1 Introduction

In recent years, the voting rights of senior citizens have come under considerable scrutiny.Footnote 1 As the votes of senior citizens have been perceived as decisive in important electoral decisions, questions have been raised regarding the legitimacy of letting those who will experience relatively few of the consequences be decisive. Following the 2016 Brexit referendum, a poll showed that a third of British voters did not believe citizens aged 70 should have a say in major decisions (The Scottish Sun 2019).Footnote 2 Unsurprisingly, such sentiments were much stronger among the young. The age-based differences between those who voted ‘remain’ and ‘leave’ were significant. In 2018, an analysis made by former YouGov director Peter Kellner showed that, by January 2019, the ‘leave’ majority would vanish (Kellner 2018). How could a majority disappear in two years? The analysis did not assume that anyone had a change of heart regarding Brexit. The forecasted shift was due to developments in the demographic composition of UK voters. With young people – who tended to prefer ‘remain’ – reaching the voting age, and the death of senior citizens – who tended to prefer ‘leave’ – the ‘leave’ majority would evaporate.

While the merits of giving senior citizens a full vote have come under considerable scrutiny following Brexit, this is not a new discussion. In a 1970 essay, Stewart suggested disenfranchising people when they retire or turn 70 (Stewart 1970). The primary motivation for this proposal resembles that identified in the post-Brexit poll. Stewart complained that people close to the end of their lives make up a large proportion of eligible voters and – by his account – thereby have a disproportional influence in state elections, given that the time in which they will be affected by any outcome is much shorter than for younger citizens. A way of corroborating this argument in favor of disenfranchising senior citizens is by looking to the expanding literature which argues that we should increase the voting power of young citizens, relative to senior citizens, for reasons having to do with intergenerational justice. For example, in his influential discussion of intergenerational justice, van Parijs (1998) argues that we may have fairness reasons to ascribe less weight to the votes of senior citizens compared to the votes of younger citizens, while Gosseries is less convinced that egalitarianism requires this (Gosseries 2022). Relatedly, Bidadanure (2015) defends youth quotas in parliaments and O’Neil (2022) a voting age of 12 by appealing to intergenerational justice. And Lecce (2009) argues that disenfranchising the young is a social injustice (a topic also discussed by Gheaus (2023)). These arguments agree that we should decrease senior citizens’ influence, including their voting weights, relative to younger citizens for reasons of intergenerational justice.Footnote 3 Thus, if these arguments are correct, there are reasons of justice for restricting the voting power of senior citizens relative to younger citizens.

Our aim in this paper is to explore whether there are democratic reasons for restricting the voting power of senior citizens relative to younger citizens (i.e., non-senior citizens). To do so, we turn to the boundary problem in democratic theory, i.e., the question of deciding who should be included in democratic decision-making (Arrhenius 2005; Dahl 1991; Goodin 2007; Whelan 1983). Two prominent solutions have been proposed to the boundary problem: the all-affected principle and the all-subjected principle. By looking at the reasons underlying these principles, which explain why affected or subjected individuals should be included, we will be able to determine whether there are democratic reasons for restricting senior citizens’ voting power. It has been argued that we must solve the boundary problem by looking at why democracy is valuable (Lippert-Rasmussen and Bengtson 2021; Saunders 2012; Song 2012). A prominent view of why democracy is valuable is the relational egalitarian view that democracy is valuable because it is constitutive of equal relations (Kolodny 2014; Viehoff 2014; Wilson 2019). This view has recently been proposed as a solution to the boundary problem (Bengtson 2022). For these reasons—and because the all-affected principle and the all-subjected principle have come under criticism (Lippert-Rasmussen and Bengtson 2021; Saunders 2012; Song 2012)—we explore this relational egalitarian principle in addition to the all-affected principle and the all-subjected principle. We will argue that whether there are democratic reasons to grant senior citizens a lower voting weight than non-senior citizens depend on which principle of democratic inclusion you support and the context within which the scheme is to be implemented. Regarding the all-affected principle and the all-subjected principle, there is reason to grant most senior citizens a lower voting weight than most non-senior citizens in most decisions. What the relational egalitarian principle entails is largely context-dependent. Indeed, we will show that it is sometimes in accordance with the relational egalitarian view to grant senior citizens a greater voting weight than non-senior citizens. Thus, whether there is a conflict between democracy and justice when it comes to the voting rights of senior citizens depends on which boundary principle you support. In this way, our investigation might also contribute to testing the extensional adequacy of these different principles.

