Show Summary Details

Page of

Printed from Oxford Research Encyclopedias, Politics. Under the terms of the licence agreement, an individual user may print out a single article for personal use (for details see Privacy Policy and Legal Notice).

date: 20 January 2022



  • Ludvig BeckmanLudvig BeckmanDepartment of Political Science, Stockholm University


Democracy is a term that is used to denote a variety of distinct objects and ideas. Democracy describes either a set of political institutions or an ideal of collective self-rule. Democracy can also be short for a normative principle of either legitimacy or justice. Finally, democracy might be used to denote an egalitarian attitude. These four uses of the term should be kept distinct and raises separate conceptual and normative issues.

The value of democracy, whether democratic political institutions or democratic self-rule, is either instrumental, non-instrumental, or both. The non-instrumental value of democracy derives either from the alleged fairness of majority rule or from the value of the social relationships enabled by participation in democratic procedures. The instrumental value of democracy lends support from a growing body of empirical research. Yet, the claim that democracy has a positive causal effect on public goods is inconclusive with respect to the moral justification of democratic institutions. Normative reasons for democracy’s instrumental value must instead appeal to the fact that it contributes to equality, liberty, truth, or the realization of popular will.

Democracy as a principle of either political legitimacy or justice is a normative view that evades concerns with the definition and value of democracy. Normative democracy is a claim about the conditions either for legitimacy or justice of either public authority or coercion. Debates in normative democracy are largely divorced from the conceptual and empirical concerns that inform studies of democracy elsewhere.

The boundaries of the people entitled to participate in collective decisions is a question that applies to all four uses of democracy. The boundary question raises three distinct issues. The first is the extent of inclusion required among the members of the unit. The second is if membership in the unit is necessary for inclusion or if people that are not recognized as members are on certain conditions also entitled to participate. The third and final issue concerns the boundaries of the unit itself.


  • Political Institutions
  • Political Philosophy
  • Political Values, Beliefs, and Ideologies

Four Uses of Democracy

The word “democracy” has four distinct referents: political institutions, ideals of collective self-rule, normative principles of legitimacy, or justice and egalitarian attitudes and norms.1 The first is encountered daily: Political systems are regularly described as democratic (or not) on the basis of the way legislatures are selected, governments are appointed, and the extent of political rights enjoyed by the general population. By contrast, democracy as self-rule is an idealized state of affairs that may not be fully realized anywhere. The ideal of people ruling themselves is not modeled upon actual political institutions and is potentially realizable also in other associations. Collective self-rule can be practiced in voluntary associations, at universities, sports clubs, corporations, and so on.

Democracy may also be used as a normative principle for either political legitimacy or justice. Political institutions are legitimate if they wield authority or coercion on grounds that are acceptable. Justice is concerned with who owes what to whom. Democracy may consequently be either a precondition for political legitimacy or a requirement of justice. As a precondition for political legitimacy, the claim is that coercion or authority is legitimate only if subjects are recognized as makers of the rules that apply to them. As a precondition for justice, the claim is that political participation is a requirement of fair terms of social cooperation.

According to the three perspectives just mentioned, democracy is an attribute of the procedures for collective decision-making. The fourth sense of democracy is different as it applies to the social relations among the members of an egalitarian society. A democratic society is one where the members take responsibility to resolve shared problems in the best interest of everyone. This usage is articulated in the influential work of the philosopher John Dewey who pictured democracy as a “way of life” characterized by mutual respect and a commitment to peaceful cooperation.2 The French writer and politician Alexis de Tocqueville used the term in a similar way, though he evaluated it less favorably. de Tocqueville identified democracy with the ethos prevalent in an egalitarian society (Berlin, 1965).

The main focus in democratic theory is on the procedures for collective decision-making. For this reason, this article ignores the fourth sense of democracy and focuses on the remaining three. The exposition is structured in four sections: the definition of democracy, the value of democracy, normative democracy, and the problem of the democratic people. The sections mirror the major themes in democratic theory, though the last is primarily included for the purpose of exemplifying how the conceptual and normative aspects of democracy feed into debates about what democracy is and should be.

Defining Democracy

A nominal definition establishes the necessary and sufficient conditions for the class of objects or properties to which the usage of a term is intended to apply (Gupta, 2021; Robinson, 1963). Definitions of either democratic political institutions or collective self-rule are descriptive. They do not provide reasons for valuing democracy but reasons for correct usage of the term “democracy.”

Political Institutions and Collective Self-Rule

The core attributes of democratic political institutions are widely agreed to and correspond to the criteria summarized by the notion of “electoral democracy” (Skaaning, 2021). A political system is an electoral democracy if and only if officials are selected in free, fair, and frequent elections, freedom of expression is secure, the population has access to alternative sources of information and are entitled to form independent associations and organizations, and citizenship is inclusive (Coppedge et al., 2015; Dahl, 2005, p. 188).

The claim that there is a common core meaning to democratic political institutions contrasts with the view that democracy is an “essentially contested concept” (e.g., Dryzek, 2016). A concept that is essentially contested is one for which there is no common core meaning that is agreed as accurate. One reason why democracy may appear to be essentially contested is the tendency to impute additional criteria in the concept that are regarded as important and desirable. In addition to the core meaning, we are familiar with the claim that democratic political institutions must also include liberal rights, a deliberative public sphere, social justice, environmental rights, and so on. The result is that the meaning of democracy is fragmented into separate “models” or “conceptions.”

Another reason why democracy may seem to be essentially contested is the abundance of distinct indices for the measurement of democratic government. The core meaning of democratic political institutions is vague and must consequently be specified in order to be a useful construct for data collection. The image of pervasive disagreement on democracy may turn on the fact that the core attributes of political institutions can be operationalized and weighted in different ways (Munck, 2009).

Democracy as collective self-rule articulates the conditions for when the members of an association are subjected only to rules decided by themselves. Collective self-rule is not realized by participation in collective decisions alone. In order for self-rule to obtain, the members of a unit must be in control of the rules that regulate the procedures for decision making and the rights of all members (“secondary rules”). Collective self-rule thus entails what Dahl (1989) terms “control of the agenda” or what in constitutional theory is also known as “popular sovereignty” (Chambers, 2004; Kalyvas, 2005).

