives will say anything to get a vote, the objection goes, so campaigning takes place in a ‘suspended context’ where no expectation of performance is justified.21 However, this conclusion is too stark. There is good evidence that representatives reliably attempt to follow through on many of the central elements of their platforms.22 This tendency is rationally explicable partly because representatives are generally sincere about their political convictions, to some degree. Moreover, citizens can incentivise representative to follow through by credibly threatening to vote them out of office. Because voters can bind representatives through a reciprocal structure of backwards- and forwards-looking accountability—they can ensure follow through in the future by punishing past failures—they can take representatives’ proposals seriously.23

Importantly, citizens have an interest as democratic agents in being able to control whether their vote makes representatives accountable to them to implement their proposals. To reiterate, voters will often be in a position of having to choose between candidates whose platforms they partly or wholly oppose on moral grounds. If their vote for a candidate invariably expresses uptake of their proposals, this imposes a significant moral cost. For they would not be able to use their vote as a tool for promoting better political outcomes without making representatives accountable to them to implement proposals they morally oppose, thus implicating their moral agency in a way they might rightly feel threatens their integrity. Citizens have an interest in having the moral freedom to vote without bearing this cost. If this is right, then votes should be interpreted as expressing uptake of a proposal—and thus strengthening a representative’s mandate reasons to implement the proposal—only if voters intend their vote to express uptake of the proposal.

C. Humility

Like any system of government, democracy can be evaluated as an instrument for making better or worse choices according to the correct principles of justice, stability, efficiency, and so on. Correctness here is an independent standard that does not invoke the procedural values undergirding democracy. Epistemic democrats argue that, under favourable conditions, democratic majorities can be expected to reliably make the correct political choices.24 The most influential model for explaining the collective competence of majorities is Condorcet’s Jury Theorem (CJT). CJT demonstrates that when voters are individually more competent than random and vote sincerely and independently of one another, the probability that the majority supports the correct choice increases and approaches one asymptotically as the number of voters increases.25 The larger the electorate the more accurate the majority becomes, and, with enough voters, the majority’s choice is almost certainly correct.

There are well-known debates about whether the assumptions of CJT hold in real-world democracies. I do not want to wade into these debates here.26 I defend the conditional claim that if the assumptions of CJT hold, then when more voters express the belief that a representative ought to implement a proposal, this plausibly strengthens their reasons of humility to implement the proposal. This suggests an epistemic account of mandate reasons grounded in the moral importance of pursuing the correct political ends. Crucially, few real-world voters will believe the representative they vote for ought implement all their proposals—many will believe that specific proposals ought not be implemented while others will have no beliefs at all about whether those proposals ought to be implemented. Because such votes do not provide evidence that proposals ought to be implemented, they do not strengthen a representative’s mandate reasons of humility to implement the proposals. An epistemic account of mandate reasons should thus affirm the dependence thesis—votes strengthen a representative’s mandate reasons to implement a proposal when and only when voters express the belief that the proposal ought to be implemented.

An important limitation of this account is that the beliefs of people when voting do not necessarily have greater evidentiary value than the beliefs of people who are not voting. This is true of the beliefs of people before and after they vote and the beliefs of people who don’t vote at all, both within and outside of one’s voting district. If a competent American citizen believes that the UK ought to implement a given health care policy, this provides the same evidence for the policy as the beliefs of competent British voters. Because the epistemic account has trouble explaining the electoral component of mandate reasons (i.e. why voters’ beliefs have a special role in grounding reasons of humility), I believe it rightly plays only a supplemental role in grounding mandate reasons. By ‘supplemental role’, I mean that it can supplement mandate reasons of equality and accountability—insofar as representatives are considering implementing a proposal justified by these reasons, the fact that voters in the epistemically engaged context of an election believe that the proposal ought to be implemented can help alleviate, to some degree at least, worries that it ought not be implemented by procedure-independent standards.

D. The Dependence Thesis and Real-World Elections

I’ve argued that core democratic values convergently support the dependence thesis, which says that votes strengthen a representative’s mandate reasons to govern in specific ways when and only when voters express support for governing in those ways. The ‘ways of governing’ that mandate reasons justify are determined by the content of representatives’ campaign proposals. The discussion vindicates mandate reasons as genuine pro tanto moral reasons that representatives must weigh in determining how they all-things-considered ought to govern.

