Children,Voting,and the Meaning of Universal Suffrage

Abstract

Although universal suffrage is a broad franchise model, it allows for exclusions provided they are robustly justified. In practice, therefore, suffrage is never universal. Every modern democracy operates with its own exclusion principles, but they are all bounded by some sort of age exclusion screening for competence. However, there is another way to conceptualise universal suffrage – a conceptualisation that finds credence in existing international treaties, and which better fulfil democracies governing promise of political equality. In this model, inclusivity and universalism remain the default, and franchise exclusions are subjected to more rigorous testing. To demonstrate the potential of this framework, I apply it to questions of children’s suffrage, arguing that the theoretical grounds for excluding children are insufficient to overturn the default principles of universalism and inclusion.

Introduction

The advent of ‘universal’ suffrage in the twentieth century saw women enfranchised and secured greater protection for the voting rights of ethnic minority groups. However, given the persistence of other franchise exclusions, no modern democracy operates with truly universal suffrage. In the twenty-first century, the most significant exclusion (by size) is structured by age and is designed to ensure a basic level of electoral competence. Age qualifications have varied over time and place (Blais et al., 2001) but in no democracy or suffrage model have children been allowed to vote.

For some, these age stipulations are self-evident (Runciman, 2021). Many people regard child enfranchisement as too ridiculous to warrant contemplation. And although the prospect of enfranchising 16-year olds gained some traction in UK political culture since the voting age was lowered in Scotland, even this modest proposal has been met with considerable resistance (Chan and Clayton, 2006; Cowley and Denver, 2004). Likewise, in political theory, the idea of children voting is mainly ignored or treated with scepticism. Most theorists remain concerned with questions of universal adult franchise and give children little consideration, treating the latter’s exclusion as ‘unproblematical’ (Dahl, 1991: 123). Others explicitly reject the idea of child enfranchisement as ‘absurd’ (Chan and Clayton, 2006: 540), owing to children’s apparent lack of reason or capacity or maturity (Brennan, 2011; Cohen, 1975; Dahl, 2000: 75–76).

This child-excluding model of universal suffrage is so familiar it is easy to overlook the way it subordinates political equality to the goal of voter competency. The trade-off is problematic, however, because political equality is the basis for democratic legitimacy, and not something to be discarded lightly. Thankfully, therefore, this is not the only model of universal suffrage. Universal suffrage can be reimagined, and there are alternatives that escape (or substitute) this trade-off and produce better socio-political outcomes. One such conceptualisation – explored in this article – takes universalism as its default but sets a higher bar for exclusion – removing age thresholds and creating a broader franchise. This model prioritises political inclusion and equality, and does so – given certain assumptions about the nature and meaning of voting – without jeopardising policymaking outcomes.

Though never realised in practice, this view of universal suffrage chimes with a recent uptick in academic sympathy for the idea of child enfranchisement. In most instances, this literature addresses itself to the arguments against children voting, showing them to be hollow or arbitrary. For example, it challenges the notions that children, uniquely, are cognitively or psychologically unfit for suffrage (Archard, 2015; Lau, 2012); or that the right to vote should only be granted alongside, or subsequent to, other legal rights and responsibilities (Munn, 2012b). Others have gone further, setting out a positive case for children voting – namely, that enfranchisement would improve and better protect the lives and interests of children, and that including children in democracy will result in broader, more sensitive, and more agile public policy (Wall, 2021). In a more empirical vein, scholars studying Youth Parliaments and Councils have noted improvements in participants’ civic and leadership skills, self-esteem, and confidence (Percy-Smith and Thomas, 2010; Shephard and Patrikios, 2013). Drawing these strands together, some commentators have sought to demonstrate how different iterations of democratic theory – instrumental, value-derivative, value-instantiating, and the like – are congruent, or at least not incompatible, with the idea of giving children the vote (Schrag, 2004; Umbers, 2020).

This article takes a different route, exploring the question of children’s rights through competing frameworks of universal suffrage, and thus situating the prospect of child participation within broader and more fundamental democratic priorities. There are (at least) two ways to understand the meaning of universal suffrage, and by analysing their differing implications for children’s franchise rights we can assess which definition best fulfils democracy’s conceptual promises, and which, by contrast, is likely to inhibit democratic flourishing or visit the greatest damage on democratic credibility. Section 1 of this article lays out two conceptions of universal suffrage and their potential risks for democracy’s functionality or legitimacy. Then, in Section 2, I assess which risks are more plausible and more damaging, and argue that a more open, child-inclusive definition of universal suffrage is the safer option. To conclude, in Section 3, I indicate what a child-inclusive conception of suffrage might mean in practice and address some potential reservations.

Universal Suffrage

For much of the history of democracy, suffrage was an exclusive right. Certain groups, often comprising a minority of the adult population, were allowed to vote because they possessed qualifying attributes, like property, or the right religion, sex, or race. For instance, in the United States before 1870, only white men were allowed to vote, while in the United Kingdom, the franchise was gradually extended to men who met particular property qualifications – such that, before 1867, one in five men were enfranchised, but by 1900 it was 3 in 5 (Weale, 2007: 212). It was not until the twentieth century that democracy began operating with a (more) inclusive franchise – so-called ‘universal suffrage’ – culminating in the enfranchisement of women and the constitutional protection of pre-existing ethnic minority franchise rights.