A few remarks before we proceed. First, we should make clear what we mean by ‘senior citizens.’Footnote 4 We do not want to set the age limit for senior citizens too high, say, at 90 years old, for two reasons. The first is that, in public discussions on whether the voting weights of senior citizens should be lowered, it is not merely those at the age of 90 and above that discussants have in mind, as is revealed, for instance, in discussions on Brexit (such as those mentioned above). There, they speak of senior citizens as people at or above 70. Second, defining senior citizens as those above 90 makes for an easy and less interesting argument. Moreover, it is such a small group, so lowering their voting weights would most likely not make much of a difference in many cases. Thus, we want a harder, more interesting case where senior citizens still have a significant amount of life years left and in which the group is of significance to elections. We admit that there will be some arbitrariness to where exactly we set the limit (whether it should be 65, 70 or 75 years of age). We will set it at 70 years of age (as is done, for instance, in the Brexit poll and Stewart’s piece). Moreover, we want senior citizens and non-senior citizens to be mutually exclusive categories. This means that we by senior citizens will understand anyone at the age of 70 and above, and by non-senior citizens will understand anyone below the age of 70.

Second, there are several ways of restricting voting power. As we might evaluate such restrictions in different ways, this paper takes up two ways of differentiating the voting weights of senior and non-senior citizens. The first is disenfranchisement of people above a certain age (in our case, 70 years of age).Footnote 5 In this paper, we refer to this as the disenfranchisement of senior citizens. The second measure discussed in the paper is age weighting. Under such a scheme, senior citizens would retain their voting rights, but their votes would count for less than those of non-senior citizens.Footnote 6 Discussing both will illustrate that whether differential voting weights can be justified for democratic reasons will sometimes depend on how we differentiate such rights. Of course, there could also be other ways of ensuring that senior citizens have less influence (at least some of which are practiced in some contemporary political systems). For instance, one could make voting more costly as that is assumed to affect senior citizens disproportionately; adopt separate constituencies with weaker representation of senior citizens in parliament; or gerrymander districts.Footnote 7 While we agree that discussing some of these options would be interesting, we focus on differentiating senior citizens’ voting rights for three reasons. First, as Beckman (2017: 887) says, “voting rights are among the essential requirements of democracy.” And, furthermore, voting rights have a special place, partly for historical reasons having to do with disenfranchisement of women, people of color, etc., for many people in contemporary democracies (Wall 2007: 430–431). This means that voting rights have a particular expressive significance (which we will discuss later). Second, and relatedly, as Bengtson and Lippert-Rasmussen (2021: 574) say, “focusing on the right to vote is clearly in line with how most theorists of democracy have thought about democratic inclusion, e.g., in relation to universal suffrage.” Third, instead of short discussions of many different institutional solutions to lowering the influence of senior citizens, we have decided to focus only on voting rights to make space for a thorough discussion of these. But hopefully, some of the insights from our discussion can be transferred, at least to some extent, to these other solutions as well (e.g., if it is democratically objectionable to lower the voting weights of senior citizens, it is likely also democratically objectionable to gerrymander to lower the influence of senior citizens).

2 Theories of Democratic Inclusion and the Voting Rights of Senior Citizens

As explained, we want to turn to the boundary problem in democratic theory to discuss whether there are democratic reasons for granting senior citizens a lower voting weight than non-senior citizens. The boundary problem, recall, is the problem of deciding who should be included in democratic decision-making (Whelan 1983; Dahl 1991; Arrhenius 2005; Goodin 2007). Solving this problem is not straightforward. We cannot say, for instance, that a prior democratic decision should decide it because then we are left with the question of who should be included in that decision, ad infinitum (Goodin 2007).

The two most prominent solutions to the boundary problem are the all-affected principle and the all-subjected principle. We will investigate the voting rights of senior citizens in light of those principles. However, as explained, we will also investigate it in relation to the relational egalitarian principle. The reason for this is that it has been argued that we must solve the boundary problem by looking to why democracy is valuable (when we later introduce the principle more thoroughly, we point to some remarks in this regard). Now, a prominent view of why democracy is valuable is the relational egalitarian view that democracy is valuable because it is a constitutive part of equal relations (Kolodny 2014; Viehoff 2014; Wilson 2019). Moreover, this view has recently been put forward as a solution to the boundary problem (Bengtson 2022). For these reasons, we discuss the relational egalitarian view in addition to the all-affected principle and the all-subjected principle. In doing so, we address what we take to be an omission in the theoretical debate over the voting rights of senior citizens. While the question has been addressed as one of justice, it has so far not been thoroughly addressed from the perspective of democratic inclusion.Footnote 8 This points to another contribution of our article: fleshing out what the various principles offered as solutions to the boundary problem imply for the voting rights of senior citizens may in turn affect people’s views on the plausibility of these solutions. Thus, our discussion will also be valuable to the debate on the boundary problem in democratic theory.

