Democracy is often defined as the form of government in which citizens, save for appropriate exceptions, such as minors, have an equal say in making at least the most fundamental political decisions (Buchanan 2002, 710; Christiano 2008, 9; Anderson 2009, 214; Kolodny 2014a, 197; Viehoff 2014, 337; Goldman 2015, 236). What does this mean precisely? Having an equal say must mean more than universal franchise, i.e., the equal formal right to vote. Before the Voting Rights Act of 1965 African Americans in southern US states had the formal right to vote, but arguably did not have an equal say. Electoral autocracies or hybrid regimes also often grant the equal formal right to vote without letting most citizens have a genuine say in politics (Morlino 2021). Some argue that universal franchise is not only insufficient, but also unnecessary for having an equal say; think of recent proposals for proportional voting (Brighouse and Fleurbaey 2010) or enfranchisement lottery (López-Guerra 2010). Perhaps citizens should not have any say at all, and we should opt for a purely sortition-based system (Guerrero 2014). All these proposals are said to be compatible with the democratic ideal. Whether they are indeed, can only be determined once it is clarified what it means to have an equal say.

The goal of this paper is to offer such clarification, i.e., to identify the most plausible understanding of having an equal say compatible with common reasonable views about democracy. In doing so, I will seek to remain neutral about the value or justification of democracy, a topic I discussed elsewhere (Kapelner 2022), and focus only on the definitional question of what having an equal say is. One may argue, however, that drawing such a sharp distinction between meaning and value in political philosophy is impossible (Dworkin 2004). Indeed, one’s judgements about the value of democracy likely influence what common views about it one accepts as reasonable, which, in turn, affects what definitions of having an equal say one finds plausible. However, my argument seems to be compatible with most views about the value of democracy, whether instrumentalist (Arneson 2009), liberal egalitarian (Christiano 2008; Wilson 2019), or republican (Pettit 2012). It even seems to be compatible with the view that democracy is ultimately not a good thing (Brennan 2016). If so, then the question of the value of democracy can safely be bracketed for the purpose of this discussion.

In this paper I argue that having an equal say means having a say as an equal. To explain and defend this view, I begin in Sects. 1 and 2 by discussing two well-known accounts of the meaning of having an equal say, i.e., equal power views, and J. L. Wilson’s (2019) appropriate consideration view. While they both have strengths, I show them to be ultimately incomplete and in need of improvement. In Sects. 3 and 4 I defend my own view, according to which one has a say when one occupies a specific social role, i.e., the public office of the democratic citizen, in which can make constitutive input in a collective political decision-making process, and one has an equal say if one is an equal decision-maker, i.e., no other decision-maker is one’s superior qua decision-maker; they all have a say as equals.

1 Equal Power

A rather intuitive and widely shared view is that having an equal say, in the sense relevant to democracy, simply means having equal power to influence political outcomes in some way. There are different views on how the relevant concepts of power and equality should be understood. For example, Brighouse (1996) defines them in terms of the equal availability of political influence among individuals, where the

“amount of influence available to them is given by the probability we would assign to their getting their way, if they and everyone else engaged in political activity, and we knew nothing of what any other citizens wanted. Equal influence would be available if for everyone we would assign the same probability to their getting their way, assuming that they and everyone else engaged, and knowing nothing of what the other citizens wanted.” (Brighouse 1996, 119)

Goldman (1999, 2015), and following him, Kolodny (2014a, b, 2023a) define the relevant power in terms of equal contributory causal power measured as equal a priori decisiveness, where each citizens’ choice has an equal chance of being decisive assuming that all patterns of others’ choices have equal probability. Dworkin (2011) defines the relevant form of causal power as “impact,” a concept that contrasts with “influence” where the two are defined as follows: “A person’s influence includes his power to persuade or induce others to his side; his impact is limited to what he can achieve through his own opinion without regard to what others believe.” (Dworkin 2011, 388) People have an equal say when “no adult citizen’s political impact is less than that of any other citizen for reasons that compromise his dignity— reasons that treat his life as of less concern or his opinions as less worthy of respect.” (Dworkin 2011, 388)Footnote 1 Arash Abizadeh (Abizadeh 2021a, b) criticizes these views; for him, equal power is not equal difference-making ability considered independently of others’ choices, but equal social ability to bring about certain political outcomes given the actually existing distribution of preferences and preference-shaping social structures in the relevant political setting. Despite their differences, however, these views similarly define having an equal say in terms of having equal power, broadly understood as some sort of ability to influence political outcomes.