Nevertheless, modern representative democracies still operate with several exclusions, only now they require more robust justification (Olsson, 2008: 57–59). Immigrants are commonly excluded, as are foreign visitors, newly settled resident non-citizens, long absent non-resident citizens, and in some instances, prisoners. These constituencies have an interest in the results of elections – that is, are likely to be affected by the outcome – but are debarred from the franchise for want of a deep investment in the society in question (Weale, 2007: 215–216) – lacking citizenship, residing elsewhere, or having violated the social contract (Blais et al., 2001: 52–58). In most instances, however, they constitute only a small percentage of a given population.

The only other common basis for exclusion is incompetence. That is, withholding the vote from those deemed too irrational, or too lacking in understanding, or autonomy, or maturity, to use it ‘properly’ or ‘responsibly’. In select instances, voter competence is assessed – to ascertain whether people with severe mental health problems are eligible to vote (Munn, 2012a: 1054). But most scholars regard franchise testing as undemocratic – see Brennan (2011) for a rare defence – so competence is usually established by proxy; that is, by age. Above a particular threshold (commonly 18 years old), people are assumed to be competent to vote; below it, they are assumed to be incompetent. The exclusion of children from suffrage is unique and consequential insofar as it disenfranchises a sizable part of every democratic population – even in ageing countries like England and Wales, 21% of people are under 18 (UK Government, 2020). And thus, one-fifth of the population are liable to have their basic interests and policy concerns deprioritised relative to the interests and concerns of older (voting) cohorts.

This franchise model – structured by age and screening for competence – is nevertheless thought to fulfil Article 21 of the United Nations’ (1948) Universal Declaration of Human Rights, which says: ‘The will of the people shall be the basis of the authority of government . . . expressed in periodic and genuine elections which shall be by universal and equal suffrage’. This rendering of universal suffrage is somewhat contradictory; a universalist concept that, in practice, has major exclusions (Dahl, 1991: 123). Furthermore, although the age qualification is familiar and uncontroversial to most modern observers, it is not inherent to the idea of universal suffrage, but rather a condition justified by a set of contested assumptions about the meaning of the demos and the role of elections in democratic politics. It assumes, for example, that voting is a mechanism for extracting or distilling public opinion, and that the views of the electorate have a powerful influence on the content of law and public policy (Dahl, 1991: 126). These assumptions turn voter competence (and sensible voting) into political imperatives. And they implicate children – assumed to lack the independence or cognitive facilities necessary to make sensible, non-risky democratic choices – as a problem for electoral politics (Dahl, 1991: 126–127). To dodge this problem, we conceptualise children as less than full members of the political community (Cohen, 1975). We make voting a privilege of competency (or (full) citizenship), and we deem children incompetent (or partial citizens, or citizens-in-waiting), rendering them ineligible for suffrage (Smith et al., 2005). Competence and full citizenship are realised at 18, and with this status, the right to vote.

Nevertheless, the scope of suffrage theory is broad, and there are other conceptions of universal suffrage beyond this familiar rendering. A second – stricter and more intuitive – reading of universal suffrage would extend the franchise to every resident citizen, including children, and in doing so resolve the above contradiction (Hinrichs, 2002: 39). This alternative interpretation finds support in Article 12 of the United Nations’ (1989) Convention on the Rights of the Child, which says that any child ‘capable of forming his or her own views [has] the right to express those views in all matters affecting the child’. Article 13 goes on: a child’s ‘right to freedom of expression’, should ‘only’ be restricted if necessary to protect the rights of others, national security, and public order and health. This rendering of universal suffrage prioritises – in fact, makes salient – a commonplace democratic belief in human equality. It assumes the fundamental properties of humanness (whatever they are) are evenly distributed, and that all individuals possess equal human dignity. It also holds that every citizen stands in more-or-less equal relation to the state – the demos is an inclusive, single entity – and that each member’s formal interactions with the state must be regulated in the same way and to the same degree (Schrag, 1975: 441–442; Weale, 2007: 207–208).

In systems of representation, the implications of this basic equality-of-worth and relational equality are twofold. First, each person’s interests and perspectives have equal value, none are self-evidently superior. And second, each person and their distinct perspectives deserve equal consideration by decision-makers. By extension, each person must be entitled to vote – ensuring their perspectives are not systematically overlooked by those in power, and that decision-makers are held accountable to all available perspectives (Cummings, 2020: 396–397; Olsson, 2008; Wall, 2021). Children, on this basis, are members of the demos or political community in the same way and to the same degree as adults. They are citizens whose perspectives and concerns deserve the same respect and protection, and therefore, they are as entitled to, and as in need of, the vote as adults. ¹

Why, then, has only one of these models been put into practice? What makes the narrower version of universal suffrage preferable to the broader, and on what basis might we re-prioritise? The answers will not be found in their normative aspirations, which in both cases is to broaden the franchise to its fair and legitimate limit and establish a basis for inclusive conceptions of democracy. Instead, we must compare their disadvantages – the ways in which they fall short of or injure their democratic aspirations. Because on the question of risk, each model of universal suffrage is quite distinct. In the open framework, the risk is confused or adverse voting outcomes: if children are irrational or ignorant or immature (in a qualitatively different way to, or to a significantly greater degree than, adults), the votes of children will reflect these deficiencies. Allowing them to participate in democracy will therefore distort or undermine policymaking, and possibly render it meaningless. In the more closed framework, on the contrary, the risk is arbitrariness and unfair exclusion. When voting rights are hitched to an age threshold, a competency standard is applied exclusively and arbitrarily to the young. Incompetent adults are therefore allowed to vote, but competent young people are excluded.