The dependence thesis is not meant to vindicate all the assumptions of conventional democratic discourse surrounding electoral mandates. Indeed, it has significantly revisionary implications for this discourse. As I mentioned in the Introduction, when representatives claim especially strong mandates after large electoral victories,27 they imply that their margin of victory has given them special moral justification to implement their platforms. The dependence thesis implies that this is generally false, for several reasons. First, a representative’s mandate reasons to implement a given proposal will be weaker than they would have been had all those who voted for the representative supported the proposal. Second, representatives who win larger electoral victories do not necessarily have stronger mandate reasons to implement their proposals than representatives who win smaller victories, since the former might enjoy weaker levels of voter support for their proposals than the latter. Third, the strength of a representative’s mandate reasons is not determined by how many more votes they receive than the opposition, but by how many voters express support for their proposals.

Just how revisionary the dependence thesis is will depend on how many real-world voters express support for the proposals of the candidates they vote for. As I argued above, many ill-informed voters and negative voters fail to express the relevant kind of support. There is good evidence that both kinds of non-supportive voting are ubiquitous in real-world democracies, especially ill-informed voting. Studies consistently find that most voters are ignorant about most of the platform of the candidate they vote for.28 Most voters do not reference candidates’ policies when asked why they favour them.29 It seems only a small proportion of democratic electorates can discriminate between parties’ platforms.30 There is good reason to think that pure ‘retrospective’ voting—voting based on perceptions of the overall success of a representative rather than support for specific proposals—and ‘group’ voting—voting based on social identities—is far more common than voting in support of a candidate’s proposals.31 According to any plausible view, then, only a fraction of those who vote for a representative will support any given element of their platform.

However, there are difficult issues that arise in determining the relationship between voter support and mandate reasons. One issue is that while most real-world voters are largely ignorant of the details of candidates’ proposals, many more know of and support at least some of the more general political ends those proposals aim to serve.32 For example, a voter might support a party’s general goal of dramatically reducing immigration even though they have no idea that the party proposes to achieve this goal by eliminating certain visa categories, freezing all non-essential immigration, implementing changes to how asylum claims are handled, implementing length-of-residency requirements to claim state benefits, and so on.33 Does voter support for the general goal ground mandate reasons to pursue at least some of the available means of realizing the goal? If so, does it ground mandate reasons to implement the specific policies that the party proposes for realizing the goal, even if they are not the only feasible means for realizing the goal? I believe that, in many cases, the answer to these questions must be yes given a plausible interpretation of the general principle that reasons for ends generate reasons for means. But since I don’t have the space to fully defend this position here, I will leave it to another day to do so.34

A second issue has to do with the scope of voter support. Suppose that a voter expresses support for a party’s proposal to expand the skilled worker visas their nation awards to 10,000 per year. Does this strengthen their mandate reasons to expand the visas to 10,000 or 9,999 per year? Intuitively, the answer is yes, since mandate reasons for policy X can justify policy Y if Y is sufficiently similar or ‘close’ to X, and the relevant visa policies satisfy any plausible criterion of closeness. As the literature on the scope of consent demonstrates, establishing a principled criterion of closeness is a difficult and complex task.35

There are also difficult questions about how to interpret votes motivated by representatives’ personal characteristics. There is good evidence that voting driven by perceptions of candidates’ qualities—such as their integrity, empathy, patriotism, or intelligence—is common.36 Does such ‘candidate trait’ voting strengthen a representative’s mandate reasons to govern with integrity, empathy, patriotism, or intelligence? The answer will depend in part on whether candidates have proposed to govern with those qualities. I argued above that what makes it possible for votes to express support for representatives governing in specific ways is that candidates have proposed to govern in those ways. If voters express support for a candidate’s proposal to govern with certain personal qualities, I think that this can strengthen their mandate reasons to do so. But the mere fact that voters are motivated by their perceptions of the candidate’s personal qualities does not by itself strengthen their mandate reasons to govern with those qualities, both because (a) they might not have proposed to govern with those qualities and (b) being motivated by perceptions of a candidate’s personal qualities does not entail expressing support for governing with those qualities.

Regardless of how these issues are resolved, the dependence thesis will have significantly revisionary implications for real-world claims about electoral mandates. Indeed, it provides a critical edge for subjecting such claims to moral scrutiny.