One problem with this approach is that there are many ways to influence political outcomes, not all of which are relevant for democracy. Bribery, manipulation, and sheer force are ways to causally influence political outcomes, for example. Yet a political arrangement in which power to influence outcomes is equalized through constitutionally guaranteed equal opportunity to bribe and manipulate decision-makers or to stage violent mass revolt is not a democracy (cf. Wilson 2019, 155). Similarly, merely equalizing opportunity for more benign forms of informal influence, such as lobbying or mass demonstrations, without other participation rights, would not suffice for democracy.Footnote 2 Equal power views, of course, may stipulate that the relevant power is exercised within some formal decision-making procedure. It is unclear, however, if such a stipulation can be made in a non-arbitrary way. If what matters is equal power to influence political outcomes, then why distinguish sharply between power exerted through formal and informal channels (Kolodny 2014b, 333)?

Furthermore, there are many ways to exert power through formal decision-making procedures. Take Philip Pettit’s (2000) concept of contestatory democracy wherein citizens have the right not only to make decisions together, e.g., by casting ballots, but also to contest decisions already made, e.g., by vetoing decisions, initiating judicial review, and so on. Contestation gives individuals what Pettit calls editorial rather than authorial power over political decisions, or as he more recently put it, virtual, rather than active influence (Pettit 2012, 156). Now, we can imagine a purely contestatory regime where decisions are made by lot, or perhaps by officials selected by sortition, but all citizens have equal editorial power within the decision-making process. Such an arrangement clearly grants citizens power over political decisions; not only can citizens get certain decisions revoked or revised, but this fact also prompts decision-makers to take citizens’ views and interests into account, lest their decisions are challenged within potentially time- and energy-consuming procedures. Provisions can be made to ensure that contestatory power is distributed equally according to the relevant measure of equality.

It may not be too far-fetched to say that under such a purely contestatory regime, citizens would have some sort of say in political decision-making. And yet there is a clear sense in which it is not the kind of say we usually expect democratic citizens to have. The claim that democracy is a form of government where citizens have an equal say is usually understood as referring to an authorial say. Democracy is quite naturally understood as rule by the people, i.e., an arrangement wherein citizens make, create or author political decisions together, rather than merely edit or supervise them (Richardson 2003, 56; Brettschneider 2007, 4; Ober 2008, 2017; Waldron 2012, 188). As Eric Beerbohm notes, “standing in an authorial relation to the political system is constitutive of democratic citizenship.” (Beerbohm 2012, 45) While equal power views undoubtedly have the merit of rejecting the merely formal equality of the say by demanding that everyone be equally empowered, they struggle with distinguishing the authorial from other kinds of says, and in this way, they cannot properly explain what it means to have an equal say in the sense relevant for democracy.

One may object that this conclusion is rushed. The fact that equal power views imply that there is nothing special about the authorial say shows precisely that we need to understand the meaning of having an equal say in a more capacious way. Perhaps there really is nothing special about the authorial say, and it is an advantage, rather than a disadvantage, a feature, rather than a bug, of equal power views that they show this to be the case. Perhaps so. But the view that having a say in a democracy implies some kind of authorial, rather than, say, merely editorial, contribution to political decision-making is deeply entrenched not only in scholarly discussions about democracy, but also in everyday conceptions of it; after all, the general public would likely not see the introduction of a purely contestatory regime as a democratic improvement. This gives us at least some reason to think that giving up this view requires a very strong justification. It is what Rawls would call a considered judgement. Considered judgements can be given up if there are strong theoretical reasons to do so. If we cannot find a better theory that accommodates this judgement while coherently specifying the meaning of having an equal say, then we may indeed conclude that a purely contestatory regime, for example, can suffice for democracy. But we should carefully look at other alternatives before drawing such a conclusion.

