1.

It’s an honour and a challenge to respond to a comment on my engagement with democratic theory, from the person who wrote the book on political equality, Chuck Beitz. It’s gratifying, if also daunting, that he should take what I say in the abstract seriously enough to ask how much of it might survive the collision with matters in the concrete. Beitz concentrates:

Let me take up the second of the two questions first, about whether the asymmetries associated with a representative system might answer to values other than tempering inferiority. As Beitz observes, my stance is already pluralistic. In addition to claims against inferiority, there are also rights against invasion and claims to improvement. So, it remains open to me to grant that representative institutions reflect a compromise between noninferiority and other values. The idea would be, in other words, that the move from direct to representative democracy involves some cost in terms of inferiority, but this is compensated for by the gain in noninvasion and improvement, as well as perhaps still other values. Indeed, the move from direct to representative democracy, while involving a cost in terms of inferiority along one dimension, may alleviate inferiority along another dimension, such as if representative democracy leads to better protections against discrimination.

Are there other values, beyond noninvasion and improvement, that countervail against noninferiority, such that representative institutions can be seen as a compromise between noninferiority and those other values? Beitz seems to suggest that there are. He cites ‘competitive fairness’ as an instance. However, it is unclear to me that competitive fairness is an independent value. This may be in part because I struggle to identify a standard of competitive fairness that doesn’t boil down to equal opportunity for influence, which I in turn take to be rooted in noninferiority. Talk of unfairness suggests that something is not being distributed to relevant stakeholders according to the appropriate standard. What is being distributed, among which stakeholders, and according to what standard, if not opportunities for influence, among individuals, equally? Suppose one grants that everyone had equal opportunity to influence the result. There was a range of parties representing different ideological points of view. Everyone had an equally weighted vote, the same airtime to press their case and convince others, under the assumption that a majority in the legislature would decide on policy. Is there still room to say, ‘yes, but nevertheless the competition was unfair’? To whom, and in what respect?

It is in part this difficulty – namely, in discerning other values, beyond noninvasion and improvement, that countervail against noninferiority, such that representative institutions can be seen as a compromise between noninferiority and those other values – that makes me wary, as Beitz notes, of talk of ‘representation’. Such talk often seems to suggest that there are special values that are realized only or best by representation. For instance, it might be said that is important in itself that legislation be the outcome of deliberation, which is not possible outside of a sufficiently small deliberative body. Of course, it is possible that legislation that is the result of deliberation will lead to substantively better results: both because it has been thought about more, but also because the legislative record provides guidance that would otherwise be lacking in the mere wording of the statute. But imagine that plebiscites would lead to equally substantive good results. And imagine that there was then vigorous discussion of the merits of the decisions after the fact, so that people understood why the right answers were the right answers and their faculties were engaged. Would something be lost? Does the fact that the decision has a deliberative pedigree matter in itself? Why? Because a collective is self-governing only if it explicitly deliberates? But why should this be? Is an individual, by analogy, self-governing only if they explicitly deliberate? What about the person of practical wisdom who sees what the right thing to do is and does it without explicit deliberation?

Set aside the question of whether there are other values besides noninvasion and improvement that countervail against noninferiority, such that representative institutions can be seen as a compromise between noninferiority and those other values. There remains the point that at least those values – noninvasion and improvement – countervail against noninferiority. If representative democracy necessarily involves some cost in terms of inferiority, and if representative democracy is, in principle, unavoidable, given the intolerable cost to those other values, then I may need to retreat from some remarks made in the book. I have in mind those where I suggest that whereas noninferiority might, in principle if not in practice, be realized, self-rule cannot be. Nevertheless, the main structural point stands: if we see our ideal as being noninferiority, rather than self-rule, then we can come closer to meeting the demands of our ideal, simply because our ideal is less demanding.

2.

This leads to the other of Beitz’s two main questions. How high is the price in inferiority that must be paid for the benefits, say, in improvement, of representative institutions? Beitz suggests that the price is higher than I might suppose. This is because my stylized discussion overlooks features of representative institutions that make them even less tempered than I imagine. Here I wish at least to question whether Beitz doesn’t overstate the case.