2.1 The All-Affected Principle

Let us start by discussing the voting rights of senior citizens from the perspective of the all-affected principle. Generically speaking, the principle specifies that those affected by a decision ought to be included in making that decision in the first place (Dahl 1991; Goodin 2007).Footnote 9 This suggestion, however, is too general. Thus, we follow Bengtson and Lippert-Rasmussen (2021: 572) in understanding the all-affected principle as specifying that ‘all whose interests are affected by the relevant collective decisions should have an influence on them’.Footnote 10 For our purposes, we do not have to settle exactly which are the relevant collective decisions – e.g., whether decisions in the family or at the workplace fall within this category – since decisions at the state level clearly do fall within this scope. In order to investigate what the all-affected principle implies for the voting rights of senior citizens, we must know why it is that an affected individual has a claim to inclusion in democratic decision-making. The most plausible ground underlying the all-affected principle specifies that the affected individual must be included to be given an opportunity to protect their interests (Whelan 1983: 17; Goodin 2007: 60; Miller 2009: 216; Bengtson and Lippert-Rasmussen 2021: 575).Footnote 11

Clearly, senior citizens have interests that are affected by democratic decisions. To mention a few, they might have an interest in how much money is allocated for pension payments; how well funded the health care system is; whether their children and grandchildren can live good lives; and the quality of retirement homes. Since non-senior citizens who are affected must be included to have an opportunity to protect their interests – an assumption that we do not challenge in this paperFootnote 12 – it is clearly the case that disenfranchising senior citizens would similarly violate the all-affected principle. If they were disenfranchised, they would not have the opportunity to protect their interests in collective decisions affecting them. But might that which explains why it would be undemocratic to fully exclude them from democratic decision-making also justify why senior citizens should have a lower voting weight than non-senior citizens? To see why this might be the case, it is important to note that while providing people with the opportunity to protect their interest is incompatible with disenfranchisement, it may allow providing people with opportunities of different sizes. The opportunity for interest protection grounding does not specify how good this opportunity must be. One view would be to say that as long as the affected individual partakes in the decision-making – irrespective of the degree to which they are affected and the weight of their vote – they have been given the opportunity to protect their interests, and they have been treated in accordance with the all-affected principle. Suppose a decision is to be made on which A’s interests are affected to a degree of 0.99 whereas B’s interests are affected to a degree of 0.10.Footnote 13 Suppose A and B are each given one vote on the decision with the same weight. Would we say that A’s and B’s claim to an opportunity to protect their affected interests has been satisfied? Intuitively, it seems that whether the individual has been given an opportunity must not only depend on an absolute matter – whether the individual has been given a say – but also on a relative matter – the say the individual is granted given the degree to which their interests are affected compared to the say other individuals are granted given the degree to which their interests are affected. So, it might be that in the example with A and B, A has been given a sufficient opportunity in absolute terms, but the opportunity seems insufficient in relative terms. A is affected almost ten times as much as B, yet they are given the same weight.

We must therefore ask, whether it is the case that non-senior citizens people’s interests are generally more affected than senior citizens’ interests. If that is the case, then they should have a greater voting weight than senior citizens on the opportunity-for-interest-protection grounding of the all-affected principle. But is this actually the case?