2 Appropriate Consideration

Recently Wilson (2019) critiqued equal power views and offered his alternative which may be called the appropriate consideration view. Wilson argues that rather than having equal power, the fundamental democratic requirement is equal authority defined as follows:

“One has authority over common life when one’s decisions about what others (often including oneself) ought to do create obligations on the part of others to adjust their deliberations and actions in ways that take positive account of those decisions. Political equality requires some way of arranging common life—what we do together—so that each citizen’s decision about what we ought to do in that domain has equal authority over what, in fact, we do.” (Wilson 2019, 99)

Having equal authority, in turn, is understood in terms of appropriate consideration. One has authority when one’s judgement has appropriate consideration, i.e., can generate obligations for others to assign positive deliberative weight to said judgement (Wilson 2019, 106). Sometimes this deliberative weight is decisive; then, one has authority to command others to their bidding (Wilson 2019, 108). But this is not always the case. Appropriate consideration in deliberation, for example, cannot involve the power to command others to do anything; rather, it involves obligating others to give a “fair hearing” to what one has to say (Wilson 2019, 159). What counts as appropriate consideration, then, differs from context to context. What matters, ultimately, is setting up political institutions in such a way that throughout the whole procedure of political decision-making, citizens’ judgement is given appropriate consideration. If this is achieved, citizens have equal authority, and consequently an equal say.

The appropriate consideration view shares some strengths with equal power views, e.g., it can distinguish between merely formal and substantive empowerment. It demands not only the formal right to vote or to free political speech, but institutional arrangements wherein individuals’ judgement is in fact given consideration. Furthermore, Wilson’s view seems better positioned to grasp the meaning of having an equal say than do equal power views. For one thing, it can distinguish between power exercised in voting and deliberation, and power exercised through, say, lobbying and revolt. Lobbyists and insurrectionists can generate prudential reasons for decision-makers to consider their judgement but have no standing to obligate them to give positive weight to their judgements. Although decision-makers may be prudent to listen to them, lobbyists and insurrectionists have no authority and no claims for consideration.

However, the appropriate consideration view, similarly to equal power views, cannot specify what is special about an authorial say. Recall that for Wilson, appropriate consideration involves the ability to “create obligations on the part of others to adjust their deliberations and actions in ways that take positive account of” one’s judgement (Wilson 2019, 99). There are many ways to create obligations for others to give positive deliberative weight to our judgement, not all of which imply that we have a say, let alone an authorial say, in a decision. Suppose my spouse and I deliberate about having a party at our house. Our neighbour overhears our conversation and asks us to end the party early because she needs to get up at 5 AM the next morning. The neighbour’s request may very well obligate us to give positive weight to her judgement about the proper length of the party. Requests deserve consideration, yet requestors do not have a say. It is precisely because only my spouse and I have a say in deciding when the party ends that the neighbour needs to make a request in the first place, rather than deciding the matter herself.

Wilson concurs. Discussing the case of counsel or advice, another way to create obligations for others to positively weigh one’s judgement in their deliberation, he writes: “we—the counseled—have some obligations of minimal courtesy or politeness to pay attention to the counselor, and to acknowledge the counselor’s point. […] The obligations of politeness, when they do exist, may be trivial in the context of the counseled person’s deliberations. But the counselor nevertheless has obligation-creating power.” (Wilson 2019, 110) This obligation-creating power of the counsellor or advisor, however, does not entail that they have a say; my colleague, for example, can advise me about my career, thereby creating an obligation to positively weigh their judgement, but they do not have a say in where I work.

One may think that the neighbour in the previous example should have a say, e.g., because her important interests are at stake. Even then, however, giving her a say means giving her powers that differ from the powers of the requestor. It is hard to see how this difference can be captured by Wilson’s conceptual framework, since in both cases, the neighbour is given appropriate consideration, in the first case as a requestor, and in the second, as a fellow decision-maker. Similarly, one may argue, and Wilson probably agrees, that it would be inappropriate to treat citizens in politics simply as requestors or advisors. For example, one may hold that in the political realm such fundamentally important interests are at stake that the deliberative weight assigned to citizens’ judgements must differ both qualitatively and quantitatively from what is owed to mere requestors and advisors. But even if this is so, this consideration is beside the point. The question at hand is not about who is entitled to a genuine say and who is allowed only to be a requestor or advisor. The question is more fundamental, i.e., what is the difference between having a say, and merely having the power to request or to counsel? All of these involve obligation-creating powers, and Wilson’s framework of appropriate consideration seems too coarse-grained to pinpoint what precisely makes a particular obligation-generating ability not the ability to request or counsel, but a genuine say.