The historically most influential version of collective self-rule is found in Rousseau’s The Social Contract (1762/1997) where democracy is properly used only for a political unit where no adult is subject to rules other than those they themselves agreed to. Rousseau’s grand idea is to reconcile individual freedom (autonomy) with political community. The members of a political community remain free only if the laws they abide by are reflections of the general will.

By contrast, Robert Dahl has proposed that the conditions for collective self-rule can be identified without appeal to the disputed notion of a “general will.” According to Dahl (1989), democracy is a procedure for collective decision making in a political unit that is characterized by five conditions: members have equal rights to participate, members have effective and adequate opportunities to participate, members enjoy enlightened understanding with respect to items on the political agenda, members fully control the agenda, and membership is inclusive of all adults subject to the rules. These five conditions represent necessary and together sufficient conditions for democratic self-rule.

Collective self-rule is democracy under ideal conditions. Democracy defined as political institutions is not. Yet, both the concept of collective self-rule and the concept of democratic political institutions are descriptive in the sense of not depending on judgments about the value or justification of democracy. Collective self-rule is an ideal type but not a normative ideal. The conditions that must be satisfied for a people to rule itself are not by definition either desirable or justified (Dahl, 2006, p. 8; Ross, 1952, p. 87). The value of democracy is determined by argumentation, not by definition.

Procedural and Substantive Democracy

The predominant understanding of democracy is procedural. A procedural definition identifies the necessary and together sufficient criteria for democracy by criteria for how to make collective decisions. The decision procedure, not what is decided, determines the extent to which decisions are democratic.

The criteria for democratic procedures can be defined either narrowly or broadly. The narrow understanding was influentially captured by Joseph Schumpeter’s (1950) dictum that democracy is a method for the competitive selection of rulers. The unique feature of political democracy is that no single party or ruler monopolizes access to public power. A significantly broader account is Robert Dahl’s procedural criteria of democratic self-rule that are explained in the section “Political Institutions and Collective Self-Rule”. Indeed, the procedural requirements for democratic self-rule are so demanding that they are unlikely to ever be fully realized.

A substantive definition of democracy includes criteria that apply to the output of collective decisions. Based on a substantive definition, decisions made by a perfectly democratic procedure are nevertheless undemocratic if their content violates substantive democratic demands. An example is Ronald Dworkin’s “partnership conception” according to which “a majority’s decisions are democratic only when […] the status and interests of each citizen as a full partner in that enterprise [are protected]” (Dworkin, 2006, p. 131). It seems to follow from this view that a decision that introduces discriminatory policies—to take an example—would be undemocratic even if the decision had been made by a democratic procedure. Another example of a substantive definition is one that says that democracy requires “human rights for everyone” (Goodheart, 2005, p. 135). Neither the institutions required for electoral democracy nor the criteria for collective self-rule are sufficient for an entity to be democratic, thus understood. In addition, the content of the decisions must be such that they protect and secure human rights.

Critics argue that substantive definitions of democracy lend credibility to interventions in democratic practices by experts and guardians (Walzer, 1981). They do seem to provide a justification for the power of courts to strike down decisions made by democratic bodies in order to protect the substantive content of democracy. A more radical objection against substantive definitions is that they depend on controversial claims to moral truth. But some have objected that such truths do not exist. The fact that no normative claim is more valid than another leaves one with no alternative but to make decisions by democratic procedures (Kelsen, 1955, p. 6). That is a non sequitur, however. If all normative claims are equally valid or invalid, it follows that the claim that decisions should be made by a dictator is just as valid as the claim that decisions ought to be made democratically. In the case that moral claims cannot be justified, there can be no moral justification for procedural democracy either.

However, the contrast between substantive and procedural definitions of democracy does not depend on normative reasons. Nominal definitions of democracy are statements about linguistic meaning. Reasons for and against a proposed concept are ultimately either reasons about what is a useful construct, given one’s purpose, or reasons about which construct best fits with linguistic convention (Collier & Adcock, 1999; Gerring, 1999).

Democracy’s Value

Descriptive definitions of democracy, like those looked at in the previous section, leave open the grounds for valuing democracy. The grounds for valuing democracy are either instrumental or non-instrumental. Democracy’s instrumental value depends on democracy being reliably connected to outcomes of moral importance. The value of democracy thus depends on the capacity of democratic institutions to deliver valuable outcomes. The idea is that the value of democracy is contingent on its performance. This is denied by the claim that democracy’s value is non-instrumental. Democracy has non-instrumental value if there is reason to value democracy for what it is and if these are reasons that do not depend on how democracy performs. Democracy’s non-instrumental value is explained either by it being a part of something else that is morally important or by its symbolic value (Destri, 2021; Zilotti, 2020).

Not all reasons for valuing democracy fit neatly with the distinction between instrumental and non-instrumental value. A renowned argument is that democratic participation has valuable educative effects. Mill (2010, Chapter 3) famously argued that participation in collective decisions is a school in “public spirit” as it invites people to take responsibility for public ends. Rawls (1971, p. 234) similarly argued that opportunities to partake in political life develops the “intellectual and moral faculties” of the citizenry. This view appears to value democracy instrumentally despite the fact that no appeal is made to how well democracy performs. Yet, it does not seem that to value an activity because of its byproducts is to value it non-instrumentally (cf. Elster, 1997).

The Non-Instrumental Value of Democracy

Non-instrumental value can be attributed to different elements of democracy. One view is that non-instrumental value is attached to the fairness of majority rule. The grounds for the non-instrumental value of democracy thus derives from some specific feature of the democratic procedure. A different view is that the non-instrumental value of democracy derives from the relationships enabled among the members of an association that is ruled by democratic procedures.

Majority Rule as Fairness

The most intuitive argument for the non-instrumental value of democracy is arguably that it represents a fair procedure for the resolution of political disagreement. The premise is that disagreements that need to be resolved should be adjudicated by a method that is fair to all parties involved. Democracy is fair to the parties involved because it affords them an equal opportunity to influence the outcome. This is in effect an argument for majority rule as the fair procedure for collective decisions. The rule that decisions should be determined by the majority is fair, provided that the group deciding is inclusive, because it gives “each expressed opinion the greatest weight possible compatible with giving equal weight to all opinions” (Waldron, 2016). The fairness of majority rule is thus reason to conclude that democracy has non-instrumental value.