IV. THE STRENGTH OF MANDATE REASONS

Given the dependence thesis, the strength of a representative’s mandate reasons to implement a proposal is determined by how many voters express support for it—the more voters that express support, the stronger the reasons. In this section, I address two issues that arise in trying to make this general claim more precise. The first is whether the strength of mandate reasons is determined by (a) the percentage of actual voters who express support for a proposal or (b) the number of eligible voters who express support for the proposal. I think the answer is (b). Consider the following two cases:

Election 1: 300 of the 600 people who vote in a district with 1,000,000 eligible voters express support for representative Alice to implement policy V.

Election 2: 300,000 of the 600,000 people who vote in a district with 1,000,000 eligible voters express support for representative Bel to implement policy W.

If the strength of mandate reasons is determined by the percentage of actual voters who express support for a given proposal, then Alice’s mandate reasons to implement V are just as strong as Bel’s mandate reasons to implement W. But that seems wrong—because more eligible voters in Election 2 express support for W, the failure to implement W wrongs more people in the ways implied by the core democratic values discussed in Section III.

If this is right, it raises a further question about whether the strength of mandate reasons is determined by (a) the raw number of eligible voters who support a proposal or (b) the percentage of eligible voters who support the proposal. Consider the following set of cases:

Election 3: 4,000 actual voters in a district with 5,000 eligible voters express support for representative Charu to implement policy X.

Election 4: 4,000,000 actual voters in a district with 5,000,000 eligible voters express support for representative Danish to implement policy Y.

Election 5: 7,000,000 actual voters in a district with 300,000,000 eligible voters express support for representative Edha to implement policy Z.

Who has the strongest mandate reasons? It depends, I think, on what is meant by ‘strongest’. Strength can be understood in terms of how close a representative’s mandate reasons are to being as strong as they might be (i.e. if every eligible voter expressed support for a proposal by voting). In this sense, the strength of mandate reasons is determined by the percentage of eligible voters who express support for a given proposal. This implies that Charu’s and Danish’s mandate reasons are equally strong, and both of their mandate reasons are much stronger than Edha’s mandate reasons.

There is another sense, however, in which the strength of mandate reasons must depend on raw numbers. This becomes salient when representatives weigh their mandate reasons against other sorts of reasons, for example reasons grounded in procedure-independent principles of justice and beneficence. The strength of reasons of justice and welfare plausibly depend on raw numbers—other things equal, it is worse to harm and exploit more people and it is better to promote the rights and interests of more people, regardless of the size of the group or community to which they belong. Because representatives in larger districts generally have a greater overall impact on justice and welfare than representatives in smaller districts, the former generally have stronger reasons of justice and beneficence to govern in specific ways. This means that if the strength of mandate reasons does not depend on raw numbers in any sense, mandate reasons will have radically different roles in the practical reasoning of representatives in large and small districts. If there is no sense in which Danish’s mandate reasons are stronger than Charu’s mandate reasons, then either (a) Charu’s mandate reasons will almost certainly massively outweigh their other moral reasons or (b) Danish’s mandate reasons will almost certainly be massively outweighed by their other reasons. Both conclusions are problematic, since (a) seems to give mandate reasons too significant a role in representatives’ practical reasoning while (b) gives mandate reasons too insignificant a role. If there is a sense in which the strength of mandate reasons is determined by the raw numbers, this implies that Edha’s mandate reasons are stronger than Charu’s and Danish’s mandate reasons in this sense. However, Edha’s mandate reasons are also much more likely to be outweighed by their other moral reasons.

V. WHEN DO MANDATE REASONS MAKE A DIFFERENCE

When do mandate reasons make a difference to how representatives ought to govern? I will assume that representatives are required (and permitted) to implement the feasible political ends that are best considering all relevant reasons. Suppose that a candidate for democratic office named Satya wins an election campaigning on a proposal to implement health care policy A. Policy A is a member of the feasible set of health care policies {A, B, C}. Consider the following three elections:

Case 1: 10% of eligible voters express support for Satya to implement policy A.

Case 2: 51% of eligible voters express support for Satya to implement policy A.

Case 3: 90% of eligible votes express support for Satya to implement policy A.

Is Satya ever required to implement policy A in some but not all the cases?