To begin with, Beitz suggests that because I overlook two important facts about legislative procedure, I overlook ways in which a representative legislature fails to realize equal opportunity for influence. The first fact about legislative procedure that I overlook is that ‘if hierarchy in the legislature enables some representatives to exercise more influence than others, the underlying equal influence of their constituents may be subverted’.

For one thing, this does not seem to me to be a necessary feature of representative democracy. Here, as elsewhere in his comment, it seems to me that Beitz may tend to conflate representative democracy with American institutions. It is hard to see how the problem that Beitz describes arises under a party-list system. In such a system, there is no representative that counts as mine; there are only the representatives of the party that I support.

Moreover, even in a parliamentary system with representatives elected by districts, party members vote to determine the leadership of the party, who will occupy which hierarchical positions within the legislature. If the leadership is responsive to the aims of the party as a whole, rather than somehow specially to preferences of their local constituency, as is arguably the case, and as one might expect leadership would need to be in order to retain their position, then it is not clear how the underlying equal influence of their constituents is subverted.

Things may be different in a district system in which key roles in the legislature are determined, say, by seniority. Then it may be that whereas voters in a district with a senior representative have a say over policy that voters in a district with a junior representative do not. And even here there is the question of whether the representative with seniority is responsive to their particular constituency, or the aims of the party as a whole, a party which members of all constituencies had equal influence in putting into power. But if such a seniority system introduces inequalities, why isn’t it to that extent regrettable? In any event, it isn’t an inevitable consequence of a representative system.

The second fact of legislative procedure that Beitz claims I overlook is ‘the fact that bills emerge from intra- and interparty negotiations’. This ‘means that they may not present a question to which a representative’s response might have been anticipated by voters when the representative stood for election, so it is implausible that voters can realistically regard their votes as exercising directive control’. I agree that if directive control requires a series of conditional directives of the form, ‘If you find yourself in precisely this negotiating position, then do X’, then voters could not realistically be understood as issuing such directives. And let me grant that the entirely open-ended, ‘“Vote as you think best” might be a directive in name only’. But it seems to me that there is a broad category of directives in between these extremes, including directives of the form: ‘Pursue these objectives, with the understanding that negotiation may require compromising some of these objectives for the sake of others’.

Beitz offers the following illustration:

I am sceptical of Beitz’s scepticism, at least with respect to this example. Why wasn’t the directive something like as follows? ‘We have the policy aims of expanding healthcare coverage in a way that reduces costs, as we see is possible from the examples of healthcare systems in other countries. Work towards healthcare legislation that goes as far as possible, given the need for negotiation, toward those policy aims’. I personally voted for Democrats in that election, and I don’t think that I understood the relevant policy stance as, ‘A public option or nothing at all! No movement in the direction of expanded coverage, a more efficient system of payments, etc., unless my ideal outcome is realized!’ Moreover, insofar as the choice was between a Democrat and a Republican in 2008, wasn’t it clear, simply based on party affiliation, which candidate would be more likely to pursue such an objective, and which would instead concentrate on tax cuts for the wealthy, as was the Republicans’ legislative priority in 2017 and 2025? It seems worth adding that the very focus on healthcare as the paramount issue to address in the short time in which the Democrats held a filibuster-proof majority in the Senate, or so it seemed to me, was the result of its prominence in intraparty debates in the primaries.

Of course, there is a difficult question of whether that is enough directive control. But the point is that there is something in between the extremes of explicit conditional directives for every contingency and the completely open-ended, nominal directive, do whatever you think best. There are more general priorities – e.g., expanded healthcare, rather than tax cuts for the wealthy – and a direction in which you want policy to move. To be sure, many people voted in the election with no clear idea of what aims the parties were likely to pursue in office. But again what matters, on my account, is that they had the opportunity to vote with an idea of what aims the parties were likely to pursue in office.

3.