It should be noted that majority rule can be put to use in both democratic and undemocratic associations. The distinctive mark of a democracy procedure is not majoritarianism but inclusion (Pasquino, 2011).

The claim that democracy is valuable because it is fair should be distinguished from the claim that democracy is valuable because it ensures that decisions are responsive to the preferences of the members. Responsiveness is promoted by majority rule. But the claim that democracy is valuable because it is responsive implies that democracy is appreciated only as a means to an end and that the value of democracy is in the end instrumental (Estlund, 2008, p. 76).

One objection is that it is sometimes unfair to let the majority decide when the stakes are very high. Dworkin invites us to consider an overcrowded lifeboat with the survivors of a wrecked ship that must find a way to reduce the number of people in the lifeboat in order to save the rest (Dworkin, 2011, pp. 483ff.). Would it be fair to decide who should leave the boat and die by counting the hands raised for and against each person in the boat? Dworkin thinks not. Majority rule is a fair method for collective decision making only if the basic needs and interests of participants are off the table. Dworkin does not deny that majority decisions can be fair though he denies that majority rule is unconditionally fair.

In response, Jeremy Waldron (2010) suggests that decisions about outcomes should be distinguished from decisions about rules for deciding outcomes. While majority rule is not necessarily fair in deciding who gets what and when, it may still be a fair procedure in decisions about the rules to use in deciding who gets what and when. Even for the shipwrecked in the lifeboat, majority rule is a fair procedure in deciding the rule by which to select the unlucky.

But is Waldron right that majority rule is necessarily fair in decisions about rules? Consider decisions that determine who should be entitled to vote. These are undoubtedly decisions about rules. Imagine the not altogether hypothetical situation that only White men are entitled to vote and that they disagree about whether the vote should be extended to Blacks and women. In order to resolve the matter, they make a decision by majority rule. Is there any reason to conclude that the decision is fair just because it is made by majority rule (Beckman, 2017)?

It seems to be the case that majority rule is fair in some situations but not in others. In order to clarify that issue, a distinction can be made between two reasons for why majority rule is fair. One reason is that majority rule leaves the decision in the hands of the majority (majority sensitivity). Another reason is that majority rule gives participants an equal say (equality sensitivity). The point is that when stakes are heavily skewed and unequal, it is unclear that equality sensitivity is a source of fairness. It does not follow that majority rule is necessarily unfair in such situations, however. Majority sensitivity might still be a source of fairness among people with unequal stakes. Hence, fairness in conditions of unequal stakes mandates majority rule with proportional and therefore unequal influence (Brighouse & Fleurbaey, 2008). This conclusion is controversial, however. Influence in proportion to stakes arguably negates other fundamental interests of the parties involved. Kolodny (2014b, p. 229) points out that proportional influence subordinates some people to decisions made by others. Given strong interests in avoiding relations of subordination, proportional influence must be rejected.

Equal Relations

The last point suggests that democracy can be valuable because of the social relationships that it enables. Thus, Ober (2007) points out that democracy is unique in providing an inclusive opportunity to develop the “capacity to associate in decisions.” That is a major virtue of democracy given that associating with others is an element of human happiness and of what it means to be human. Democracy’s non-instrumental value thus consists in it promoting a basic human capacity. This might appear reminiscent of arguments for democracy’s instrumental value. But remember that political institutions have instrumental value only by virtue of their performance. The opportunity to associate with others does not depend on how well democracy performs but is an attribute of what democracy is. If associating with others is valuable, democracy therefore has non-instrumental value.

Saffon and Urbinati (2013) argues that democracy embodies the intrinsic importance of equal political liberty. Political liberty requires equal political rights and equal opportunities for political participation—the very same qualities Dahl defines as integral to democratic procedures. The value of democracy is thus conditioned by the normative significance of the rights and liberties that are necessary and indeed constitutive elements of democratic procedures. The value of political liberty is a reason to value democracy non-instrumentally.

By contrast, Kolodny (2014a) argues that a democratic distribution of political influence is non-instrumentally valuable because it is a constituent part of a society where members relate to each other as equals. There is reason to value democracy for what it is, given that a society where people relate as social equals is valuable. A society ruled by democratic procedures is one that reduces to a minimum the relationships of superiority and inferiority.


Every account of the non-instrumental value of democracy must be able to explain why these values are uniquely embodied by democracy and why they are sufficient in justifying democracy. Even if true that democracy embodies features that are valued non-instrumentally, it may be that other and potentially non-democratic procedures embody the same non-instrumental values. This possibility is particularly clear if democracy is valued as a fair procedure. Democracy is evidently not the only fair procedure. There are other ways to make decisions that give each participant an equal chance to influence the outcome. In particular, why not make decisions by lot or by taking turns? (Beitz, 1989, p. 76; Estlund, 2008). A related but distinct point is concerned with the sufficiency of non-instrumental values for the justification of democracy. Any theory that claims to justify democracy by recourse to its non-instrumental value must be able to explain why this value is so weighty as to overshadow the importance of outcomes in terms of well-being and substantive justice (Estlund & Landemore, 2018). The point is that if one sincerely believes that democracy is justified because of its non-instrumental value, then one is committed to the view that democratic procedures are valuable even if they consistently produce results that make everyone worse off in terms of well-being and justice.

The Instrumental Value of Democracy

The claim that democracy is instrumentally valuable lends support from a wealth of empirical work in comparative politics. Amartya Sen famously showed that democracies are superior in terms of protecting their inhabitants from starvation (Sen, 1999, p. 152). Additional studies supply evidence to the effect that democratic institutions improve public health and education (Mesquita et al., 2003), economic growth (Acemoglu et al., 2019), human development (Gerring et al., 2012), and peaceful international relations (Dafoe et al., 2013). This body of work provides ample support for the instrumental value of democratic government. Of course, the support provided is only as strong as permitted by available empirical evidence. Causal explanation is a complex undertaking that is open to revision. The provisional character of empirical knowledge implies that the validity of the claim that democracy has instrumental value is also provisional.

Empirical support for democracy’s instrumental value is speaking primarily to democratic political institutions. Little or no empirical evidence exists to support the instrumental value of democracy as collective self-rule. Though many actual political systems are electoral democracies, few, if any, meet the conditions of collective self-rule.