Risks: Policy Confusion or Political Exclusion

Of course, incoherence, confusion and exclusion are inevitable aspects of every democracy. No democratic process claims to be completely or perfectly logical; democratic outcomes are never entirely clear or unambiguous; and no franchise model is fully inclusive. However, while each of these idiosyncrasies is ineradicable, they differ in scale and importance. It is thus appropriate to ask: which is more damaging, or stands in greatest tension with the basic principles of democracy – distorted electoral outcomes and confused policymaking, or political exclusion and inequality? ² Moreover, when weighing constitutional or institutional change, we should also ask: which by-product is most likely to emerge or be exacerbated by this or that democratic innovation? That is, which danger is more credible? Together, these inquiries provide a framework for assessing each interpretation of universal suffrage and determining whether children should be given the vote. In this context, confusion is a less likely danger, while exclusion is more damaging – for although open models of universal suffrage might produce incoherent politics or policymaking, less open models necessarily entail exclusions that violate the principle of democratic equality.

Confusion

The belief that enfranchising children would lead to reckless or chaotic democratic outcomes assumes that children are not competent to participate in democratic life. The outline of this argument runs as follows. According to various studies in developmental psychology, reasoning facilities are acquired gradually and come with age. For example, most children under 4 do not realise that synonyms (like rabbit and bunny) refer to the same object (Perner et al., 2002), and children under 7 are prone to class-inclusion errors – not understanding that a subcategory (a rose) is also an example of a broader category (flowers) (Borst et al., 2013). Those under 12 years old tend not to reason from abstract moral principles (Umbers, 2020: 737). Owing to these deficits in reason, as well as their general immaturity and ignorance of the world, children are supposed to be cognitively and existentially dependent on adults (Hinrichs, 2002: 41). And together, these limitations in reason, knowledge, and experience disqualify children from voting, which is an activity that requires care, insight, and responsibility. Given the opportunity to vote, children’s incompetence – expressed through arbitrary and ill-conceived preferences – would confuse or invalidate electoral results or policy outcomes. Therefore, to protect democracy from epistemic disorder – to ensure sensible and stable policymaking – children must be excluded.

Aspects of this analysis are undoubtedly true. In certain respects, children are less competent than adults; they have less life experience and less worldly knowledge, and are, on average, less cognitively developed. Still, the broader exclusionary inferences are nevertheless unwarranted. Take, for example, the stages of cognitive development, which are imprecise and only a crude indication of a child’s (ir)rationality (Borst et al., 2013: 1366). An age group is deemed capable of class-inclusion judgements if 50% can answer the relevant test question correctly (Politzer, 2016). To set a classification boundary this low turns a vague and porous distinction into a hard one, and makes it possible (even likely) that significant numbers of children will pass the test but still be deemed incapable because 51% of their age cohort did not. That the number of correct answers increases significantly when test questions are written in child-friendly language further illustrates the arbitrariness of group distinctions (Politzer, 2016).

Moreover, even if testing was precise and accurate, it is not clear how the abilities tested for are relevant to voting. Many people continue to make class-inclusion errors well into adulthood (Borst et al., 2013) – and such failures do not imply a lack of astuteness or experience. At the same time, it is easy to imagine someone passing these tests despite being politically clueless or reckless (Schrag, 1977: 171–172).

This points to a general question about which competencies – possessed to what threshold – are necessary for citizens to vote responsibly? What is it, in other words, that ensures voting does not lead to unclear outcomes and chaotic policymaking? Specific, largely abstract, cognitive abilities may not map onto voting requirements, but what about the general ability to make rational choices (Cohen, 1975), understood, broadly, as a capacity for logical syllogising (Lau, 2012: 862)? Alternatively, must voters be capable of self-governance (Dahl, 1991: 97)? Or need knowledge of the machinery of the state – the different branches of government, for instance, or the legislative process itself? Should they be able to distinguish between different parties and candidates and platforms (Archard, 2015: 100–101)? Or must they show awareness of the details of different bills and legislative programmes (Schrag, 1975: 450)? Aside from these attributes or types of knowledge, what should voters consider when voting; what are their normative guardrails – the common good or self-interest (Schrag, 1975: 445–446)? Most importantly, by which of these standards would children fare significantly worse than adults? And if voter competence is so integral to democracy, why, given the demonstrable paucity of adult voter knowledge and rationality, have democracies been so historically prosperous? (Hannon, 2022; Somin, 1998).

These questions are connected to a larger issue of democratic legitimacy. For scholars (e.g. Estlund, 2008: 6–14) who think voters ought to surpass a cognitive or competency threshold (however defined), democratic legitimacy has an epistemic component. If democratic authority is justified by the likelihood that democratic processes will produce good (or right) outcomes, it is important that electorates are rational and oriented towards the ‘truth’. On the contrary, if legitimacy is a function of citizens’ political equality (as discussed below), the demand for a rational electorate recedes. The same is true if democratic legitimacy derives from the protection that democracy provides to citizens, both from one another and from the state itself (Weale, 2007: 50–55), or the fact that its decision-making processes are provisional and revisable (Anderson, 2009: 215–217). It might be preferable to have competent electorates, but they are not, in these models, essential to democratic legitimacy.