This depends on senior citizens’ number of affected interests, the extent to which these interests are affected (which is also a matter of how important they are to the person) and the duration for which these interests will be affected relative to non-senior citizens’ interests. Perhaps senior citizens and non-senior citizens will have the same number of affected interests. And perhaps the extent to which their interests are affected will be the same overall, e.g., it might be that non-senior citizens’ interests in education are affected to a larger extent than senior citizens’ interests in education, whereas senior citizens’ interests in health policy are affected to a larger extent than non-senior citizens’ interests in health policy, etc. Perhaps.Footnote 14 But even granting this, it seems that there is at least a difference in the duration for which their interests will be affected, i.e., that senior citizens’ interests will be affected, generally speaking, for a shorter amount of time than non-senior citizens’ interests. Whether that is the case will depend on how long the decisions that are a result of the election will last. In general, we might expect that regular decisions, such as laws, will last for a shorter period of time than non-regular decisions, such as referenda. To give an example, it is possible that the Brexit decision (a referendum) will exist and thus affect UK citizens for longer than their latest election (a regular decision).Footnote 15 If so, we should in general expect that if senior citizens should have less of a say than non-senior citizens, this is particularly the case in non-regular decisions. But there are two reasons to tread carefully here. First, the distinction between regular and non-regular decisions is ultimately only a proxy for that which really matters: how long their respective interests will be affected. In principle, nothing precludes regular decisions from lasting longer than non-regular decisions.

Second, the fact that a non-regular decision, such as a referendum, was organized because the parliament so decided shows that the election (the non-regular decision) was at least as important, duration-wise, as the non-regular decision.Footnote 16 Thus, we cannot simply look at regular and non-regular decisions in general and say that senior citizens should have less of a say over non-regular decisions than over regular decisions because they will be affected by those for a shorter amount of time (compared to non-senior citizens). In the end, what matters is for how long senior and non-senior citizens will be affected by these decisions. But at least in relation to some people within the group of senior citizens, their expected remaining life years suggest that they will be affected by a shorter period of time than non-senior citizens.Footnote 17 This was also what was suggested in the debate about Brexit: they do not get to live with the consequences for as long as non-senior citizens, so they should have less of a say. And path dependency theories suggest that once laws are in place, they often stay in place (Pierson, 2000). If so, and to that extent, senior citizens will be affected for a shorter amount of time than non-senior citizens and should have less of a say.Footnote 18

But there might also be a duration consideration which speaks in the opposite direction. As Poama and Volacu (2021: 16) suggest when discussing the voting weights of senior citizens:

for some […] elections […], older citizens should have more voting weights than younger ones. This is because the electoral process is generally reversible, i.e., it allows for prior electoral decisions to be undone over time. But the opportunity to influence electoral outcomes is also temporally bound and by comparison more constrained for older citizens. For elections where citizen’s interests are equally affected, older citizens’ lower life expectancy might recommend that we grant them more electoral weights to compensate for the higher electoral influence that life expectancy gives to younger citizens.

Even if this is correct,Footnote 19 the importance should not be overstated. It is, after all, only one consideration among others. And since we can expect, due to the considerations mentioned in this section, that at least many senior citizens will be affected by most decisions for a shorter amount of time than non-senior citizens—even if we assume that the number of affected interests and the importance of these affected interests are the same for senior and non-senior citizens—it is not undemocratic to grant them a lower voting weight according to the all-affected principle.

2.2 The All-Subjected Principle

Let us now turn to the second prominent answer to the boundary problem, the all-subjected principle. Instead of affectedness, the all-subjected principle specifies that those subjected to the relevant collective decisions should have an influence on them (Dahl 1991: 122; Beckman 2008: 351; Abizadeh 2012: 878; Erman 2014: 539; Goodin 2016: 370–373).Footnote 20 We may distinguish three understandings of what it means to be subjected to a decision. A subject could be (i) anyone to whom the law ascribes legal duties; (ii) anyone conferred a legal power by the law; or (iii) anyone subject to the institutions enforcing the law.Footnote 21,Footnote 22 As is common, we will understand ‘being subjected to’ in the third sense. This understanding most clearly separates the all-subjected principle from the all-affected principle (Goodin 2016: 370–371). The most promising value taken to underlie the all-subjected principle is self-government (Lopez-Guerra 2005: 221; Abizadeh 2008: 39–40; Miller 2009: 214; Näsström 2011: 120–122; Goodin 2016: 369; Bengtson and Lippert-Rasmussen 2021: 579–580). On this understanding, the subjected individual must be included to be (or remain) self-governing. We follow Raz in understanding self-government as involving three conditions which are necessary and jointly sufficient for an individual to be self-governing: (i) mental abilities; (ii) an adequate range of options; and (iii) independence (Raz 1986: 154–155).Footnote 23

Consider first disenfranchisement of senior citizens. If we assume that non-senior citizens who are subjected to democratic decisions must be included to be self-governing,Footnote 24 it seems to also be in violation of the all-subjected principle to disenfranchise senior citizens. Senior citizens are clearly subject to democratic decisions in the same way that non-senior citizens are. For instance, it is not the case that senior citizens would not be subject to enforcement if they were to break the law, say, by driving too fast.