A final observation is that instrumental justifications that are grounded in empirical evidence are inconclusive. In order to establish that democracy is morally justified, it is not enough to identify outcomes to which democracy is causally related. In addition, one needs an account of why these outcomes are morally important. For an outcome to be morally important, it needs be a “bearer” of normativity. The claim that some states of affairs are bearers of normativity in turn depends on an account of the “sources” of normativity (Chang, 2009; Väyrynen, 2013). The point is that claims about the normative significance of the facts causally related to democratic institutions push for moral explanations that are beyond the ambit of empirical investigation.

There are two main strategies in explaining the instrumental value of democracy: (a) arguments to the effect that democracy is instrumental to first-order values, and (b) arguments to the effect that democracy is instrumental to second-order values. The distinction trades on the difference between values with a specified content and values that are “content neutral” (Colburn, 2010).

Equality, Liberty, and Truth

To the extent that equality, liberty, and truth are valuable, they are first-order values in the sense of having a particular content. According to a variety of arguments, democracy is uniquely equipped in promoting each of them. Equality is examined first, specifically equal enjoyment of well-being. Though well-being is a complex notion, it can be explained as engagement with ends that are good and experienced as such. Well-being so conceived has both a subjective and an objective component that together represent an intrinsic good to which everyone has an interest. Now, if the well-being of each person is of equal moral significance, there is reason to accept the principle of equal advancement of well-being. The more well-being each person achieves, the better. And equal well-being is better than unequal well-being.

Following Christiano (2008), the principle of equal advancement of well-being plays an important role for justifications of democracy. Nondemocracies may, of course, be able to advance equal well-being too. Yet, given assumptions about the nature of society as it is, there is reason to believe that only a democracy will advance equal well-being in the long run. The assumptions introduced by Christiano (2008, Chapter 2) are that the members of society tend to be in disagreement, that their individual capacities and backgrounds are diverse, that all are prone to cognitive bias, and that all judgments are fallible. In view of these assumptions about living together in society, equal advancement of well-being is likely only if public decisions are inclusive and if public officials are regularly held accountable. In other words, democracy is a necessary albeit insufficient condition for equality with respect to well-being.

The liberty-based view holds that democracy is the best safeguard for the protection of individual rights and liberties. The rights one has are not preordained but are established through political struggle. A society that secures individual rights is therefore premised on providing its members with the means to establish individual rights. The basic means needed to this end are rights to participation. Only through rights to political participation are people able to fight for and maintain other rights and liberties. Political rights are the “rights of rights,” in Waldron’s memorable phrase (Waldron, 1999). Thus, democracy is a valuable instrument to individual rights and liberties.

The third view is that democratic procedures are instrumental to the realization of collective judgments about “reasonable,” “true,” “correct,” or “rational” public policy (Landemore, 2013). The value of democracy derives from its special relationship to knowledge, or what is generally referred to as its “epistemic” qualities. The epistemic case for democracy is made in various ways. One view is that majority rule is superior to dictatorial rule as it pools the judgments of the many. The crucial premise is that individual participants are more likely to be right than wrong (Goodin & Spiekermann, 2018). Another view is that public decisions will be more informed and impartial when based on reflective and public deliberation (Cohen, 1986). The argument about the epistemic virtues of democracy is related, but not identical, to the claim that only democratic rights and liberties are conducive to practices of public criticism that promote the growth of scientific knowledge (Popper, 1963).

Popular Will Theories

Democracy is arguably unique in providing the means for making public decisions conform to the preferences of the population. The point is that a democratic process allows the majority to “induce the government to do what they most want it to do and to avoid doing what they most want it not to do” (Dahl, 1989, p. 95; Mill, 1861/2010). The value of democracy derives from its being instrumental to the realization of the popular will.

An attractive feature of this argument is that it identifies the instrumental value of democracy without specifying the content of that value. This is crucial, according to some writers, since otherwise the reason why people should be included in political decisions is contingent on them making a contribution to the realization of ends defined as valuable without consulting them (Rostböll, 2015; Shapiro, 1997).

The popular will argument is premised on the claim that majority rule is a feature of democratic decision making. Democracy satisfies the will of the majority because democratic decisions are made by the majority rule. However, the social choice literature has dissected a range of problems with the tenet that majority rule produces outcomes that are preferred by the majority (Coleman & Ferejohn, 1986).

The most obvious problem is that majority rule is not a single unique method for the aggregation of individual choice to a collective decision. For example, both proportional elections and majoritarian elections are instantiations of the precept that the will of the majority should prevail. Yet, they represent different interpretations of that principle. Proportional systems seek to mirror the will of the majority in elected bodies by ensuring that seats are allocated in proportion to the votes received. But that leaves open the creation of different majorities within that body. A proportional system mirrors the will of the people but does not ensure that any specific majority of the people is influential. By contrast, majoritarian elections virtually ensure that a unique majority is influential. The government appointed following a majoritarian election is more likely to be that which is preferred by the majority of voters exactly because it does not mirror the will of the majority in elected bodies. Yet, depending on the distribution of electoral support in different electoral districts, it is possible that a majoritarian election produces a winner that is not, in fact, preferred by the majority, as frequently happened in U.S. presidential elections. In sum, the claim that majority rule provides a clear-cut method for the satisfaction of the majority will is fraught with ambiguity (Weale, 2018).

The social choice-based objection to majority rule was influentially pressed by William Riker (1982), who argued that the connections between democracy and collectively preferred outcomes are spurious, since no non-arbitrary method for aggregation of individual preferences into collective decisions exists. Riker’s conclusion has subsequently been challenged as widely overstated; only rarely does majority rule produce arbitrary results (Landemore, 2013; Mackie, 2003). These disputes do not detract from the insight that majority rule is unable to establish an unambiguous interpretation of the claim that the will of the majority should prevail.


“Pure instrumentalism” is the view that the justification of democracy depends only on its causal contribution to morally worthy ends (Griffin, 2003). The choice between democratic and nondemocratic forms of government depends exclusively on instrumental considerations, or what Arneson calls the “standard of best results” (Arneson, 2003, p. 41). In fact, pure instrumentalism is not the claim that democracy has instrumental value but the claim that democracy is justified if and only if it has instrumental value.