The second feature of representative institutions that Beitz claims that I overlook is the distinction between legislative and the executive. ‘Intuitively’, he suggests, ‘we should want [executive] implementers to be guided in their interpretation of legislative directives by the aims of the legislature in enacting the directive, insofar as they can be known (for example, from the legislative history) rather than by their own, freestanding interpretations of the public good’. I find this puzzling, since I would have thought that my account offers a straightforward explanation of this. Executive implementers should be guided by the aims of the legislature, rather than their own, freestanding interpretations of the public good, out of respect for the value of equal opportunity for influence. What other value does Beitz thinks accounts for this? He doesn’t say.

What Beitz says about administrative agencies puzzles me for similar reasons. He writes: ‘It is not even clear that equality of influence for voters should have any bearing on the conduct of agency implementers… The regulative concern would not be voters’ equality of opportunity for influence but rather adequate deliberative contestation among otherwise unaccountable executive officials’. But are what are the terms of the deliberative contest, and why is important that there be a contest? Is the thought that agency implementers are putting forward their own freestanding views of the public interest, and that deliberative contestation is a way of insuring that the best such view prevails? But in the previous paragraph, Beitz writes, to the contrary, that ‘agency officials are instructed to exercise discretion to implement the law in whatever way they judge is most consistent with the law as written and interpreted through the legislative record provided’. So it would seem that deliberative contestation is a way of insuring that the discretion exercised is consistent with the law as written and interpreted through the legislative record provided. But why is it important to hew to the law and to the legislative record, rather than their own freestanding view, even if correct, of what would serve the public interest? One answer is that it is because people had equal opportunity to influence the law, whereas they did not have equal opportunity to influence the freestanding views of agency officials. If so, then the ultimate ‘regulative concern’ is indeed voters’ equal opportunity for influence, and ultimately equality of influence for voters should have ‘bearing’ on the conduct of agency implementers. Wouldn’t we want an agency official who is tempted to depart from the law to advance a freestanding view of the public interest, to think, ‘I wasn’t elected to make the law, but appointed (or hired) to execute it. If I think the law should be changed, then I should lobby for that change or stand for election myself’?

Or is Beitz’s worry that because the executive is independently elected, they have a title to rule on their own mandate without regard for the legislature? Again to assume that the executive is independently elected might be to focus on American institutions. In a parliamentary system, there is not this difference between the legislative and executive. In any event, even in the American system, the executive was independently elected to implement the law, not to make it.

When it comes to legislative oversight of the executive, Beitz writes, ‘there does not appear to be any informative sense in which we can say that they have duties to their constituents to act as they led voters to expect, because it’s overwhelmingly unlikely that they led voters to expect that they would behave in any particular way when exercising the role of overseer’. Is it, I wonder, so unrealistic to think that voters are in a position to know what stance a particular party will take with regard to oversight of the executive? Consider the present situation in the United States as an example. One of the reasons that, at the time of writing this, I am so desperate to see Democrats win back legislative control of at least one of the chambers in the 2026 mid-term elections is the expectation that it would lead to tighter oversight of a Republican executive that seems to be out of control. Among other things, there is the Justice Department’s defiance of the law in the release the Epstein papers and its politically motivated investigations and indictments, as well as the policy and conduct of the Department of Homeland Security’s Immigration and Customs Enforcement. And who knows what further misbehaviour the executive will have gotten up to by the time this response is published? This seems to me an eminently reasonable expectation. But on Beitz’s telling, my vote for Democrats is a shot in the dark, its being ‘overwhelmingly unlikely’ that I have any reason ‘to expect that they would behave in any particular way when exercising the role of overseer’.

4.

The final feature of representative institutions that Beitz says I overlook is the distinctive role of parties, which I do by way of a ‘conflation of the roles of candidates and parties’.