But, once again, one might suspect that this is a rushed conclusion. Perhaps the fact that the appropriate consideration view does not pick out a qualitative difference between having an equal say and other types of consideration is best seen as an indication that there is no distinction to be picked out at all. In other words, although intuitively democracy seems to require more than simply appropriate consideration, we should give up this intuition; perhaps, because it is based on an implausibly strong conception of what it means to have proper authority in communal decision-making, one modelled on command, rather than counsel or request which may be more appropriate for understanding how democratic citizens make decisions together. Again, I am not, in principle, opposed to this theoretical move, but what is at stake here is a considered judgement about democracy, i.e., that it requires equal authorial say for citizens. Before giving it up, it is worth exploring if there is an alternative view which can make sense of the relevant distinction and possess the theoretical advantages of the appropriate consideration view and equal power views. In the remainder of this paper, I present such an alternative view.

3 Constitutive Input

To understand what it means to have an equal say in decision-making, we first need to understand what it means to have a say at all. As we have seen, it cannot simply mean having power to influence the decision; lobbyists and insurrectionists have such power and influence, but they do not have a say. Similarly, it cannot mean having appropriate consideration in Wilson’s sense; requestors and advisors can have that without having a say. So, what is it to have a say? Consider the example of a jury trial. In a jury trial, witnesses have power to influence the verdict, e.g., by giving persuasive testimony. Others can have input worthy of appropriate consideration, e.g., judges can instruct the jury, sustain or deny objections, attorneys can deliver arguments, cross-examine witnesses, and so on. Still, despite their causal power and claims to consideration, neither the witnesses, nor the judge, nor the attorneys have a say in the verdict; only jurors do. They don’t make the decision, the jury does. What exactly is the difference between the jury and other parties in this setting?

The key difference seems to be this: the judge, the witness, and others merely shape the decision by providing different kinds of input; in contrast, jurors’ input constitutes the decision. Jurors’ input not only causally influences whether the defendant will be acquitted or convicted but makes it the case that the defendant is acquitted or convicted. Speech act theorists have long recognized that certain utterances do not merely shape or influence states of affairs but make them the case. For example, if the right person, going through the right motions utters the phrase “I name this ship the Queen Elizabeth” at a naming ceremony, or at a wedding the couple says “I do,” at the right moment, then this act makes it the case that the ship is named Queen Elizabeth and that the couple is married (Austin 1962, 5). Here the act of naming a ship and saying “I do” are constitutive of the ship becoming named and the couple becoming married respectively. Similarly, the jury’s input does not shape (causally or otherwise) facts about the verdict, but rather it constitutes the defendant being acquitted or convicted, i.e., it makes it the case that there is a verdict. What it is for the defendant to be acquitted or convicted is for the jury to provide the appropriate kind of input under the appropriate circumstances.

It is important not to construe the jury’s constitutive input too narrowly. The jury’s input is not a single pronouncement or declaration. It is not simply the verdict they hand over to the judge or the relevant official who pronounces it. For it to count as a genuine verdict, the document containing the jury’s decision must be the result of the appropriate deliberative procedure informed by the witnesses’ testimony, the attorneys’ arguments, and so on. If at the last moment, someone replaced the jury’s verdict with another one which, then, were delivered to the judge through the right procedure, then there is a sense in which nonetheless no valid verdict would be produced. Giving constitutive input in this context, then, is a complex, temporally extended process requiring the cooperation of many individuals. Just as at the wedding both members of the couple must say “I do,” for there to be a marriage, at the jury trial, all jurors must participate in the right way for there to be a verdict.

Which inputs count as constitutive of a decision is largely a matter of convention and institutional design. In some settings, it is the couple’s assent that makes them married, in others, it is the pronouncement of the officiating person, e.g., a priest or a public official. In some settings, juries make verdicts, in others, judges do. In each case, a particular person or group, e.g., the couple or the jury, has constitutive input when they occupy a special role within the relevant institutional context in which they are empowered to make the final decision, rather than offering information, advice, instruction, and so on.Footnote 3 Consider, for example, that in some contexts couples need to acquire a valid marriage licence from a notary. The notary certainly has input, but since she cannot decide whether the couple actually marries, this input is not constitutive of the final decision; only the couple’s is, for they perform different roles within the relevant institutional context.