One potential reason for pure instrumentalism is that democracy lacks non-instrumental value. For example, Wall (2007) points out that equal votes—which are commonly regarded as a requirement of democracy—do not invariably secure either fairness or equal status. Therefore, no reason exists for attaching intrinsic significance to equal voting. Arneson argues that epistemic reasons support the contention that the value of democratic procedures is “driven by convictions about the likely or certain consequences of following the procedure.” Since one can never be certain about what morality requires, the value of democracy is at best instrumental (Arneson, 2003, p. 130). Against this, it can be pointed out that the uncertainty of moral value does not seem to disqualify claims to the effect that democracy has non-instrumental value. If moral value is uncertain and if non-instrumental values are moral values, there is no epistemic reason to conclusively reject that democracy has non-instrumental value.

A distinct reason for pure instrumentalism is the claim that non-instrumental values are immaterial to the justification of democracy. Arneson and Wall both embrace a version of the “deeper egalitarian ideal” according to which everyone’s life is equally important and that one should be concerned exclusively with how people’s lives are affected (Wall, 2007, p. 418). The egalitarian principle arguably implies that a political order is justified only if it generates outcomes that in the long run affect people’s lives equally. If non-instrumental values make no difference to how people are affected by political institutions, it follows that non-instrumental values are immaterial to the justification of democracy.

One possible objection is to deny the presumption that non-instrumental values are not affecting people’s lives. People are affected by the extent to which institutions affirm values that are important to them—such as the principle of equal respect. If that is correct, a consistent egalitarian should care also for what Beitz (1989) has called “higher-order interests” in the public recognition of equal status. A person excluded from the vote is not just refused an opportunity to influence political decisions. That person also receives a message that she is not an equal member of society. And that message is affecting her.

Mixed Justifications

Instrumental and non-instrumental value are distinct but not mutually exclusive categories of value. Democracy can be valued both for what it is and because of its consequences. According to Dahl, a democratic process is valuable because it embodies a “just distribution of freedom” but also because it allows each citizen “to protect and advance their most fundamental rights, interests, and concern” (Dahl, 1989, pp. 91, 322). Weale (2007) similarly argued that democracy is unique in respecting individuals as political equals in addition to being effective in securing collective goods.

Democracy’s instrumental and non-instrumental value was defended also by Beitz (1989) and Christiano (2008). To them, equal opportunities for political influence are valued as a means for the advancement of interests, but also because the institutionalization of equal opportunities to influence publicly affirms each person as an equal. The latter consideration is reason to value democracy for what it is and not just for what it can be expected to achieve. Political equality is not, in this view, a requirement of justice—as Dahl would argue. Instead, political equality is necessary for the realization of justice as a public ideal, where justice is seen to be made as well as achieved.

Mixed justifications of democracy seem less vulnerable to the challenges facing the alternatives previously considered. Elisabeth Anderson (2009) has pointed out that people would prefer a democratic distribution of political rights to an authoritarian distribution even if the latter turned out to perform better in terms of instrumental outcomes. A democratic order is a constitutive part of the way of life that most people value. This point might be valid and yet valid only to an extent. Democracy is valuable both for what it is and because of its performance. Democracies need not perform better than its rivals in order to remain the best political system overall. However, unless democracy performs sufficiently well, its non-instrumental value is unlikely to be enough.

Normative Democracy

The word “democracy” is frequently used as shorthand for a normative principle of either political legitimacy or justice. The content of normative democracy is determined by principles of either legitimacy or justice.3 In one view, democracy corresponds to the principle of public equality such that only a political system that coheres with the requirements of that principle is democratic (Christiano, 2008). A different view is that democracy refers to the principle that collective decisions should be decided by an inclusive process of public deliberation among free, equal, and reasonable citizens (Cohen, 1997, p. 74).

Taking democracy as a normative standard for the assessment of the legitimacy of political institutions implies that democracy is a normative concept. Indeed, some tend to believe that normative judgments are inevitable in deciding when an entity is either democratic or not. An account of democracy must articulate the normative reasons that together constitute the necessary and sufficient conditions for either political legitimacy or justice. The result is that democracy emerges as an “interpretative concept” that is inexorably normative (Dworkin, 2011, p. 379).

Democracy as Justice

The distinction between legitimacy and justice is not always acknowledged. For example, following Allen Buchanan (2002), a political entity is legitimate only if it protects the basic human rights of the subject population. Legitimacy so conceived is virtually indistinguishable from minimal justice. Valentini (2012) similarly defends the view that liberal justice and legitimacy are not distinct but both premised on the democratic principle of equal respect.

The view that political legitimacy and justice overlap received its most powerful articulation in John Rawls’s Political Liberalism.4 Rawls’ “liberal principle of legitimacy” holds that exercises of public power must agree with a constitution that all reasonable citizens can be expected to endorse. The normative force of this principle is grounded in a concern both with the conditions for political fairness and political legitimacy (Rawls, 1993, pp. 217, 225). Justice requires equal and inclusive participatory rights because no one could reasonably accept a constitutional order that affords others greater influence in decisions that determine coercive uses of public power. Legitimacy requires equal and inclusive participatory rights because the exercise of public authority is permissible only if consistent with the canons of public reason.

The overlap between legitimacy and justice is confounded by disagreement about the meaning of political legitimacy. One view is that the state is legitimate if and only if morally permitted to employ coercion in the pursuit of public ends. The point is that legitimacy so conceived does not imply a duty of obedience on the part of subjects. A legitimate state is one that enjoys a liberty and right to rule and where subjects are not entitled to interfere with the state’s doings (Wellman, 1996). The notion of legitimacy as morally permissible coercion seems closer to prevailing understandings of justice. If justice is about who owes what to whom and claims of justice are enforceable, it seems to follow that it is permissible for the state to implement claims of justice. The requirements of political legitimacy are, in this view, largely reducible to those of justice.

Democracy as Political Legitimacy

The alternative view is that the conditions for political legitimacy are distinct from justice. Democracy is a requirement for legitimate public decisions; “the job of democratic institutions is to ensure the form of popular control required for political legitimacy” (Pettit, 2012, p. 132). Public decisions are legitimate if and only if subjects are morally obligated to obey them. Legitimacy is not a mere liberty and right to rule but also a claim and right by the ruler to be obeyed, and the conditions for political legitimacy are, in this view, equal to the conditions for political obligation. The subjects of the state have political obligations if and only if they are morally obligated to obey the laws and decisions made by the state.