However, assuming (as is quite common) that we ought to place a competence demand on citizens – what does this imply for the role of children in democracy? Fundamentally, the fact there is no consensus for what is required of voters means it is impossible to say adults have it – whatever it is – and children do not. Still, for arguments sake, let us imagine both a minimalist and a maximalist voting requirement and explore what each might mean, first, for children’s suffrage, and second, for potential policy outcomes. Assuming a minimal competency requirement for voting – for example, the ability to identify and compare a small number of parties or leaders or policies – many children would qualify as competent, but, due to their age, remain ineligible to vote. They would fall short of the age threshold but exceed the underlying qualification (Munn, 2012a: 1049–1050; Schrag, 1975: 454). To remedy this irregularity, and recognise their adequate voting competence, the voting age could be lowered or removed. We cannot know exactly what this would mean for voting outcomes. However, by definition, the additional participants – included for their competence – would not introduce new or excessive ignorance into democratic policymaking.

Now imagine a more stringent or maximalist competence requirement – for example, detailed understanding of existing and proposed legislation, and the ability to reason from generalities to particulars and vice versa, with an eye to common interests. Most children would probably fail these requirements. But then so too would many adults, who repeatedly demonstrate deficits in basic political knowledge, policy awareness, and reasoning capacity (Achen and Bartels, 2017; Caplan, 2007). Adults commonly misunderstand and conflate their interests with those of the broader community (Caplan, 2007: 18–19; Klocksiem, 2019: 28–30). And even those who understand their own interests often struggle to identify the policy, politician, or party, responsible for catering to them (Somin, 1998: 417–433).

If we hold electorates to these high competence standards, democracies already tolerate, and in some cases thrive with, incompetent voters. When governments pursue sensible policies, it is not because they have competent electorates, but rather in spite of their large numbers of incompetent voters. In fact, the notion that flourishing democracies require knowledgeable citizens is contested by evidence that knowledge and reasoning capacity correlate with partisanship and closed-mindedness (Hannon, 2022). And on contentious and identity-defining issues like climate change, the more information one has, the more entrenched and polarised one is likely to become (Kahan et al., 2012). This should give us pause when thinking about the supposed dangers of allowing children to vote. If electorates do not need to be competent (and might even suffer for it), how would enfranchising (allegedly incompetent) children make matters worse? We already admit children and young people into other critical democratic activities – activism, deliberation, protest, and social association (Nishiyama, 2017) – and the outcomes of these activities are not obviously more chaotic or irresponsible as a result. So why not allow children to vote as well?

The logical end point of this argument is captured in the Condorcet Jury Theorem, which says larger electorates (such as those including children) are better at making good decisions than smaller ones, provided voters, on average, have a 51% chance of picking the ‘right’ policy answer (Goodin and Lau, 2011). There are several problems with this hypothetical scenario, not least the assumption that electoral choices have certifiably ‘right’ or ‘wrong’ outcomes – a technocratic outlook that discounts the existence of value disputes and ignores the effects of partisanship on decision-making. Nevertheless, the Theorem plausibly suggests that including children in the franchise is unlikely to be electorally disruptive. Like adults, children are susceptible to cognitive biases, but also have different worldviews that give rise to dissimilar preferences and priorities (Klocksiem, 2019: 32–33). As such, enfranchised children would not skew election results by all selecting the same candidate or party. Rather – like adults – their votes would be (randomly) distributed across the political spectrum (Olsson, 2008: 67). They may be less competent than adults, but children’s votes would not undermine policymaking or governance (any more than adults’ votes).

That said, unleashing a wide distribution of new votes is less likely if the voting age is only lowered to 16. This group’s partisan leanings (to the left) are a matter of record, and their enfranchisement is therefore opposed by right leaning voters and administrations (YouGov, 2016). A wider distribution is more likely among younger cohorts, and thus, from a party-political perspective, enfranchising younger children is more intriguing than enfranchising teenagers because the former’s partisan and policy interests are basically unknown (Runciman, 2021). In either case, however, larger electorates are probably less prone to groupthink.

Finally, the idea that children’s incompetence poses a plausible or grievous danger to the coherence of democracy, is undercut by the nature and implications of voting itself. Some commentators treat voting as a way to participate in law-making (Goodin and Lau, 2011: 159), or as a licence to wield power over others (Brennan, 2011). If electoral decisions directly impacted how laws were written or how people led their lives, then voting would be a highly consequential and potentially dangerous activity, and it would make sense to guard against the threat of ‘bad’ voters, and promote electoral competence. However, voting is not the same as making law, which is the job of elected representatives and civil servants. And voters do not have the power to decide legislative battles, nor, by themselves, the power to set (or disrupt) a political agenda. In short, voters do not acquire direct authority over other citizens, nor is voting necessarily the best way to contribute knowledge or arguments to public discourse (Olsson, 2008: 74). Over time, policy is likely to be affected by electorates’ ideological preferences (Caughey and Warshaw, 2018) (with a bias towards richer and more educated interests (Elsässer et al., 2021)). But immediate shifts in public political sentiment are often more effectively realised by campaigning, protesting, and petitioning. Incompetent voters pose little threat to the coherence of democratic policymaking because there are several degrees of separation, and various decision-making layers, between elections and policymaking. Concerns that expanding universal suffrage (to include children) would lead to disruptive or reckless democratic decision-making are, therefore, misplaced.

This leaves the question of what voting does achieve, if not electoral control over policymaking? The answer is that voting reflects and consolidates citizens’ political equality. The franchise ensures people’s political concerns enjoy the same formal status as everyone else’s, which in turn establishes a (loose) guarantee that everyone will be listened to and not overlooked by those who make decisions and write law (Cummings, 2020: 396–397; Wall, 2021). These functions do not invest individuals with huge power. But as equality and inclusion are ‘central to the definition of democracy’ (Weale, 2007: 208), their realisation through suffrage is the cornerstone of democratic political legitimacy.