However, this assumes that it is individual self-government, which underlies the all-subjected principle. If instead of including people when subjected because it is constitutive of their individual self-government, what if we should include subjected people when and because it is constitutive of collective self-government? As Beckman (2017: 895) says, if we appeal to collective interests, such as collective self-government, “it does not follow from the claim that the collective interests of citizens benefit significantly from democratic institutions that any particular citizen is morally entitled to vote … All the argument shows is that the voting rights of some sub-set of the citizenry is significant” (see also Bengtson and Lippert-Rasmussen 2021: 581–582). In short, the collective may be self-governing even if some individual citizens are not included. And, in principle, it is possible that the collective may be self-governing even if a sub-set of the citizenry, such as senior citizens, are not granted the right to vote. So if we were to appeal to collective self-government instead of individual self-government, it might not even be in violation of the all-subjected principle if senior citizens were disenfranchised.

So, whereas assuming individual self-government entails that it would be undemocratic to disenfranchise senior citizens, that might not be the case if we assume collective self-government instead. The remaining question is then what the all-subjected principle, assuming that individual self-government is its underlying value, implies about giving a lower weight to the votes of senior citizens than the votes of non-senior citizens. As we said above, senior citizens are clearly subjected to democratic decisions in the same way that non-senior citizens are. But they might in general be subjected for a shorter amount of time because they have fewer expected life years left than non-senior citizens. If so, it seems to be in line with the all-subjected principle to give them a lower voting weight. The idea is that the longer the duration for which you are subjected, the larger the extent to which it threatens your self-government, all else equal.Footnote 25 For instance, all else equal, it is more threatening to your self-government to be a slave for five years than for one year. So, as was the case with the all-affected principle, what ultimately matters is the extent to which they will be subjected to these democratic decisions. Moreover, we can expect, at least in democracies as we know them, and because of path dependency, that senior citizens in general will be subjected for a shorter amount of time than non-senior citizens. They should accordingly have a lower voting weight than non-senior citizens.

Now, there is another distinction which is relevant when discussing the all-subjected principle. We might understand the principle as applying to particular decisions—e.g., are you subjected to X law, Y law, etc.—such that you should be included only if you are subjected to these particular decisions. However, we might also understand the principle such that decisions are made by institutionalized normative systems—such as the legal system—and that people are subjected to these systems.Footnote 26 A person should therefore be included if she is subject to such a system, irrespective of whether she is subjected to a particular decision made by that system (Beckman’s (2023) view might be understood in this way). Our discussion assumed the former understanding. If we were to assume the latter understanding, it does not make a difference to senior citizens’ voting rights—and their relative strength—if they are subjected to a lesser extent to some particular decisions made by, say, the legal system. Since subjection to the system is what matters, and since senior citizens, like non-senior citizens, are subject to the system, their claim to inclusion is as strong as the claim of non-senior citizens. But this is presumably only if we assume that they will be subjected to the system for the same amount of time.Footnote 27 If it is the case that non-senior citizens in general will live longer than non-senior citizens, then they can be expected to be subjected to the system for longer (but that, of course, depends on whether the system remains in place). And, if they will be subjected for longer, the all-subjected principle prescribes that they should have a larger voting weight than senior citizens.

To sum up, if we assume that collective self-government underlies the all-subjected principle, disenfranchisement of senior citizens might be in line with the principle. If we assume that individual self-government underlies it, disenfranchisement of senior citizens will not be in line with the principle. Assuming the latter view, and since senior citizens can be expected to be subjected for a shorter amount of time—whether to individual decisions or the system as a whole—the all-subjected principle prescribes that they should have a lower voting weight than non-senior citizens.