Suppose that policy A is independently morally permissible and the best of the feasible health care policies according to non-mandate reasons. In Cases 2 and 3, both mandate reasons and non-mandate reasons support A, and Satya is required to implement A. So, having stronger mandate reasons doesn’t make a difference to which policy Satya is required to implement in Case 3 as opposed to Case 2. However, having stronger mandate reasons in Cases 2 and 3 might make a difference to what Satya is required to do as opposed to Case 1. Suppose that in Case 1, the other 90% of voters oppose A and support B instead. There are still reasons of accountability to implement A because voters have taken Satya up on his proposal to implement it. However, equal respect of judgment and epistemic humility seem to demand that Satya does not implement it, since large majorities oppose policy A. This raises the possibility of a conflict internal to mandate reasons. In such cases, representatives must weigh these sorts of reasons against one another to determine what the balance of reasons requires them to do. I do not advance a view of how to do so here, but it is plausible that the balance of mandate reasons favours not implementing A.

Suppose that policy A is independently permissible and precisely as good as policies B and C according to non-mandate reasons. In Cases 2 and 3, there is no conflict between non-mandate reasons and mandate reasons, and the weight of mandate reasons requires Satya to implement A. Having stronger mandate reasons doesn’t make a difference to whether Satya is required to implement A in Case 3 as opposed to Case 2. The same reasoning applies when considering Case 1 as opposed to Case 2. It also applies when comparing Case 1 to Case 3.

Suppose that policy A is independently permissible but worse than policy B and policy C according to non-mandate reasons. Mandate reasons for policy A are strongest in Case 3 (90% support), weaker in Case 2 (51% support), and weakest in Case 1 (10% support). Whether mandate reasons require Satya to implement A depends on how much better B and C are and how weighty mandate reasons are relative to other sorts of reasons. This depends on how morally important the core democratic values that ground mandate reasons are relative to other values. I believe they are very important, and so mandate reasons can justify the implementation of political ends that are significantly worse than ideal according to non-mandate reasons. Plausibly, given that A is independently permissible, mandate reasons will justify its implementation in Case 1. Cases 2 and Case 3 are more difficult calls—we need to know more about how much worse A is than B and C according to non-mandate reasons. The point is that mandate reasons can make a difference to how representatives ought to govern in this sort of case.

The bottom line is that having stronger mandate reasons can, but does not always, make a difference to what political ends representatives are required to implement.

VI. THE EPISTEMIC DEMANDS OF MANDATE REASONS

It might be objected that, even if mandate reasons exist, representatives face significant epistemic barriers to determining their content and strength, since it is generally not possible for voters to indicate support for specific proposals that representatives advance on the electoral ballot. Unless there is a reasonably accurate way to determine levels of voter support for proposals, the objection goes, mandate reasons are of limited practical relevance to representatives.

My response has three parts. First, the moral importance of overcoming these epistemic barriers provides a pro tanto case for the increased use of referenda.37 The final vote tallies of well-crafted38 referenda provide much better evidence of levels of voter support for political ends than the vote tallies of representative elections. This does not necessarily imply that the use of referenda ought to be increased all-things-considered, since voter ignorance and the need to weigh mandate reasons against other sorts of reasons provide a pro tanto case against referenda.

Second, in representative elections, I think it is possible to roughly determine levels of voter support for proposals through well-designed surveys administered to a sample group of voters. Because the aim is to measure and aggregate attitudes expressed through votes, surveys conducted soon before or after an election are likely to have the greatest evidentiary value. It is obviously possible to construct and administer surveys that measure whether individuals support specific proposals.39 The question is whether these surveys are accurate enough to draw justified conclusions about the approximate strength of mandate reasons. It is not necessary for the data to be perfectly accurate, since there is bound to be all sorts of imprecision in representatives’ attempts to weigh up their reasons. It is enough that the data is reasonably accurate.