Could we set up institutions in such a way that, for example, not only the jury, but also the judge, or not only the couple, but also the notary would count as co-decision-makers, capable of providing constitutive input? There are many ways to set up marriage ceremonies and judicial procedures, and many ways to distribute constitutive input within them. But this cannot be done wholly arbitrarily. There are certain limitations as to what we can reasonably recognize as constitutive input. We can certainly imagine that the judge deliberates and votes with the jurors. We can even imagine that the notary is party to an extended deliberative process wherein she and the couple decide together whether getting married is a good idea; if so, they get the licence, and must proceed with the marriage. But we will have trouble, for example, seeing events too far removed from the actual decision as constitutive inputs into the final decision, e.g., the initial appointment of the judge or the notary, or their acceptance to school decades before the decision is made. Crucially, we will have much trouble seeing inconsequential input as constitutive. If the jury pronounces the verdict, but no one acts accordingly, then, in fact, their role as decision-makers is not recognized. They do not have a say in the decision, for their input, in fact, decides nothing.

These considerations now can be applied to democratic decision-making as well. Citizens in a democracy have a say when they occupy a social role wherein their input into a political decision-making process is constitutive of the decision being made. One might find it somewhat unusual to think of citizens in a democracy as occupying a specific social role. After all, the whole idea of democratic citizenship seems to be that everyone is entitled to a say regardless of their social role. Recall, however, that in a democracy citizens create or author fundamental political decisions. This is indeed a specific social role, and a very familiar one at that. People who make fundamental political decisions are rulers. Democracy, as mentioned, is naturally understood as rule by the people. My account of what it means to have a say specifies the meaning of this well-known term. Democracy is rule by the people precisely in the sense that in a democracy all citizens are co-rulers of the polity (Brettschneider 2007, 3).

The claim that in a democracy, citizens are co-rulers of the polity does not commit one to any radical view of popular sovereignty, whereby the people are granted absolute power to rule however they like. To the contrary, the idea that democratic citizens occupy a specific social role, a public office, sits quite comfortably within the liberal tradition. After all, to be a democratic citizen is to be entrusted with public power to shape the system of binding rules that govern social interaction in the polity; and a “claim to political power is a claim to political authority over others—a public office—rather than a claim to a resource that is personally beneficial.” Thus, to be “a voter is to occupy a political office.” (Gaus 1996, 250–51) Having a democratic say, then, is a matter of occupying the political office of the democratic co-ruler wherein one has the ability to provide constitutive input into the making of fundamental political decisions.Footnote 4 But what does it mean to have an equal say?

4 Having an Equal say

Since having a say is a matter of having constitutive input, one may try to define having an equal say in terms of the equality of this input. The principle of “one person, one vote,” for example, defines equality for a type of input. So perhaps when all types of constitutive input, including not only the vote, but also deliberative input,Footnote 5 is equal, people have an equal say. However, it seems that such equality of input does not entail an equal say. Some argue, for example, that persistent minorities lack an equal say, even when their input is equal to others’ (cf. Abizadeh 2021a). One may disagree, but to decide whether persistent minorities’ say is equal or not, we need to first specify what having an equal say means. Thus, the meaning of having an equal say cannot be straightforwardly defined in terms of equality of input.

I propose that having an equal say means having a say as an equal. Having a say is a matter of occupying the political office of the democratic co-ruler with constitutive input in political decision-making, and people have an equal say when all co-rulers are equals. This means not that people are both co-rulers and also equals, independently of their role as co-rulers, but rather that they are equals qua co-rulers. What makes co-rulers equals in this sense? Co-rulership here is understood primarily in terms of political decision-making. Democratic co-rulers make fundamental political decisions; when they are equals qua decision-makers, they are all equal makers of these decisions. When are people equal makers of decisions? Clarity can be gained about this by considering when people are equal makers of something more generally. Let me, then, briefly sketch an account of equality in making.