2.3 The Relational Egalitarian Principle

Now, it has been argued by several people in the boundary problem literature that we must solve the problem by looking to the value of democracy. For instance, Miller (2009: 204) says that the boundary problem “cannot be solved by appeal to democratic procedure. But this does not mean that it cannot be solved by appeal to democratic theory, understood to mean the underlying values, such as political equality, that justify procedures like majority voting.” Similarly, López-Guerra (2005: 221) says that “the principle of inclusion must be derived from the distinctive normative ideals of democracy.” And, finally, Lippert-Rasmussen and Bengtson (2021: 1027) argue, “we believe that a demos is democratically constituted if its constitution derives from the value underpinning democracy, i.e. that which makes us care about democracy in the first place. Basically, our thought is that if the demos is constituted in accordance with whatever value makes democracy valuable, then there can be no moral complaint against the relevant delimitation deriving from a concern for democracy” (see also Song 2012). In short, these authors argue that we must solve the boundary problem by looking to why democracy is valuable.Footnote 28 A prominent contemporary view of why democracy is valuable is the relational egalitarian view that democracy is valuable because it is a necessary, or constituent, part of relating as equals (Kolodny 2014; Viehoff 2014; Wilson 2019). In fact, this view has recently been defended as a solution to the boundary problem (see Bengtson 2022).Footnote 29 For these reasons—i.e., that it is a prominent view of why democracy is valuable and that it has been defended as a solution to the boundary problem—we investigate this relational egalitarian view.Footnote 30

Assuming this relational egalitarian view, the important question is whether senior citizens can relate as equals to non-senior citizens if we deviate from a ‘one person, one vote’ scheme. According to two of the most prominent relational egalitarians, namely Anderson (1999) and Kolodny (2014), a ‘one person, one vote’ scheme is necessary for people in society to relate as equals under ideal circumstances:Footnote 31 ‘if a procedure gives anyone a say, it should give everyone an equal say’ (Kolodny 2014: 291); ‘each citizen is entitled to the same number of votes in an election as everyone else’ (Anderson 1999: 318). However, Bengtson challenges these arguments, arguing that people can relate as equals under ideal circumstances even if we deviate from a ‘one person, one vote’ scheme (Bengtson 2020). Obviously, some grounds upon which we may grant people differential voting weights are incompatible with equal relations, e.g., if black people are given a lower voting weight than white people because they are deemed morally inferior. However, Bengtson argues, if we grant people voting weights in accordance with what they have at stake in a given decision, this will not lead to unequal moral relations between octogenarians and younger people (but, arguably, his argument extends to 70-year-olds as well, and thus to what we refer to as senior citizens). Importantly, as he argues,

a democracy with differential voting weights based on different stakes does not rank people in terms of their intrinsic worth. Some get more voting power than others in a given decision not because they are intrinsically more valuable, as would be the case if some were given a greater voting weight merely because of being white (if whites were seen as morally superior). It is because their interests are affected to a greater extent than someone else’s. Analogously, the fact that I give the candy bar, assuming I only have one, to the diabetic with low blood sugar rather than the healthy kid does not mean that the two are moral unequals. What matters to moral standing is not whether people get unequal amounts of something per se—what matters is on behalf of what they are given unequal amounts. If that something (i.e. stakes) does not have to do with intrinsic moral worth or can be justified by moral principles acceptable to all reasonable persons, unequal moral standing does not result (Bengtson 2020: 1059–1060).

But would it not lead to stigmatization of senior citizens, and thus unequal relations, if they are given less voting power than others because they have less at stake? Bengtson (2020: 1060) argues that this question is ambiguous between two understandings: (i) whether it should—given the basis on which differential voting weights are granted—lead to stigmatization of senior citizens; and (ii) whether it will lead to stigmatization of senior citizens. Only (i) is important for ideal theory (since we assume that, in ideal theory, people are reasonable and will not stigmatize, and feel stigmatized, if there is no reason for it), and we turn to non-ideal theory below. In relation to (i), he argues that there is no reason for stigmatizing senior citizens on this basis. Those who would stigmatize senior citizens—non-senior citizens—can expect to be in senior citizens’ position once, or if, they get to the same age. For this reason, non-senior citizens might react like those who interact with others in spheres where they do particularly well: “True I am in much better health than old Mike, who needs our help. However, once I get to be as old as he is now (if I do), then I will probably be no better off than he is now and I will then need younger people to help me” (Lippert-Rasmussen 2018: 133; quoted in Bengtson 2020: 1060). As he concludes, there is no reason why differential voting weights should lead to stigmatization—and, more broadly, unequal moral relations—between senior and non-senior citizens (Bengtson 2020: 1060).Footnote 32 Thus, it seems that, under ideal conditions, relational egalitarianism may be indifferent between a ‘one person, one vote’ scheme and a scheme where senior citizens are given a lower weight in decision-making insofar as they have less at stake than non-senior citizens.Footnote 33