Exclusion

The risk of universal suffrage as it is currently practised is that it excludes children from the franchise and thereby denies them political equality. Democracies are based on, and express, two connected notions of equality; First, that all citizens have the same moral standing and dignity, and second, that each person has a unique (though not perfect) insight into their own perspectives and experiences (Anderson, 2009: 219–223; Dahl, 1991: 99–100; Weale, 2007: 63–64, 215). In turn, these principles give rise to two injunctions. On the one hand, people endowed with equal moral status are deserving of equal respect and protection – no one should be given more authority or offered greater security simply because of who they are – and on the other, those possessed of self-knowledge have a right to forge their own preferences. Taking these ideas together, and assuming an electoral democratic environment, it follows that citizens – children included – have a shared and equal right to participate in electoral processes.

The franchise embodies these principles of equality in two interconnected ways. First, it respects the dignity of the political community, and the plurality of its members’ perspectives. Citizens all have different concerns and priorities, and yet their moral equivalence – as persons – confers an equal right to publicly express or advance their preferences (Brighouse, 1996; Christiano, 2008: 2–4). This principle underwrites the basic liberal freedoms of religion and assembly and expression (Sigman and Lindberg, 2019: 598–600). And in representative contexts, where voting is a form of speech or expression, it grants citizens the right to convey their views at the ballot box. ³ The importance of this relationship – between equal voting rights and political equality – is illustrated by a counterfactual in which only half a state’s people are allowed to vote, while the other half are excluded; a scenario that violates the idea that citizens, and their different perspectives, have shared and equal dignity.

And second, the franchise ensures citizens have equal access to power and political decision-making. This is a form of relational equality. Voters do not become decision-makers themselves; they do not decide on policy or make the law. But voting rights give every individual the same formal opportunity to shape policy and inform the political agenda (Sigman and Lindberg, 2019: 600). Much of the time, this capacity is fairly minimal. At elections, citizens can approve a party-political programme by voting for it, or abstain or vote for another party if they do not. However, more than the democratic rights to speech or deliberation, voting allows individuals and groups to hold their representatives to account; making it harder (though not impossible) for lawmakers to systematically overlook the views and perspectives of their constituents (Umbers, 2020; Wall, 2021). Voting ensures citizens’ preferences or perspectives are minimally heeded in the political decision-making process. And voting rights ensure this basic political influence – or insurance – is available to citizens equally (Brighouse, 1996).

Of course, political equality has limits. Although individuals likely understand the challenges, limitations, and opportunities of their own experiences, they do not necessarily apprehend their own interests or the common good. Pace Dahl (1991: 97–98), political equality does not stem from citizens’ equal qualification to govern themselves, nor does it suggest every citizen can or should contribute equally to decision-making. People have different capacities and limitations, notwithstanding their moral equality, and not everyone is suited to self-governance or equipped to defend or prosecute their interests (Caplan, 2007: 157–158). However, it is because people are unequal in these ways that formal political equality – of respect and protection – is so important (Harris, 1982: 49–50). Procedural voting equality protects citizens from being ruled over unequally (Viehoff, 2014: 374), and only by including all citizens in the political process will the least advantaged be guaranteed a hearing and sufficient security. Equality through suffrage is the basis for democratic legitimacy, and because children have as much moral value and are as privy to their own experiences as adults, this ought to include children as much as it includes adults.

It might nevertheless be argued that age-based discriminations (preventing children from voting) are compatible with political equality when equality is considered diachronically – that is over the course of a lifetime (Daniels, 1983; Weale, 2007: 214). This is a distributive view of equality, concerned with whether particular goods – in this case, franchise rights – are allocated equally over people’s entire lives. Given that every child should – all things being well – pass from childhood (and disenfranchisement) to adulthood (and enfranchisement), in the long run, everyone ought to enjoy the same rights, at the same time, and to the same extent.

The problem with this view is that the same underlying principle could also justify raising the voting age to 40 or taking votes away from 80-year olds. These scenarios are distributionally even-handed – that is, everyone receives and/or loses the same right at the same time in their lives. But they are also intuitively unappealing. They are unappealing because they overlook or discount the importance of unequal social relations – inequalities that exist at particular moments. By itself, then, complete-life egalitarianism does not make society just or equal (McKerlie, 1989). And in certain domains, it must be supplemented, or trumped, by considerations of relational and/or synchronic equality (Bidadanure, 2016). This, for example, is why we do not withdraw voting rights from the elderly, even though doing so would be consonant with complete-life equality; it would diminish the present-day status and moral equality of old people (Lippert-Rasmussen, 2016: 155).

Not all relational forms of inequality are problematic. The requirement that children attend school, for example, is not un-dignifying for children, nor does it establish relationships of shame or marginalisation between children and non-school-attending adults. However, this is not the case with unequal voting rights, which, despite equalling out over time, give rise to damaging present-day inequalities in social relations (Anderson, 1999: 289; Umbers, 2020: 746–747). Voters have more influence, and are shown more consideration, than the disenfranchised; they are more respected, receive more attention and deference, and their interests are more likely to be satisfied by politicians (Kolodny, 2014: 297–298). In short, adult voters and disenfranchised children do not stand in equal relation to one another (Bidadanure, 2016: 247). And children, at this present time, are therefore (more) vulnerable to marginalisation, exclusion, domination, or exploitation (Harris, 1982: 50). Moreover, because politicians respond to short-term electoral incentives more than considerations of long-term structural justice, present-day political exclusion can be disadvantageous in ways that will not necessarily be compensated for, or equalised, later (Peto, 2018: 290–291). For example, governmental inaction on climate change, and its consequences for currently disenfranchised children, will be hard to balance against the effects of subsequent governmental failures on future children.