As a heuristic exercise, let me consider a rather distant example of making, i.e., artistic creation, such as painting. Of course, painting and political decision-making are very different kinds of making. Political decision-makers, much like jurors and the marrying couple, create not physical artifacts, but social facts (Gilbert 2013; Tuomela 2013). They may not be involved in writing the actual document which records the decision, but only in bringing the decision itself into existence. We care about who the makers of a painting are, e.g., a famous artist or a forger, and who the makers to a political decision are, e.g., a democratic public or a dictator, for very different reasons. Still, painting and decision-making are instances of making which share enough general features for insights about the one to elucidate matters about the other.

When are, then, individuals equal makers of a painting? Mere participation in painting a picture is not enough. Many of the great Renaissance paintings were made with the help of assistants working under the supervision of a master. It seems right to say that Michelangelo and the assistants who painted elements of the background are not equal makers of the frescos of the Sistine Chapel. While they are all makers of the paintings, Michelangelo is more of a maker than the assistants. It makes sense to say that The Creation of Adam is Michelangelo’s painting, although he did not paint it alone. He is the primary maker of the painting while assistants are secondary or auxiliary makers. This is not because Michelangelo was a better painter than the assistants, or because he put in more hours. Neither needs to be the case.

Michelangelo is the primary maker of the painting because of the way in which the process of making the painting was organized. Michelangelo and the assistants perform different roles within the painting process. Assistants paint, while Michelangelo does not only paint, but by painting he also revises and approves assistants’ contribution to ensure that, at least in essential respects, the final product conforms to a unified vision determined by him. There is a sense in which assistants’ contribution is only provisional, waiting to be finalized by Michelangelo, at least when it comes to contributions affecting the essential aspects of the painting, i.e., perhaps not the colour of angels’ hair, but, say, the expression on Adam’s face. Suppose, however, that Michelangelo is absent; there are no primary or secondary makers. Everyone gets a brush, and they start painting. No one approves or revises the contribution of others, no one paints over what others paint. Or perhaps everyone can paint over anything anyone else paints. In any case, no maker occupies a role in which they do not only contribute to the final product, but also approve and revise others’ contribution in order to impose a singular vision on the essential aspects of the painting. This, I believe, is a case of egalitarian making; in the absence of primary and secondary makers, all painters are equal makers of the painting.

This suggests that equality in making is a matter of not having relationships of primacy, or superiorityFootnote 6 more generally, between individuals qua makers. Someone is a superior maker compared to others when they do not merely contribute, but by contributing they also revise and approve said others’ contribution, at least when it comes to essential aspects of the final product, so that it conforms to a unified vision determined by them. This makes the contribution of the subordinated makers provisional, waiting to be approved, at least in the sense of not being overridden, by superior makers, and as such, less than a full contribution. It is important that primary makers modify the essential aspects of the final product. If Michelangelo’s role is to revise and approve, however extensively, only insignificant details, e.g., angels’ hair colour, or anything in the upper-left corner, then he is not a primary maker. He has another specialized role, i.e., that of angel-hair-painter or the upper-left-corner-painter, which does not necessarily stand in a hierarchical relationship to others’ roles.

Note that the hierarchy discussed here is organisational, rather than social hierarchy. Following Kolodny (2023a), I understand social hierarchy as asymmetry in power or authorityFootnote 7 over persons and disparities of regard, i.e., the unequal enjoyment of certain favourable responses, e.g., respect or courtesy. Superiority in making requires none of these. Of course, the historical Michelangelo was likely a social as well as an organisational superior of his assistants. But we can imagine a hypothetical Michelangelo who is the primary maker of a painting without being a social superior of the secondary makers. Hypothetical Michelangelo can revise and override all his assistants’ contributions, but without any power or authority over them, i.e., he, and only he, can paint over what they paint, but he cannot tell them what to paint. And this role may not entail asymmetric regard for Hypothetical Michelangelo, i.e., he has no social standing compelling his assistants to show greater respect or more curtesy towards him within or outside the context of painting. He is a primary maker because he is in charge of determining the essential aspects of the painting, while assistants merely assist with this, but he is not their social superior.