To begin with, there may again be a tendency to focus on American institutions. In a party-list system, a candidate doesn’t owe responsibilities to a constituency, since there is no constituency. And it seems that we can imagine electoral systems without parties, such as perhaps the elections of consuls in the early Roman republic. In any event, it seems to me that the relevant, general distinction here is between being in office and seeking office: between an individual or organization – an elected official or government – who or that wields the powers of office and an individual or organization who or that competes to attain those powers of office by winning elections – a candidate or party. Hence my tendency in the book to speak of ‘candidates and parties’ in the same breath. In the passage quoted above, it seems to me that Beitz is unhelpfully opposing an elected official – not a candidate in competition for office, but a former candidate who has now attained office – with a party – an organization in competition for office, which conflates the less important distinction between an individual and organization, on the one hand, with the more important distinction between being in office and contending for office, on the other.

How, then, should a candidate or party behave when contending for office, as opposed to how an elected official or government should behave in office? It is a difficult question. But, appealing just to the materials on offer in the book, I would imagine that the candidate or party should seek to gain office in a way that is compatible with equal opportunity for influence of all, with the aim of pursuing, if office is gained, some bona fide conception of what is in the public interest. How can the candidate or party do this? Principally, by persuading voters that their or its conception of what is in the public interest is correct, standing ready to advance that conception if and when they are elected. What the candidate or party shouldn’t do is, say, lie to other candidates’ or parties’ likely voters about the date of the election, or intimidate them from going to the polls. For reasons that I give in the book, if I simply convince more voters than you do, I have not undermined your equal opportunity for influence. By contrast, if I successfully deceive you about the date of the election, then I have undermined your equal opportunity for influence. There are difficult cases, such as offering rides to the polls to your voters, but not to other voters (as pointed out to me by Daniel Wodak). But it still seems to me that the constraint that one seek office in a way that preserves everyone’s equal opportunity for influence explains much of the distinction between acceptable and unacceptable ways of gaining office.

Beitz’s main example of a violation of the rules of contending for office is as follows:

Again, it seems worth noting, Beitz is focusing on American-style institutions, here in the form of mostly first-past-the-post, single-member districts. In any event, Beitz seems to be suggesting that, while there is a clear objection to such collusion, we cannot explain what the objection is in terms of the materials I have on offer, and so I need to appeal to a further, independent value of competitive fairness to explain it. As I explained earlier, it isn’t clear to me how to explain the concern about competitive fairness in a way that doesn’t bottom out in a concern about equal opportunity for influence in any event. But set that aside. Why doesn’t the objection sound directly in the value of equal opportunity for influence? Why isn’t the objection straightforwardly that the major parties are preventing those who might support fusion candidates from having the same say as those who support the two major parties? Suppose that the two major parties were simply to throw out the ballots of anyone who voted for a third party. Wouldn’t that plainly be a violation of equal opportunity for influence? Things would be different if the two major parties were to persuade those who might support a third party to vote for them instead.

5.

I turn now to Pamela Hieronymi’s superbly penetrating comment. Imagine a hypothetical state that meets interests in improvement in a way that is fair to all and, importantly, that will continue to do so. It seems to me a commonplace in contemporary political philosophy that there is a further complaint against such a state, which calls for it to be ‘justified’ or ‘legitimated’. This further complaint would have to be met by some ‘legitimating condition’ – such as that those subject to the state consent to it or can reasonably accept it – or some ‘limit of legitimacy’ to what the state does – that the state is minimal or that it respects John Stuart Mill’s harm principle. And, in an effort to substantiate the idea that it is a commonplace claim, I give a list of philosophers – not only libertarians such as A. John Simmons and Robert Nozick, but also liberals such as John Rawls and Ronald Dworkin, as well as philosophers otherwise wary of liberal pieties, such as Joseph Raz and Bernard Williams – who insist that there is some such further complaint against the state. I then argue against the idea that the further complaint against the state has to do with coercion – whether understood as force or threat. I conjecture instead that it has to do with the hierarchical relations between those whose decisions the state’s decisions are and those who are subject to those decisions.