All surveys have potential sources of error, including respondent selection issues (e.g. sampling and nonresponse errors), response accuracy issues (e.g. framing and specification errors), and survey administration issues (e.g. survey mode errors).40 One especially salient source of error is what Philip Converse calls the problem of ‘nonattitudes’, which occurs when individuals do not have any attitude towards a policy issue but express one anyways when asked to avoid the embarrassment of seeming ignorant or negligent.41 This problem threatens the reliability of survey data as a measure of mandate reasons because, as I’ve argued, only genuine voter support strengthens mandate reasons. Researchers attempt to mitigate this problem in different ways, including by measuring attitude stability and consistency.42 I think it is reasonable to be optimistic about these attempts, and thus reasonable to be optimistic about the possibility of designing and administering surveys that measure the strength of mandate reasons. If I am wrong about this, then while the argument defending the existence of mandate reasons in Section III still holds, there is good reason to be sceptical about attempts to determine their strength and content without increasing the use of referenda, and thus good reason to be sceptical about their practical relevance to representatives in existing democracies.

Third, if it is possible to determine a representative’s mandate reasons through survey data, then representatives, political parties, and other relevant political actors have a moral duty to try to do so. This duty is grounded in the moral importance of mandate reasons themselves. If reliable survey data is already available, representatives and parties have duties to seek it out and use it to determine their mandate reasons. If such data is not already available, they have an additional duty to support the creation and administration of reliable surveys. Mandate reasons thus ground two sorts of demands in a representative democracy. There are first-order demands to implement the proposals that voters support. And there are higher-order, epistemic demands to determine what proposals voters support.

VII. FURTHER QUESTIONS

My aim in this article has been to examine foundational questions about the nature and grounds of mandate reasons that have been ignored in the literature. There are several remaining questions about the implications of the account. In this final section, I identify three important kinds of questions. The list is not meant to be exhaustive.

First, there are questions about whether it is better from the standpoint of democratic legitimacy for representatives to have stronger mandate reasons rather than weaker mandate reasons and, if so, whether this justifies compulsory voting. Perhaps democracies that consistently produce stronger mandate reasons are morally better than democracies that do not. If they are, does this provide an argument for legally requiring eligible citizens to vote? Any such argument must contend with the moral costs of compulsory voting, including the potential cost of reducing voters’ average competence.

Second, representatives might receive widespread support for their proposals from voters in their district, but the district might contain only a small share of an overall electorate that overwhelmingly opposes the policies. It is natural to think that there can be individual mandate reasons for representatives given by voters in their districts and collective mandate reasons for political parties given by whole electorates. How should representatives understand the balance of mandate reasons when voter support for certain proposals within their district deviates from voter support for the proposals within the broader electorate?

Third, there are questions about how to interpret mandate reasons when split-ticket voting is and isn’t available. In elections in the UK and other Westminster systems, for example, voters lack the option to vote for a local MP candidate without voting for their party’s candidate for Prime Minister. When there is strong coordination within political parties, MP and PM candidates generally endorse roughly the same platform. In many democracies, however, partisans share a general political orientation but advance quite different proposals—there are significant differences between the progressive, moderate, and conservative wings of the Labour Party in the UK or the Democratic Party in the US, for example. However, there are also challenges in interpreting mandate reasons that arise from the possibility of divided government when split-ticket voting is possible. Executives often belong to a different party than the party that controls the legislature. This can lead to cases where heads of government have strong mandate reasons to implement proposals that the majority party in the legislature has strong mandate reasons to oppose, and vice versa.

VIII. CONCLUSION

I have tried to shed light on the neglected topic of how voter support affects how representatives ought to govern. I defended the existence of mandate reasons characterised by the dependence thesis, which says that votes strengthen mandate reasons to govern in specific ways when and only when voters express supportive attitudes for governing in those ways. If the dependence thesis is true, then having stronger mandate reasons can make a difference to how a representative ought to govern. However, the dependence thesis also implies that mandate reasons have a more limited normative influence and are more difficult to establish than is commonly presumed.