Although organisational hierarchy does not require social hierarchy, social hierarchy often leads to organisational hierarchy. Suppose that a person of high social standing enters the egalitarian art collective. Although she formally has the same role as everyone else, the rest of the group shows such deference to her, e.g., out of respect or perhaps fear of her power, asking for her feedback (and implicitly, her approval) about their own contribution that she ends up, in effect, performing Michelangelo’s role within the collective, imposing her vision on the painting. Then, it seems to me, it is fair to say that despite formal equality within the group, this person has in fact become the primary maker of the painting. Or suppose that a person with very low social standing enters the collective; while other painters mostly leave each other’s contributions untouched out of respect for one another, they do not respect this newcomer, and therefore systematically erase her contributions. If this happens extensively enough, then her social inferiority can lead to her becoming a secondary painter whose contribution needs approval from others, while these others assume primacy qua makers.Footnote 8

Of course, these hierarchies are informal in the sense that they do not involve formally recognized roles of superiority; they are not necessarily laid out in any organisational chart or involve special titles. Indeed, they may contradict the formal rules of the art collective; its charter may explicitly stipulate members’ equality as makers. Yet what matters is not titles and charters, but the way in which the group de facto functions. If painters are not quite willing to say that the painting is done, unless one of them approves it, and if they defer to this person on decisions about the essential aspects of the painting, then this person occupies the role of the primary maker whether they bear a special title or not, indeed, whether or not other members of the group admit this to each other or to themselves. This brief sketch, of course, leaves many questions open that only a full theory of making could answer. Although here I cannot hope to present such a full theory, I believe these considerations already provide us with the basic theoretical resources to spell out what it means to have an equal say in the sense relevant to democracy.

People have an equal say when they are equal co-rulers in the sense that there are no co-rulers who occupy a role in which they do not merely contribute to fundamental political decisions, but also revise and approve others’ contribution so as to impose their vision on the essential aspects of the final decision. There are no primary or secondary makers of fundamental political decisions. Consider J.S. Mill’s plural voting scheme in which the educated have extra votes. An extra vote is, in effect, a means to cancel someone else’s vote, i.e., to override their contribution (cf. Kolodny 2023b, 25). The plural voting scheme defines a political decision-making procedure where, while everyone has a say, the educated occupy a role in which their job is to ensure that political decisions are by and large in line with informed opinion possessed by them. The goal of the plural voting scheme as a way of organizing political decision-making, as Mill makes it quite clear (Mill 1998, 340), is to ensure that political decisions, in essential aspects, reflect the judgement of the educated, while others are assigned an auxiliary role in which they can help shape outcomes, but not in this primary capacity. Here individuals clearly do not have an equal say.

Compare this with a counter-majoritarian setting wherein a persistent minority is given extra votes to counterweigh its inability to influence political decisions. This institutional setup does not necessarily assign the role of primary makers to the persistent minority. To the contrary, here extra votes are meant to ensure that the role which the persistent minority plays in political decision-making is not an inferior, auxiliary, or secondary one. To some extent, it resembles the egalitarian art collective assigning greater time or resources to those members who, for some reason, cannot work as much on the painting as others. One may still argue, of course, that giving extra votes to persistent minorities is not necessary to ensure their equality qua co-rulers. Here I do not want to take sides on this issue. What matters to the present discussion is that now we have a criterion for determining whether this is the case or not. What matters is not persistent minorities’ causal power to influence decisions or the consideration they receive, but the role they end up playing within this institutional context, and whether it is that of a secondary co-ruler or not. It is this question we need to consider to decide whether persistent minorities have an equal say.

As we have seen, one does not have to occupy a formally recognized superior role, e.g., that of bearer or plural votes, to be a secondary co-ruler. Informal hierarchies matter too. Sometimes, these informal hierarchies are introduced through institutional design. For example, in hybrid regimes, skewed election laws, unequal access to media or financial resources for opposition parties make it the case that despite democratic appearances, political decisions are by and large determined by a select elite. This makes them the primary makers of fundamental political decisions, i.e., primary co-rulers, despite formal political equality. Or think of Jim Crow laws which suppressed the African American vote without explicit disenfranchisement establishing white citizens as primarily in charge of collective political decision-making, despite formally recognized universal franchise. Other times, it is social hierarchy that disrupts equal co-rulership. Consider the excessive influence of money on politics. When billionaires exercise disproportionate influence on political decisions, e.g., through private campaign finance or lobbying, we find this objectionable, not only because it creates and entrenches injustice, but also because it enables them to effectively erase the contribution of less affluent citizens (Gilens 2012). The disproportionate power, authority, and regard commanded by the wealthy, i.e., their superior position within social hierarchies, can make them de facto primary co-rulers without them officially occupying any special role within the democratic process.