The biting question, then, is not whether uneven voting rights undermine political equality (they do). But rather, whether procedural voting equality is an adequate response to, and reflection of, citizens’ moral parity and equality in self-knowledge? Enfranchisement does not rule out substantive inequalities in resources – money, status, information etc. – and extreme inequalities in economic power or physical health or cultural cache can depress citizens’ willingness or motivation to participate. As such, several theorists argue that, as well as political equality, democracy requires strong principles of material and relational equality to ensure citizens are equally capable of utilising their rights and freedoms (Dahl, 1991; Sen, 2001: 89; Sigman and Lindberg, 2019: 597–600).

Certainly, a shared basic level of capacity and empowerment is important. However, realising this shared capacity is not – as some suggest – entirely beyond the purview of procedural equality. Voting is an expression of formal political equality, but it is also, by extension, the grounds for democratic legitimacy. Substantive inequalities that undermine political legitimacy, therefore, stand in tension with procedural theories of equality. And so, in the interests of political legitimacy, procedural equality commits representative democracies to guard against basic inequalities of outcome or social relations that might inhibit some people’s capacity to vote or democratically participate (diZerega, 1988: 465; Post, 2006: 33–34). The picture is complicated by the fact that relational inequalities between children and adults are, in some cases, necessary and inevitable, and child enfranchisement cannot (and should not) change the facts of children’s youth and dependency. Unequal voting rights are not a fact of life, however. And the social relations between children and adults would undoubtedly reach greater parity if the former were allowed to vote.

At the same time, every representative democracy – whatever its franchise model or socio-economic makeup – operates with and through some inequalities of power and influence. Politicians make laws, and, via processes of judicial review, judges define and curtail citizens’ rights and freedoms (Spector, 2003). Political equality is not therefore an equality of decisive political authority – citizens’ preferences are too often overlooked by politicians or overruled by judges. Instead, it is a shared and equal role in the political process. Voting rights ensure citizens are equally (though imperfectly) protected from persecution, and that they exert equal (though limited) influence over political decision-making. Other forms of equality may follow from this provision, but voting rights are the first, fundamental, and most feasible form of political equality.

So, if the principle of universal suffrage dictates that franchise inclusions be the norm or default, and exclusions be well founded, is the political equality between children and adults well founded? Can children’s democratic rights be denied, and political equality violated, on the grounds of incompetence? The answer, as discussed, is no. We do not know which competencies children lack, nor which competencies voters require, and due to its vagueness, the competence criterion is crudely assessed (according to age) and only selectively enforced (to those under a particular age threshold). As such, the fact we hold children in lower moral standing than adults (as half or partial citizens) is arbitrary (Smith et al., 2005); it leaves the promise of universal suffrage unfulfilled; and it calls into question democracy’s claims to inclusivity and universality (Hinrichs, 2002: 51).

These charges – violating political equality and undermining democratic legitimacy – are serious but rather abstract. However, construing universal suffrage in a way that stymies children’s participation also has concrete political repercussions. The disenfranchised cannot hold elected representatives to account, and therefore their perspectives and priorities are liable to be overlooked (Dahl, 1991: 104, 129). Children’s disenfranchisement is transitory and ends in adulthood (unlike women’s erstwhile exclusion, which was lifelong). But before becoming adults, children form a constituency whose concerns and perspectives lack formal representation, and are likely to be ignored in political discourse or given less consideration by political decision-makers.

This is precisely the outcome described by the UK Children’s Commissioners (2020), who warned that ‘Children’s right to be heard and involved in decision-making processes. . .is being denied’ and therefore ‘The UK government does not prioritise children’s rights or voices in policy or legislative processes’. As such, the Commissioners argue, in England, the initial COVID-19 pandemic responses ‘overlooked children’s needs’ and prioritised the reopening of the hospitality and retail sectors. Children’s political marginalisation is also reflected in the United Nations (2019) reporting that, in 2018, 30% of UK children were living in poverty, whereas, in the two decades up to 2013, pensioners (a key electoral constituency) saw their poverty levels decline by half. The fact that children feel betrayed by governments’ inadequate responses to climate change suggests they are aware of, and either angry or despondent about, their political exclusion (Hickman et al., 2021; Strife, 2012).

Although suffrage is no guarantee that children’s interests will be accommodated, historically, the acquisition of voting rights has given rise to various socio-political advantages. Between 1880 and 1938, for example, franchise expansions in western Europe had a positive impact on redistribution and the provision of public goods – albeit mediated by specific institutional arrangements and legislative processes (Abou-Chadi and Orlowski, 2015). And in 2011, when various Norwegian municipalities reduced the voting age to 16, youth engagement and representation increased, as did the initiatives designed to include young people in local politics (Godli, 2015). Not all these effects were immediate – although the Norwegian case was only a matter of years – but evidently there is a connection between having the vote and protecting one’s (broadly defined) interests.