Notes

  1. See e.g. Christiano 2008; Estlund 2008; Stilz 2009; Viehoff 2014. [^]
  2. McGraw and Allison 2024. [^]
  3. The Economic Times 2019. [^]
  4. Epstein 2020. [^]
  5. Georgiadis et al. 2019. [^]
  6. For an influential sceptical view, see Dahl 1990. [^]
  7. Guerrero 2010. [^]
  8. Ibid., pp. 275–77. [^]
  9. Ibid., p. 283. The language of trustee and delegate is borrowed from Pitkin 1967. [^]
  10. Guerrero 2010, p. 285. [^]
  11. Ibid., p. 287. [^]
  12. For a systematic defence of this view, see Christiano 2008. [^]
  13. For more on the nature and grounds of normative powers see: Owens 2014; Tadros 2020. [^]
  14. See Kolodny 2023, ch. 27. [^]
  15. For an interesting exploration of the relationship between democratic equality and representation, see Lovett 2021. [^]
  16. For more on the relationship between democratic self-rule and moral control of the general aims of society, see Christiano 1996. [^]
  17. I am thinking here of referenda and ballot measures that appear alongside the choice of representative on the ballot. [^]
  18. For contemporary defences of this sort of view see: Christiano 2008, chs. 1, 2, and 6; Wilson 2020; Waldron 1999, ch. 5. [^]
  19. I think the same basic argument applies to relational egalitarian views that understand the value of democracy not in terms of equal respect for judgment, but in terms of equal power or influence in political relationships (see e.g. Viehoff 2014; Kolodny 2023). Individuals relate as equals in the process democratic self-rule only if they have equal power to determine which political ends their society ought to pursue, and this requires equal control over whether their vote strengthens a representative’s mandate reasons to implement their proposals. [^]
  20. Beerbohm 2016, p. 399. [^]
  21. For an illuminating discussion of suspended contexts, see Shiffrin 2014, pp. 16–20. [^]
  22. There is a large literature demonstrating this so-called programme-to-policy linkage. For a study of over 20,000 pledges made in 57 election campaigns in 12 countries, see Thomson et al. 2017. [^]
  23. Of course, in real-world democracies, representatives are not equally responsive to the preferences of voters. [^]
  24. Different theories rely on different models to explain this tendency. I focus on Condorcet’s Jury Theorem because it is the most well-known model and because it can straightforwardly explain the structural features of mandates reasons. Other prominent epistemic theories rely on the ‘miracle of aggregation’ (see Page and Shapiro 1992) and the ‘diversity trumps ability’ theorem (see Landemore 2012). [^]
  25. For a systematic exploration of the conditions of CJT and its relationship to democracy, see Part I of Goodin and Spiekermann 2019. [^]
  26. See ibid. for the most sophisticated attempt to defend the applicability of CJT to democracy in light of the debates about whether its assumptions hold. [^]
  27. See text accompanying footnotes 2–5 above for examples of mandate claims. [^]
  28. For an overview, see Somin 2016, ch. 1. [^]
  29. This is famously demonstrated by Campbell et. al. 1960. See also Lewis-Beck et. al. 2008. For a nuanced philosophical discussion, see Lovett 2022. [^]
  30. See Carpini and Keeter 1996. [^]
  31. The prevalence of retrospective and group voting is a central theme of Achen and Bartels 2016. [^]
  32. This is one way of understanding the implications of Ansolabehere, Rodden, and Snyder 2008. [^]
  33. These proposals are extracted from Reform UK’s (2025) political platform entitled ‘Our Contract with You’. [^]
  34. An important exploration of this issue in the context of a general theory of promissory obligation can be found in Scanlon 1998, ch. 7. [^]
  35. Important contributions to this debate include: Dougherty 2021; Tadros 2022. [^]
  36. This is an important theme of Campbell et. al. 1960. See also: Popkin 1991; Hayes 2005; Lau and Redlawsk 2001. [^]
  37. For an important exploration of the use of referenda in representative democracies, see Budge 1996. Although referenda provide the most obvious and direct method for determining mandate reasons, citizens’ assemblies or other empowered deliberative fora might also do so. [^]
  38. For example, the Brexit referendum was arguably not well-crafted because the general binary question posed—’Should the United Kingdom remain a member of the European Union or leave the European Union?’—obscured deeply impactful differences and complexities between different ways of ‘remaining’ or ‘leaving’ the EU. [^]
  39. For an overview of survey methods in democratic theory, see Gastil 2022. [^]
  40. For a comprehensive analysis of potential sources of error, see Weisberg 2005. [^]
  41. Converse 1964/2006. For an overview of issues related to the nonattiude problem, see Saris and Sniderman 2004. [^]
  42. See Ibid. [^]