If legitimacy and justice are distinct, it may be that states that comply with principles of justice are nonetheless illegitimate. A just state may be entitled to rule but is not thereby automatically entitled to obedience from its subjects. This possibility is illustrated by the situation where one state is occupied by another but where the occupier implements a just social order. Even if the occupier is just and therefore entitled to rule, the occupation can be condemned as inconsistent with principles of legitimacy premised in the value of self-determination (Stilz, 2019, p. 92). This position is not available if one believes that political legitimacy and justice are overlapping normative principles such that justice is all there is to political legitimacy.

Following Raz’s (1986) influential “normal justification thesis,” an authority is morally justified (or legitimate) only if it provides reasons that apply to subjects independently. The state is morally justified only if compliance makes subjects better comply with the demands of practical reason; the authority of the state is legitimate to the extent that obedience to the state is morally required.

Raz’s view does not imply that democracy is necessary for justified authority. It is conceivable that a nondemocratic state provides reasons that apply to subjects independently. Democracy may not be a condition for political legitimacy.

Critics of Raz argue that legitimacy depends on the procedural qualities of public decisions and not just on their content. If one has reason to comply with reasons that are created by procedures that embody equal respect and fairness to others, then one has stronger reasons to comply with laws made by democratic procedures than with laws that are not. In fact, noncompliance with the laws of a democratic authority can be contrary to the moral requirement of treating other citizens as equals (Christiano, 2004). The test for legitimate authority is partly procedural and uniquely satisfied by democratic procedures (Hershowitz, 2003; Scott, 2002; Viehoff, 2011).

The position defended by critics of Raz may be grounds to believe that democracy is necessary for the state to be legitimate. It does not follow that democracy is sufficient for legitimate authority, however. Legitimate authority is, following Raz, premised on de facto authority. The state has de facto authority only if it is effective in securing compliance and is widely believed to be legitimate. A democratic state may or may not be a de facto authority in this sense, and democratic procedures are consequently insufficient to establish that political institutions enjoy legitimate authority.

The Democratic People

Debates in democratic theory have brought to attention the intricacies of the boundaries of the people. Democracy is rule by the people; hence democracy presupposes criteria that delimit the people as a collective entity. The conventional understanding is that the demos (the people entitled to participation) comprise the citizens of the state. A democratic state is inclusive only if citizens are entitled to participate in collective decisions through voting rights and other legally regulated means. Of course, given that democracy is not exclusively an attribute of the state, this claim can be stated more generally as the requirement that the demos include all the members of the relevant unit. For a unit to be democratic, its members should be included in the demos. But this is precisely the claim that turns out to be more complicated than expected.

The consequent “boundary problem of democracy” raises three distinct challenges. The first is whether democracy does, in fact, require that the demos include all members of the unit. The second challenge is whether membership of the unit is necessary for membership in the demos. The third and final challenge concerns the boundaries of the unit itself. Is democracy consistent with taking the boundaries of the unit for granted or does a democratic people depend on the legitimacy of the unit and its boundaries?

The first challenge stems from the observation that states recognized as democratic in fact are never fully inclusive. Nowhere is membership in the state, by means of citizenship, sufficient for rights to democratic participation. This is clear from the practice of excluding children from the vote but also from regular exclusion of people with mental disabilities, prisoners, expatriates, and “nomadic peoples” (Carlsen Häggrot, 2018). The question is when the exclusion of citizens from the right to vote and from participation is either normatively justified or consistent with the conceptual criteria of democracy (Beckman, 2009)?

The second challenge is to question that membership in a unit is necessary for inclusion in the demos. In the context of the state, migration and increasing human mobility virtually ensure that not all residents in the territory of the state are citizens. Moreover, because of the trans-border effects of the laws and policies enacted by the state, it is clear that not just the citizens of the state are affected by public decisions. The state is a leaking “container” with economic, social, and environmental consequences extending far beyond the territory of the state. In view of these facts, the principle that membership in the unit is necessary for democratic inclusion is problematic. In lieu of membership, one might think that the demos should presumptively include anyone relevantly affected by collective decisions (Goodin, 2007) or that the demos should presumptively include anyone subjected to the normative authority of the unit (Beckman, 2014).

The third challenge is highlighted by the fact that the boundaries of units are often contestable or “debated” (Miller, 2014). The fact that the borders separating one unit from another are contested puts pressure on two of the premises that undergird the claim that democratic inclusion applies to the members of the political unit. The first is that democratic participation contributes to the legitimacy of the unit. This is by no means evident in cases where the boundaries of the unit in which democratic participation takes place are either unjust or illegitimate. Instead, there might be reason to think that the relationship between democratic participation and legitimacy is the reverse: The “criteria of the democratic process presuppose the rightfulness of the unit itself” (Dahl, 1989, p. 207). The second premise under pressure is that democratic participation and inclusion are desiderata only for decisions made within a given unit. If the boundaries of the units are normatively problematic, the claim can be made that also the creation of units must be subject to democratic procedures. The boundaries of a unit for democratic decision making are legitimate only if they are established by democratic decisions (Abizadeh, 2008).

The three challenges to the democratic people are distinct, although they may be actualized at the same time. An illustrative case are so-called independence referendums, or what is also known as “constitutive referendums.” In the situation where the people are to decide on the future status of the political unit, the question unavoidably arises regarding who among the members should be included in the demos, what the relevant criteria of membership should be, and what the borders of the political unit are or should be (Tierney, 2012: Ziegler et al., 2014).

Through the various aspects of the boundary problem runs contrasting perceptions about what democracy is. Is the boundary problem a problem for the justification of democratic institutions, a problem in the specification of the meaning of collective self-rule, or a problem for the normative understanding of democracy as either justice or legitimacy? The debate on the extent of voting rights in a democracy, the criteria for democratic inclusion and the boundaries of political units illustrate the continuing relevance of the three accounts of democracy identified here.

Democracy’s Meaning and Value

Democracy as a property of collective decisions is used as a description of either a specific set of political institutions or collective self-rule. While the distinction between them is not always maintained, there is little fundamental disagreement on their meaning. Major disagreements are nevertheless found on the grounds for valuing and ultimately justifying democracy in either of these senses. A basic contrast is between the claim that democracy is instrumentally valuable and the claim that democracy is valued for non-instrumental reasons. Mixed accounts, according to which democracy is recognized as both instrumentally and non-instrumentally valuable, offer a promising attempt to bridge these conflicting views.