In the face of this evidence, one might argue that, because children are represented by their parents, they do not require additional formal representation. Adults have paternalistic authority over children (unlike husbands who once made similar claims on behalf of their wives), which gives them the right to direct various aspects of children’s lives. Given this paternalistic licence – the argument goes – the costs of children’s disenfranchisement are not that severe. Aspects of this argument are correct; adults do have some authority over children and there is much that adults should decide for them. However, from these limited adult rights, it does not follow that children’s interests can or will be comprehensively represented by their parents (or other adults) (Schrag, 2004: 374–378). This assumption discounts the possibility that parents can be neglectful or indifferent, and ignores the fact that childhood and adulthood are entirely different experiences, and that, despite their memories of the former, adults are not necessarily well placed to interpret or speak to the perspectives of children (Wall, 2021). Children have unique perspectives, inaccessible to other groups. And like any other constituency, they are entitled to seek formal representation for these (multifarious) perspectives.

Still, given the reality of paternalistic authority, might the enfranchisement of children effectively multiply the votes of parents, whose influence weighs heavily on children’s opinions and actions? Clearly, we cannot answer this with certainty. Although, ballot secrecy means children could just as easily disobey their parents and defy electoral instructions. Presumably, many parents would exert a significant influence on their children’s votes. And yet, parents do not have a monopoly on influence. Like adults, children are affected by multiple, sometimes competing, sources – grandparents, friends, teachers, media. These influences endure well into adulthood, so, as Umbers (2020: 746) notes, unless we disenfranchise adults still intellectually beholden to their parents, the weight of parental influence cannot be grounds to withhold the vote from children.

It might further be argued that, because children are not campaigning for the vote, as women did a century earlier (Runciman, 2022), they are broadly unconcerned by their disenfranchisement. We rarely, if ever, ask children what they think about their exclusion, so these sorts of conclusions are highly provisional (Holt, 1975: 122). Perhaps they are not interested in the franchise. Or maybe, owing to their age and relative social status, they simply lack the resources or wherewithal to mobilise effectively. Granted, if children are content with the status quo, the psychological burdens of exclusion – self-doubt, low self-esteem etc. – will be less pronounced. But this would not change the underlying conditions of inequality; the situation itself would be no less unjust, nor would the practical consequences of the injustice be any less real (Schrag, 1975: 454–455).

Practice and Reservations

The choice between an open or a less open definition of universal suffrage – in which children are respectively included or excluded – is a choice between an outside chance of electoral or policy incoherence, on one hand, or inevitable exclusionary injustice, on the other. The unlikely worsening of democratic outcomes (with the open model) or the necessary corrosion of the democratic ideal, and attendant neglect of one constituency’s perspectives (with the less-open one). On the balance of risk – the likelihood of damage and the nature of the respective threats – this is not much of a choice. A less open model of suffrage poses the greater risk to democracy, while an open model helps democracy better realise its own promise.

This conclusion is likely to elicit two basic responses. First, a practical concern about capacity – which amounts to a rehashed version of the competency issue. And second, a conceptual challenge to do with the relationship between voting rights and other age-related rights and responsibilities.

Babies and Capacity

In a truly universal model of universal suffrage, citizens would acquire the right to vote at birth, so babies and toddlers, not just children, would have the franchise. This requires further justification. For even if one accepts that decoupling franchise rights from voter competency opens the door for 15-year olds (Schrag, 1975) or even 6-year olds (Runciman, 2021) to vote, the same logic has rarely been extended to babies.

In fact, in many ways, babies are more cognitively sophisticated than is commonly supposed. At 18 months, for example, they grasp something of the intentions of other people (Carpenter et al., 1998). And studies show that 25-month olds understand when other agents hold false beliefs (about the location of particular objects) (Southgate et al., 2007). However, the fact these tests are non-verbal points to the complicating reality that babies usually lack basic language skills, and therefore, to put it mildly, the chances of them understanding policy proposals, or knowing what political parties are, or which one(s) they favour, are slim. Babies, in other words, are more than just potentially incompetent or liable to vote recklessly and irresponsibly; they probably also lack the mental, and perhaps even physical, capacity to cast a vote at all. Many would struggle to access voting booths or even hold a pen, while those physically able (or enabled) to vote, might do so without understanding what they were doing.

Stepping back from this conundrum, it is worth asking what makes decision-making meaningful in the first place (which is not the same as what makes it good or competent). In representative democracies, the preconditions for meaningful decision-making are simple and twofold. First, voters must be capable of forming preferences. Their preferences need not be scrupulous or rational or well-thought out, nor must they be governed by assessments of the common good; they simply need to exist. And second, voters must want to express their preferences and know how to do so in a political context (Cummings, 2020: 400–401), which means understanding the nature and effect of elections (Munn, 2012a: 1054). Technical knowledge of their democratic voting structures or detailed understanding of policy options is unnecessary; voters need simply to grasp that their preference for a party or person can be expressed at a ballot box, that their ballot will be aggregated with others, and that authority will be invested in the winner of that contest. In short, for voting to be meaningful, voters must be aware of what they are doing; they must understand what elections are and mean, and personally desire to participate. Anyone who lacks this awareness, but for some reason votes anyway, votes without intention and performs a meaningless action.