ACKNOWLEDGMENTS

For very helpful feedback and suggestions, I am grateful to Sanjiv Bajaj, Jacob Barrett, Paul Billingham, Simon Caney, Tom Christiano, Matthew Clayton, Annabel Dallen, Nadine Elzein, Ben Ferguson, Simon Gansinger, the late Jerry Gaus, Michele Giavazzi, Jeff Howard, Gah-Kai Leung, Ricky Li, Kida Lin, Guy Longworth, Yael Lowenstein, Andy Mason, David Miller, Chris Mills, Zeynep Pamuk, Tom Parr, Fabienne Peter, Josh Pike, David Rischel, Lewis Ross, Cain Shelley, Adam Slavny, Victor Tadros, Patrick Tomlin, Kartik Upadhyaya, Amber Kavka-Warren, Katy Wells, anonymous referees for this journal and the editor Robert Goodin, and audiences at the Legal Philosophy Workshop, the From the Individual to the Public workshop, the Nuffield Political Theory workshop, and the Warwick philosophy colloquium.

COMPETING INTERESTS

The author declares that he has no competing interests.

REFERENCES

Achen, Christopher and Larry Bartels. 2016. Democracy for Realists: Why Elections Do Not Produce Responsive Government. Princeton, NJ: Princeton University Press.

Ansolabehere, Stephen; Jonathan Rodden; and James M. Snyder. 2008. The strength of issues: using multiple measures to gauge preference stability, ideological constraint, and issue voting. American Political Science Review, 102 (2): 215–32.  http://doi.org/10.1017/S0003055408080210

Beerbohm, Eric. 2016. The ethics of electioneering. Journal of Political Philosophy, 24 (4): 381–405.  http://doi.org/10.1111/jopp.12082

Budge, Ian. 1996. The New Challenge of Direct Democracy. Cambridge: Polity Press.

Campbell, Angus; Phillip Converse; Warren E. Miller; and Donald E. Stokes. 1960. The American Voter. Chicago: University of Chicago Press.

Carpini, Michael X. Delli and Scott Keeter. 1996. What Americans Know about Politics and Why It Matters. New Haven, CT: Yale University Press.

Christiano, Thomas. 1996. The Rule of the Many. Boulder, CO: Westview Press.

Christiano, Thomas. 2008. The Constitution of Equality: Democratic Authority and Its Limits. Oxford: Oxford University Press.

Converse, Philip E. 1964/2006. The nature of belief systems in mass publics. Critical Review 18 (1–3): 1–74; originally published 1964.  http://doi.org/10.1080/08913810608443650

Dahl, Robert. 1990. The myth of the presidential mandate. Political Science Quarterly, 105 (3): 355–372.  http://doi.org/10.2307/2150822

Dougherty, Tom. 2021. The Scope of Consent. Oxford: Oxford University Press.

Epstein, Jennifer. 2020. Biden has ‘clear mandate’, aides say as Electoral College votes. Bloomberg, 14 Dec. https://www.bloomberg.com/news/articles/2020-12-14/biden-has-clear-mandate-aides-say-as-electoral-college-votes

Estlund, David. 2008. Democratic Authority. Princeton, NJ: Princeton University Press.

Gastil, John. 2022. Survey methods. Pp. 204–217 in Research Methods in Deliberative Democracy, ed. Selen A. Ercan, Hans Asenbaum, Nicole Curato, and Ricardo F. Mendonça. Oxford: Oxford University Press.

Georgiadis, Philip; Myles McCormick; Sarah Provan; Charlotte Middlehurst; Adam Samson; and Josh Noble. 2019. UK election results: Boris Johnson says he has ‘stonking mandate’ on Brexit — as it happened. Financial Times, 13 Dec. https://www.ft.com/content/470b4c60-706e-3a62-927e-40ce5ed6f97a

Goodin, Robert E. and Kai Spiekermann. 2019. An Epistemic Theory of Democracy. Oxford: Oxford University Press.

Guerrero, Alex. 2010. The paradox of voting and the ethics of political representation. Philosophy and Public Affairs, 38 (2): 272–306.  http://doi.org/10.1111/j.1088-4963.2010.01188.x

Hayes, Danny. 2005. Candidate qualities through a partisan lens: a theory of trait ownership. American Journal of Political Science, 49(4): 908–923.  http://doi.org/10.1111/j.1540-5907.2005.00163.x

Kolodny, Niko. 2023. The Pecking Order: Social Hierarchy as Philosophical Problem. Oxford: Oxford University Press.

Landemore, Hélène. 2012. Democratic Reason. Princeton, NJ: Princeton University Press.