But what if we are under circumstances in which there is injustice (i.e., non-ideal circumstances)? Suppose that injustice disadvantages senior citizens relative to non-senior citizens; that senior citizens stand as inferiors to non-senior citizens. In such a situation, granting senior citizens a greater voting weight than non-senior citizens may actually lead to less unequal relations than using a ‘one person, one vote’ scheme. This is due to the importance of democracy for how we relate to each other. As Kolodny explains, democracy is a particularly important constituent for how we relate because democratic decisions (i) usually cannot be escaped at will; (ii) characteristically involve (the threat of) force; and (iii) have final de facto authority (Kolodny 2014: 304–307). Political decisions have final de facto authority because political decision-making cannot be moderated by a higher court of appeal, and because political decisions have authority over non-political decisions (Kolodny 2014: 306). Thus, we may boost the standing of inferior persons by giving them a larger voting weight than those who stand as superiors. Indeed, ‘[I]n some nonideal circumstances, striving for equal opportunity to influence political decisions may actually take us further away from full social equality. For example, giving greater opportunity to influence political decisions to members of groups whose acceptance as social equals is under threat in other domains, especially as a kind of temporary or remedial measure, may be warranted’ (Kolodny 2014: 309). For these reasons, in a situation where senior citizens stand as inferior to non-senior citizens, we have relational egalitarian reasons to grant the former a larger voting weight than the latter. Conversely, if non-senior citizens stand as inferior to senior citizens, we will have relational egalitarian reasons to grant them a larger voting weight than senior citizens.

We must take a further relational egalitarian consideration into account. For many relational egalitarians, whether an act or a state of affairs is just or unjust is also determined by what the act or state of affairs expresses (Anderson 1999; Schemmel 2012; Voigt 2018). For instance, ‘if I fly a Confederate flag over an official building in southern US states, I might express racist attitudes, even if I am not in any way racist’ (Lippert-Rasmussen 2018: 77).Footnote 34 This act may be unjust from the perspective of relational egalitarianism because of what it expresses: that black people are morally (and socially) inferior to white people.Footnote 35 Similarly, in a situation in which there is injustice in society such that non-senior citizens people stand as inferior to senior citizens, granting the former a larger voting weight than senior citizens may be better from the point of view of relational egalitarianism. This is due to what doing so expresses: that non-senior citizens are as worthy as senior citizens and thus that they should not stand as inferior to the latter and that we are doing something to mitigate that injustice. Conversely, in a society in which there is injustice such that senior citizens stand as inferior to non-senior citizens, it may be better, expressively speaking, to grant senior citizens a greater voting weight than non-senior citizens.Footnote 36

What this comes down to is that, from the perspective of relational egalitarianism, there is no universal answer to the question of the voting rights of senior citizens. Under ideal circumstances, a ‘one person, one vote’ scheme and a differential voting weights scheme may be equally good from the point of view of relational egalitarianism (depending in the latter case, of course, on the grounds upon which differential voting weights are granted). Under nonideal circumstances, which voting weight to grant senior citizens compared to non-senior citizens depends on the injustice in society. In some contexts, a differential voting weights solution where senior citizens are granted a lower voting weight than non-senior citizens will be preferable from the point of view of relational egalitarianism. In other contexts, the reverse may be true. And in still other contexts, a ‘one person, one vote’ scheme may be preferable for expressivist reasons.

3 Conclusion

Whether there are democratic reasons to grant senior citizens a lower voting weight than non-senior citizens depend on which principle of democratic inclusion you support and the context within which the scheme is to be implemented. For instance, as we have argued, the all-affected principle might often be in line with granting senior citizens a lower voting weight because they will be affected for a shorter period of time than non-senior citizens. But if you were in a context in which laws last, on average, only, say, two years, then it might actually be in violation of the all-affected principle to grant most senior citizens a lower voting weight (this also applies to the all-subjected principle). However, since this is not the context in most democracies—many decisions last for longer, partly because of path dependency—it will for most decisions be in line with the all-affected principle and the all-subjected principle to grant senior citizens lower voting weights than non-senior citizens (though full disenfranchisement would be undemocratic). At an ideal level, the relational egalitarian principle is indifferent between ‘one person, one vote’ and differential voting weights based on stakes (where senior citizens are given less weight when and because they have less at stake). At a non-ideal level, the relational egalitarian principle is highly context-dependent since it takes into account expressive effects. Indeed, if we are in a context in which senior citizens stand as inferior to non-senior citizens, it might even be that they should have a larger voting weight than non-senior citizens.