How can we know that electorates will cast meaningful, intentional votes? As Wall (2021) points out, the best indication that someone has the awareness and intentionality to make their vote meaningful is not a score on a test, but simply their desire to cast a ballot. The desire to vote emerges from the two basic apprehensions necessary to make voting meaningful. First, the formation of preferences, and second, an understanding that one’s preferences can be expressed through voting. After all, why would someone vote if they had no preferences and/or did not understand what elections were for? Crucially, however, the capacity (or right) to vote does not confer an obligation. If age thresholds were abolished, large numbers of newly enfranchised children would decline the opportunity to participate. Some – like their adult peers – would abstain due to indifference or disinterest; others – probably those in the very earliest years of life – because they do not understand what is on offer, and consequently have no desire to act. Importantly, though, nobody who wished to participate would be denied the opportunity. The demos would be politically equal, and exclusionary injustice avoided.

Other Rights and Responsibilities

The final objection to giving children the vote relates to the order in which other rights and responsibilities are acquired. Voting is a serious business, and it stands to reason that the right to vote should coincide with, or follow, the right to perform other activities of similar weight and consequence (Cowley and Denver, 2004). Thus, in the United Kingdom, voting is allowed at 18 – the same age that one can buy cigarettes and alcohol. The age of consent and the age one can get married (with parental approval) or join the army is 16. Incredibly, one becomes criminally responsible at 10 years old.

Again, beyond the specific issue of voting rights, it is worth asking why any of these rights or liberties are postponed. If freedom is a fundamental human and political value, why are children denied it? One answer is that when people move from one phase of life (childhood) to another (adulthood), they acquire the qualities that make them (as adults) fit for liberty (Uprichard, 2008: 306–309). But which qualities or attributes distinguish all adults from all children, and at the same time justify the use or denial of freedom? Differences in reasoning capacity are not significant enough (as discussed), nor are levels of political maturity, which emerge at different times in different people (Cook, 2013: 443–444). Most children lack self-sufficiency or autonomy, but in a general sense so do many adults (Harris, 1982: 42–44; Schrag, 1977: 173). The most coherent difference between the young and old is biological – sexual maturity. But while this might inform restrictions on children’s sexual freedoms, it is not hugely relevant to voting rights (Scarre, 1980: 120; Uprichard, 2008: 306).

Another answer rests on the claim that liberty is an instrumental value – something necessary for the realisation of other goods, like happiness. On this basis, freedom might be curtailed if its exercise was likely to produce the opposite, say, harm or unhappiness (Dworkin, 1972; Scarre, 1980: 121–122; Schrag, 1977: 174). Assuming children are immature and incontinent and liable to abuse their liberty, their rights could be withheld for their own protection. Nevertheless, the fact that immature and incontinent adults do not have their liberty curtailed means the logic of protection is not – by itself – an adequate reason to deny someone their rights. Rather, the paternalistic case rests on the intention to protect children’s future adult selves; we deny children potentially dangerous rights until such time – adulthood – as their exercise represents less risk.

When is that, though? At what point do rights become less risky? This takes us back to the problematic childhood/adulthood distinction. We assume adults think and act more systematically vis-à-vis their purposes and interests than children – that due to their greater experience, adults are better judges of what is and is not good or healthy behaviour and are able to exercise their rights with care (Scarre, 1980: 123). And yet, while some adults clearly do consider the long-term and systematic implications of their actions, many children do as well, and sometimes in a superior manner to adults (Harris, 1982: 38–39).

On one hand, then, it is impossible to say children are perfectly distinguishable from adults according to their capacities for long-term, systematic thought (or something similar). And on the other, in some circumstances, it is pragmatic and necessary to assume they are. This assumption is a fudge. But it is justifiable insofar as it creates mechanisms by which children can be protected from dangerous liberties. That is, it empowers adults to withhold rights from children (but not other adults) that might cause them physical or psychological damage and endanger their future lives. This makes sense (and is necessary) vis-à-vis the right to have sex, or get married, or consume alcohol. But the same fudge or assumption has very little bearing on voting rights, which are not dangerous in the same way, and which children do not need protecting from (Munn, 2012b: 140–141; Umbers, 2020: 741–742). In politics – as opposed to ordinary, social life – not having rights is always more dangerous than having them. Voting rights are a form of political protection, and disenfranchisement means children are powerless to influence or resist political decisions. The franchise should thus be acquired as early as possible, and certainly before the right to drink or buy cigarettes etc. (Wall, 2021).

This leaves a final question about rights and responsibilities. Abolishing the voting age would mean children under 10 (the age of criminal responsibility) would acquire voting rights without corresponding responsibilities to the law (Archard, 2015: 99–100). This may seem jarring. But not all rights and responsibilities can or need to be balanced, and the fact that visiting foreigners and resident aliens are subject to the law of their host country but do not have the right to shape or inform that law shows that the connection between voting rights and legal responsibilities is flexible. If we expect visiting foreigners to obey host country laws, we ought not to baulk at the prospect of children voting before they become criminally responsible.

Conclusion

Democracies will never be entirely free of either electoral and policymaking confusion, or political exclusion. And yet, when resolving the definition of universal suffrage, a trade-off between confusion and exclusion is inevitable. An open interpretation of universal suffrage gives children the franchise at the expense of possible policy incoherence. A less open interpretation denies children the vote and imposes exclusionary injustice. However, democracies already function with large swathes of ignorant, irrational adult voters, who nevertheless vote meaningfully under minimal epistemic conditions. So, the inclusion of children is unlikely to (further) confuse or distort electoral results; on the contrary, it might support improved longer-term decision-making. On the contrary, excluding children without compelling reason is a violation of political equality, and a blow to democratic legitimacy. Consequently, both in principle and in practice, placing children outside the scope of universal suffrage is an impediment to democratic flourishing.