Lau, Richard R. and David P. Redlawsk. 2001. Advantages and disadvantages of cognitive heuristics in political decision making. American Journal of Political Science, 45(4): 951–971.  http://doi.org/10.2307/2669334

Lewis-Beck, Michael S.; William F. Jacoby; Helmut Norpoth; and Herbert F. Weisberg. 2008. The American Voter Revisited. Ann Arbor: University of Michigan Press.

Lovett, Adam. 2021. Must egalitarians condemn representative democracy? Social Theory and Practice, 47 (1): pp. 171–198.  http://doi.org/10.5840/soctheorpract202121116

Lovett, Adam. 2022. Voter motivation. Journal of Ethics and Social Philosophy, 21(3): 299–308.  http://doi.org/10.26556/jesp.v21i3.1255

McGraw, Meridith and Natalie Allison. 2024. Trump claims ‘unprecedented and powerful mandate’. Politico, 6 Nov. https://www.politico.com/news/2024/11/06/trump-cusp-victory-lap-00187777

Owens, David. 2014. Shaping the Normative Landscape. Oxford: Oxford University Press.

Page, Benjamin I. and Robert Y. Shapiro. 1992. The Rational Public. Chicago: University of Chicago Press.

Pitkin, Hannah. 1967. The Concept of Representation. Berkeley: University of California Press.

Popkin, Samuel L. 1991. The Reasoning Voter: Communication and Persuasion in Presidential Campaigns. Chicago: University of Chicago Press.

Reform UK. 2025. Our Contract with You. https://assets.nationbuilder.com/reformuk/pages/253/attachments/original/1718625371/Reform_UK_Our_ContrCon_with_You.pdf?1718625371 (accessed Jan. 2025).

Saris, Willem E. and Paul M. Sniderman, eds. 2004. Studies in Public Opinion: Attitudes, Nonattitudes, Measurement Error, and Change. Princeton, NJ: Princeton University Press.

Scanlon, T.M. 1998. What We Owe to Each Other. Cambridge, MA: Harvard University Press.

Shiffrin, Seana. 2014. Speech Matters: On Lying, Morality, and the Law. Princeton, NJ: Princeton University Press.

Somin, Ilya. 2016. Democracy and Political Ignorance: Why Smaller Government Is Smarter. 2nd ed. Stanford, CA: Stanford University Press.

Stilz, Anna. 2009. Liberal Loyalty: Freedom, Obligation, and the State. Princeton, NJ: Princeton University Press.

Tadros, Victor. 2020. Appropriate normative powers. Aristotelian Society Supplementary Volume, 94 (1): 301–326.  http://doi.org/10.1093/arisup/akaa009

Tadros, Victor. 2022. Beyond the scope of consent. Philosophy and Public Affairs, 50 (4): 430–466.  http://doi.org/10.1111/papa.12222

The Economic Times. 2019. People gave resounding mandate for building new India: PM Modi. 23 Aug. https://economictimes.indiatimes.com/news/politics-and-nation/people-gave-resounding-mandate-for-building-new-india-pm-modi/articleshow/70803589.cms.

Thomson, Robert; Terry Royed; Elin Naurin; Joaquín Artés; Rory Costello; Laurenz Ennser-Jedenastik; Mark Ferguson; Petia Kostadinova; Catherine Moury; François Pétry; and Katrin Praprotnik. 2017. The fulfilment of parties’ election pledges: a comparative study on the impact of power sharing. American Journal of Political Science, 61 (3): 527–542.  http://doi.org/10.1111/ajps.12313

Viehoff, Daniel. 2014. Democratic equality and political authority. Philosophy and Public Affairs, 42 (4): pp. 337—75.  http://doi.org/10.1111/papa.12036

Waldron, Jeremy. 1999. Law and Disagreement. Oxford: Oxford University Press.

Weisberg, Herbert F. 2005. The Total Survey Error Approach: A Guide to the New Science of Survey Research. Chicago: University of Chicago Press.

Wilson, James. 2020. Democratic Equality. Princeton, NJ: Princeton University Press.

Share

Downloads

Information

Metrics

  • Views: 218
  • Downloads: 50

Citation

Download RIS Download BibTeX

File Checksums

(MD5)
  • PDF: 53bcf18d196e899838f940d58939a62e
  • XML: 9175e1f1d6b92c69c7dbcb90540170f6

Table of Contents