Democracy is also used in a third sense, as a normative conception of either justice or political legitimacy. Theories of this kind do not explore reasons for the justification of democracy but instead take democracy to be among the reasons that justify or legitimize collective decisions. Yet, whether democracy is a requirement of justice or a requirement of political legitimacy is controversial. The conceptual and normative controversies in democratic theory are well illustrated by the debate on the boundaries of the democratic people. Though extraordinary progress has been made in clarifying the nature of and the reasons for democracy, it is clear that democratic theory is still in its infancy and that many issues remain to explore and discover.


  • Abizadeh, A. (2008). Democratic theory and border coercion: No right to unilaterally control your own borders. Political Theory, 36, 37–65.
  • Acemoglu, D., Naidu, S., Restrepo, P., & Robinson, J. A. (2019). Democracy does cause growth. Journal of Political Economy, 127(1), 47–100.
  • Anderson, E. (2009). Democracy: Instrumental vs. non-instrumental value. In T. Christiano (Ed.), Contemporary debates in political philosophy (pp. 213–227). Oxford University Press.
  • Arneson, R. (2003). Defending the purely instrumental account of democracy. Journal of Political Philosophy, 11, 122–132.
  • Bauböck, R. (2017). Political membership and democratic boundaries. In A. Shachar, R. Bauböck, I. Bloemraad, & M. Vink (Eds.), The Oxford handbook of citizenship (pp. 524–550). Oxford University Press.
  • Beckman, L. (2009). The frontiers of democracy: The right to vote and its limits. Palgrave Macmillan.
  • Beckman, L. (2014). The subjects of collectively binding decisions. Ratio Juris, 27, 252–270.
  • Beckman, L. (2017). Deciding the demos: Three conceptions of democratic legitimacy. Critical Review of International Social and Political Philosophy, 15(2), 412–431.
  • Beitz, C. (1989). Political equality. Princeton University Press.
  • Berlin, I. (1965). The thought of Alexis de Tocqueville. History, 50(169), 199–206.
  • Brighouse, H., & Fleurbay, M. (2008). Democracy and proportionality. Journal of Political Philosophy, 18, 137–155.
  • Buchanan, A. (2002). Political legitimacy and democracy. Ethics, 112(4), 689–719.
  • Carlsen Häggrot, M. (2018). The right to vote and nomadic voter enrolment. Citizenship Studies, 22, 725–744.
  • Chambers, S. (2004). Democracy, popular sovereignty, and constitutional legitimacy. Constellations, 11, 153–173.
  • Chan, J., & Miller, D. (1991). Elster on self-realization in politics: A critical note. Ethics, 102(1), 96–102.
  • Chang, R. (2009). Voluntarist reasons and the sources of normativity. In D. Sobel & S. Wall (Eds.), Reasons for action (pp. 243–271). Cambridge University Press.
  • Christiano, T. (2004). The authority of democracy. Journal of Political Philosophy, 12(3), 266–290.
  • Christiano, T. (2008). The constitution of equality. Oxford University Press.
  • Christopher, W. (1996). Liberalism, samaritanism, and political legitimacy. Philosophy & Public Affairs, 25(3), 211–237.
  • Cohen, J. (1986). An epistemic conception of democracy. Ethics, 97, 26–38.
  • Cohen, J. (1997). Deliberation and democratic legitimacy. In J. Bohman & W. Rehg (Eds.), Deliberative democracy (pp. 67–91). MIT Press.
  • Colburn, B. (2010). Anti-perfectionisms and autonomy. Analysis, 70, 247–256.
  • Coleman, J., & Ferejohn, J. (1986). Democracy and social choice. Ethics, 97, 6–25.
  • Collier, D., & Adcock, R. (1999). Democracy and dichotomies: A pragmatic approach to choices about concepts. Annual Review of Political Science, 2(1), 537–565.
  • Coppedge, M., Lindberg, S., Skaaning, S.-E., & Teorell, J. (2015). Measuring high level democratic principles using the V-Dem data. Working Paper Series 2015, No. 6. The Varieties of Democracy Institute.
  • Dafoe, A., Oneal, J. R., & Russett, B. (2013). The democratic peace: Weighing the evidence and cautious inference. International Studies Quarterly, 57, 201–214.
  • Dahl, R. (1989). Democracy and its critics. Yale University Press.
  • Dahl, R. (2005). What political institutions does large-scale democracy require? Political Science Quarterly, 120, 187–197.
  • Dahl, R. (2006). On political equality. Yale University Press.
  • De Mesquita, B., Siverson, R. M., & Morrow, J. D. (2003). The logic of political survival. MIT Press.
  • Destri, C. (2021). Extrinsic democratic proceduralism: A modest defense. Res Publica, 27, 41–58.
  • Dryzek, J. S. (2016). Can there be a human right to an essentially contested concept? The case of democracy. The Journal of Politics, 78, 357–367.
  • Dworkin, R. (2006). Is democracy possible here? Harvard University Press.
  • Dworkin, R. (2011). Justice for hedgehogs. Harvard University Press.
  • Elster, J. (1997). The market and the forum: Three varieties of political theory. In J. Bohman & W. Rehg (Eds.), Deliberative democracy (pp. 3–34). MIT Press.
  • Estlund, D. (2008). Democratic authority. Princeton University Press.
  • Estlund, D., & Landemore, H. (2018). The epistemic value of democratic deliberation. In A. Bächtiger, J. S. Dryzek, J. Mansbridge, & M. Warren (Eds.), The Oxford handbook of deliberative democracy (pp. 113–132). Oxford University Press.
  • Gerring, J., Thacker, S. C., & Alfaro, R. (2012). Democracy and human development. The Journal of Politics, 74, 1–17.
  • Goodheart, M. (2005). Democracy as human rights: Freedom and equality in the age of globalization. Routledge.
  • Goodin, R. (2007). Enfranchising all affected interests, and its alternatives. Philosophy and Public Affairs, 35, 40–68.
  • Goodin, R., & Spiekermann, K. (2018). An epistemic theory of democracy. Oxford University Press.
  • Griffin, C. G. (2003). Debate: Democracy as non-instrumentally just. The Journal of Political Philosophy, 11(1), 111–121.