Keep in mind, however, that primacy in making is not reducible to social superiority. Recall the case of Hypothetical Michelangelo who is the primary maker of the painting but does not stand in a hierarchical social relation to assistants. To ensure that everyone has an equal say, it is not sufficient to give everyone a say, and then erase (objectionable) social hierarchy. Whether or not people are social equals, democracy demands their equality qua co-rulers as well. This is compatible with saying, as Kolodny does, that the lack of social hierarchy, i.e., social equality, requires having an equal say in political decision-making; that is, although Hypothetical Michelangelo could be both a primary maker and a social equal, in the political realm people can only be social equals if they have an equal say. Furthermore, none of this is to say anything about the value of having an equal say more generally. If social equality requires equal co-rulership, then it is tempting to say that social equality is the value of equal co-rulership. However, I have argued elsewhere that such an account fails to explain the agential value of equal co-rulership, i.e., why it is good not only to have, but also to exercise equal co-rule (Kapelner 2022). Based on a similar consideration, Lovett and Zuehl (2022) endorse a pluralist account of democracy’s value, referencing both equality and collective autonomy, while I argued that democracy has a distinct relational value. All these views and more are compatible with the preceding discussion, for, as I noted at the beginning, my argument here concerns the meaning, not the value of having an equal say.

5 Concluding Remarks

Democracy is a form of government in which citizens have an equal say in making at least the most fundamental political decisions. This means, I argued, that in a democracy citizens occupy a special social role, i.e., the public office of democratic citizen, where they do not merely shape fundamental political decisions, but make them; i.e., they have constitutive input in decision-making processes on fundamental political matters. Not only do democratic citizens all have a say, but they also have a have a say as equals; no citizen is a secondary, auxiliary, or perhaps even only nominal decision-maker, i.e., no one is in charge of approving and revising others’ input to impose their vision on political decision-making, leaving it to others merely to assist with this task. This view aligns with the traditional understanding of democracy as rule by the people. In a democracy, the people rule in the sense that citizens are equal co-rulers of the polity; they all rule together, and no one is superior to anyone in their role as co-rulers.

This view shares many of the strengths of equal power views and the appropriate consideration view. For example, like them, it can distinguish between a merely formal and a substantive say by requiring the actual empowerment of individuals to discernibly influence decisions. As noted in the previous section, inconsequential input does not count as constitutive input; if citizens’ input into political decision-making has no discernible influence on decisions, e.g., all matters are settled by elites’ backdoor deals, then citizens do not in fact have say. At the same time, not any kind of power or influence counts as having a say. Lobbyists and insurrectionists do not provide constitutive input into political decision-making. Equalizing the power to stage insurrection does not suffice for democracy, for individuals only exchanging threats of violence do not engage in a practice of equal co-rule. Thus, my view, unlike equal power views, can explain why equal informal influence does not suffice for democracy. At the same time, as we have seen, it also explains why excessive inequality of informal influence can undermine democracy.

This account can also make sense of the difference between an authorial and an editorial say, as well as request, counsel, and so on. For example, to return briefly to the example of the jury trial, a defendant’s right to appeal gives her an editorial say to contest the verdict, but it does not make the defendant a co-decision-maker. Contestatory input, i.e., an editorial say, is not constitutive input. Similarly, requests, advice, and instruction issued by attorneys and judges to the jury may be inputs into the decision-making process about the verdict, but not constitutive inputs. This explains why in a purely contestatory regime citizens would not have the right kind of say, and why, therefore, such a regime is not a democracy. The same is true of an arrangement where citizens’ input into political decision-making is only considered as request or counsel.

My goal in this paper was to present a coherent conceptual framework for specifying what it means for democratic citizens to have an equal say that is in line with common reasonable views about democracy. Much more needs to be said of the implications of this framework, e.g., what it implies about the allegedly democratic arrangements of proportional voting, enfranchisement lottery, and sortition, as well as other issues in democratic theory, such as representation or judicial review. It is also worth considering how this view bears on the question of the value or justification of democracy, which I bracketed in this discussion. I believe that exploring these issues can lead to interesting and important insights about these widely discussed topics. This, however, is a task for future research.