As I read Hieronymi, she thinks that this commonplace claim is just mistaken. The only worry about the state, in her view, is that it will not continue to meet interests in improvement in a way that is fair to all. If we really could be assured that the state would continue to meet interests in improvement in a way that was fair to all, then there would be no further complaint against the state. There is no problem with the benevolent despot, so long as we can be assured that the benevolent despot will in fact continue to be benevolent.

So, Hieronymi is in effect denying what I take to be a commonplace claim: that even if the state will continue to treat us well, there is some further complaint calling for the state to be justified or legitimated. She might well be right in denying this allegedly commonplace claim, and what she says gives me pause. But it is hard to know how to argue that I offer the right explanation of something against the view that there is nothing there to explain. I can lamely report that there seems to me a further complaint against the state. And I can cowardly take shelter in the observation that many others – including the rather august list I enumerated in the paragraph before last – seem to think that there is a further complaint against the state. But again what she says gives me pause.

6.

As Hieronymi notes, I put forward several ‘vertical’ tempering factors which are meant to address the otherwise hierarchical relationships between those whose decisions the state’s decisions are and those who are subject to those decisions. For instance, there is Impersonal Justification and Least Discretion. As explained in the précis, the point of Impersonal Justification and Least Discretion is to make it possible for subjection to the superior power and authority of an office to be something distinct from subjection to the superior power and authority of the natural person who occupies the office.

Hieronymi suggests, however, that insofar as tempering factors such as Impersonal Justification and Least Discretion assuage our concern about the state, they do so not by addressing concerns about interpersonal hierarchy. Instead, what’s comforting about the tempering factors is the thought is that if the state realizes the tempering factors today, then the state will not treat us badly tomorrow – again, as I would put it, in the sense of failing to meet interests in improvement in a way that is fair to all.

It is true, I suppose, that a state all of whose officials abided by Impersonal Justification and Least Discretion today would meet the interests in improvement of all in a fair way today. However, at a certain level of abstraction, there is a puzzle in thinking that what’s comforting about the tempering factors being realized today is that they keep the state from treating us badly tomorrow. If the worry is that the state, although it’s treating us well today, might treat us badly tomorrow, then why isn’t there a similar worry that the state, although it realizes the tempering factors today, might stop realizing the tempering factors tomorrow? If the state stops realizing the tempering factors tomorrow, then there is no reason to think that it will treat us well tomorrow.

Things would be different if the tempering factors were somehow external constraints, independent of the willingness of people to decide in certain ways, that once in place today prevented the state from treating us badly tomorrow. But tempering factors such as Impersonal Justification and Least Discretion are realized only so long as officials in fact decide in certain ways. If tomorrow officials should start to decide in other ways, then Impersonal Justification and Least Discretion will no longer be realized. Of course, we might imagine that other people would keep the officials in line, but this just pushes the problem back. What keeps those other people in line? In sum, I don’t think that there are any external constraints that once in place today prevent the state from treating us badly tomorrow. There is no ‘red tape’, as Hieronymi puts it, of that kind.

This is one of the ways in which I part company from republican views, which cling to the idea that there is a way to keep the state in line that does not depend ultimately on the willingness of people to decide in certain ways. The republican idea is roughly that so long as it is possible for another agent, individual or collective, to treat you badly, then you are dominated. I think that so long as it is actual that the natural person who occupies some office decides your case in accordance with Impersonal Justification and Least Discretion, then, to some extent, you do not stand in a relation of inferiority to that natural person. The fact that it is possible for them to decide your case differently does not matter.

The republican may reply that so long as that other person decides your case in the right way but only because it pleases them, or because you have curried favour with them, it hardly seems that you have avoided a relation of inferiority to that natural person. I would agree. But this is because someone who decides your case in the right way only because it pleases them or because you have curried favor with them is not in fact deciding your case on the basis of impersonal reasons. They are not, in the actual world, satisfying Impersonal Justification and Least Discretion. The significance of the expression ‘only because it pleases them’ is not that there is a counterfactual world in which because it does not please them they treat you badly, but instead that in the actual world they are not satisfying Impersonal Justification and Least Discretion – they are not maintaining the right sort of separation between the office and the natural person who occupies it.