  • Gupta, A. (2021). Definitions. In E. N. Zalta (Ed.), Stanford encyclopedia of philosophy. Stanford University Press.
  • Hershowitz, S. (2003). Legitimacy, democracy, and Razian authority. Legal Theory, 9, 201–220.
  • Kalyvas, A. (2005). Popular sovereignty, democracy, and the constituent power. Constellations, 12, 223–244.
  • Kelsen, H. (1955). Foundations of democracy. Ethics, 66(1), 1–101.
  • Kolodny, N. (2014a). Rule over none II: Social equality and the justification of democracy. Philosophy and Public Affairs, 42(4), 287–336.
  • Kolodny, N. (2014b). Rule over none I: What justifies democracy? Philosophy and Public Affairs, 42(3), 195–229.
  • Landemore, H. (2013). Democratic reason. Princeton University Press.
  • Langvatn, S. (2016). Legitimate, but unjust; just, but illegitimate: Rawls on political legitimacy. Philosophy & Social Criticism, 42(2), 132–153.
  • Mackie, G. (2003). Democracy defended. Cambridge University Press.
  • Mill, J. S. (2010). Considerations on representative democracy. Cambridge University Press. (Original work published 1861)
  • Miller, D. (2014). Debatable lands. International Theory, 6, 104–121.
  • Munck, G. L. (2009). Measuring democracy: A bridge between scholarship and politics. Johns Hopkins University Press.
  • Naess, A. (1956). Democracy, ideology and objectivity. Oslo University Press.
  • Ober, J. (2007). Natural capacities and democracy as a good in itself. Philosophical Studies, 132(1), 59–73.
  • Pasquino, P. (2011). The nature and limits of the majority principle. Books & Ideas.
  • Pettit, P. (2012). On the people’s terms. Cambridge University Press.
  • Popper, K. (1963). Conjectures and refutations: The growth of scientific knowledge. Routledge.
  • Rawls, J. (1971). A theory of justice. Oxford University Press.
  • Rawls, J. (1993). Political liberalism. Columbia University Press.
  • Raz, J. (1986). The morality of freedom. Clarendon Press.
  • Riker, W. (1982). Liberalism against populism. Freeman.
  • Robinson, H. (1963). Definition. Oxford University Press.
  • Ross, A. (1952). Why democracy? Harvard University Press.
  • Rostbøll, C. (2015). The non-instrumental value of democracy: The freedom argument. Constellations, 22(2), 267–278.
  • Rousseau, J.-J. (1997). The social contract and other later social writings. Cambridge University Press. (Original work published 1762).
  • Saffon, M. P., & Urbinati, N. (2013). Procedural democracy: The bulwark of equal liberty. Political Theory, 41(3), 441–481.
  • Schumpeter, J. (1950). Capitalism, socialism and democracy. Harper & Brothers.
  • Scott, J. S. (2002). Authority. In J. Coleman & S. Shapiro (Eds.), The Oxford handbook of jurisprudence and philosophy of law (pp. 382–440). Oxford University Press.
  • Sen, A. (1999). Development as freedom. Oxford University Press.
  • Shapiro, I. (1997). Components of the democratic ideal. In A. Breton, G. Galeotti, P. Salmon, & R. Wintrobe (Eds.), Understanding democracy: Economic and political perspectives (pp. 211–248). Cambridge University Press.
  • Skaaning, S.-E. (2021). Democracy: Contested concept with a common core. In G. Crawford & A. Abdul-Gafaru (Eds.), Research handbook on democracy and development (pp. 27–44). Edward Elgar.
  • Stilz, A. (2019). Territorial sovereignty: A philosophical exploration. Oxford University Press.
  • Talisse, R. (2003). Can democracy be a way of life? Deweyan democracy and the problem of pluralism. Transactions of the Charles S. Peirce Society, 39(1), 1–21.
  • Tierney, S. (2012). Constitutional referendums: The theory and practice of republican deliberation. Oxford University Press.
  • Valentini, L. (2012). Assessing the global order: Justice, legitimacy, or political justice? Critical Review of International Social and Political Philosophy, 15(5), 593–612.
  • Väyrynen, P. (2013). Grounding and normative explanation. Proceedings of the Aristotelian Society, Supplementary Volumes, 87, 155–178.
  • Viehoff, D. (2011). Procedure and outcome in the justification of authority. Journal of Political Philosophy, 19, 248–259.
  • Waldron, J. (1999). Law and disagreement. Cambridge University Press.
  • Waldron, J. (2010). A majority in the lifeboat. Boston University Law Review, 90, 1043–1057.
  • Waldron, J. (2016). Political political theory. Harvard University Press.
  • Wall, S. (2007). Democracy and equality. The Philosophical Quarterly, 57, 416–438.
  • Walzer, M. (1981). Philosophy and democracy. Political Theory, 9, 379–399.
  • Weale, A. (2007). Democracy. Palgrave Macmillan.
  • Weale, A. (2018). The will of the people: A modern myth. Polity Press.
  • Wellman, C. (1996). Liberalism, samaritanism, and political legitimacy. Philosophy & Public Affairs, 25, 211–237.
  • Ziegler, R., Shaw, J., & Bauböck, R. (2014). Independence referendums: Who should vote and who should be offered citizenship? (Research Paper No. 90). Robert Schuman Centre for Advanced Studies and European University Institute.
  • Ziliotti, E. (2020). Democracy’s value: A conceptual map. Journal of Value Inquiry, 54, 407–442.


  • 1. There may be additional uses of “democracy.” No less than 200 distinct definitions of democracy were documented by Arne Naess already in the 1950s (Naess, 1956). A large portion of these and later uses of the term can arguably be subsumed under one of the four senses identified here.

  • 2. See Talisse (2003) for a critical appraisal of Dewey’s position and Anderson (2009) for a more sympathetic take on the view that democratic government is a manifestation of the democratic way of life.

  • 3. For example, Beitz (1989) and Pettit (2012, p. 180) both dismiss the preoccupation with definitions of democracy as mere semantic quibbles.

  • 4. The distinction between legitimacy and justice is more pronounced in the later writings of Rawls (see Langvatn, 2016).