Introduction
Political authority claims the power to change the normative situation of people, such as their rights and duties. How can political authority be legitimate? Political philosophy provides many justifications, but ultimately three of them figure most prominently: reason, consent or will, and proceduralist theories.Footnote 1 Support for each one of these justifications can also be found in political discourse. Politicians often claim that their policies are just, wise, or in accord with reason; that citizens are bound by these decisions because they elected the politicians and thereby consented to the resulting policies; and that the procedures by which these decisions were reached count as fair or egalitarian ones.
Reason, consent, and proceduralist theories of political authority capture some aspects of what renders political authority legitimate. However, none of these theories, nor a combination of them, provides an adequate answer. The theories correctly identify that reason, consent, and fair procedures matter but fail to explain why. Further, they fail to account for a particularly important aspect of the decisions made by a political authority, namely, that these are our decisions and that in order for them to count as our decisions, the authority ought to make decisions that cohere as much as possible with our perspectives. Political authority, in other words, ought to represent us. By subjecting the authority to the citizens’ perspective, its decisions also satisfy an attributability condition, meaning that citizens bear responsibility for such decisions. We maintain that political authority can count as legitimate by the standards articulated by reason, consent, and proceduralist theories and yet remain tyrannical if it speaks for us, rather than in our name, which is to say, if it fails to represent us.
Consider a system of laws that is perfectly just, enjoys the consent of the governed, and is procedurally fair. We argue that it can nevertheless count as tyrannical. As mentioned in the Introduction, Exodus 24 tells us about Moses, who “took the Book of the Covenant, and read it to the people; and they said, All that the Lord hath said we will do, and we will hear.” The traditional interpretation is that the Israelites had promised to comply with God’s rules before even knowing what the rules were.
This divine system of rules seems to satisfy the basic requirements set out by the theories of political authority of reason, consent, and proceduralism. The Israelites accepted rules that they presumed to be just or, at least, grounded in reason (as they were given by God). They also consented to the authority of God and the laws it imposes on them. Finally, the procedural aspects of the making of the rules were presumably (or at least could be) fair and egalitarian ones. Perhaps, for instance, each person had an equal say in making the decision to submit to God’s authority. Nevertheless, the system should count as tyrannical as its rules were made for the Israelites rather than in their name. It seems to us that none of the traditional theories of authority (reason, consent, and procedure) could explain the flaw in this biblical account of the authority of God.
By contrast, we argue that a certain conception of the public provides an adequate basis for explaining what political authority is and what renders it legitimate. To assert that an institution or an official is public implies that the institution or the official speaks and acts in the name of its citizens, which means that it represents them. Therefore, its decisions can be attributed to them. Consequently, those represented can regard themselves as the authors of the rules; they recognize that (at least generally) those rules accommodate their concerns and reflect their worldview. They can, therefore, truthfully assert that the rules are their rules, which they have authored.
Representation, as we use the term here, does not imply voting or consent. While voting may be conducive to representation, on our view, representation implies that the authority endorses the worldview of its subjects. Yet, as we later establish, endorsing the worldviews or the perspectives of the subjects does not necessarily imply endorsing their raw preferences or judgments. Representation sometimes requires interpretation that is based on the comprehensive worldview of citizens. Therefore, faithful representation may sometimes yield decisions that deviate from those that would have been chosen by citizens. Representation also implies that the represented may bear responsibility for the decisions of the authority, as, in reality, it is the worldview of the represented that is being replicated in the decisions.
The proposed account intervenes in the contemporary theory of political authority in two ways. First, we develop a theoretical framework that better explains the role of reason, consent, and procedure in rendering political authority legitimate. Second, our account establishes what the reason, consent, and proceduralist theories of legitimate political authority fail to establish – that (unlike other typical authoritative entities such as military commanders, bosses, or parents) political authority is fundamentally nonhierarchical.
Political authority is a subset of practical authority. However, it is fundamentally different from typical cases of practical authority. Practical authorities are generally identified with hierarchical relationships. The authority decides for its subjects what to do or refrain from doing and the subjects, in turn, are required to conform to that decision. The legitimacy of practical authority depends on justifying the hierarchy between authority and its subjects – for instance, that submission to authority guides the subject to act in accordance with reason, or that it serves the subject’s own interests, or that the subject actually consented to be guided by the authority.
By contrast, we argue that political authority is nonhierarchical; it is justified because it eliminates rather than justifies the hierarchy between the authority and its subjects. Public officials in a position of authority do not decide for those subject to their rule but rather do the deciding in the name of the subjects. The answer to the legitimacy question addressed to Moses (by the Israelite who was rebuked by Moses) – “Who made thee a prince and a judge over us?” – is not that the thee is necessarily more qualified or better positioned to make decisions for us; nor is it necessarily consensual submission to the rule of the thee, and it is not necessarily a byproduct of a fair, democratic, or egalitarian procedure. Rather, it is that the thee is in reality, or, at least, can justifiably count as us. This explains why political authority is necessarily public: It represents those who are subject to it and, consequently, those who are subject to it are, in principle, accountable for the authority’s decisions.Footnote 2 It further explains the fact that political authority is a paradigm case of authority that is freedom-facilitating rather than freedom-limiting.Footnote 3
The crucial difference between political authority and other types of practical authority is that political authority is fundamentally public. To be politically authoritative, an institution ought to be public in the right sense. To be a public authority, properly conceived, implies that those in authority speak and act in the name of the public. To do so, political authority must satisfy what we label a perspectival condition: Its decisions must reflect (in the right way) the perspectives of those who are subject to it. Perspective-taking also entails attributability. If the political authority satisfies this condition, its decisions can be attributed to its subjects, as ultimately it is the perspectives of its subjects that dictate the rules. Neither acting in the name of those subject to the authority nor the attribution of so acting to them appears (typically) to be relevant to the legitimacy of the authority held by other (nonpolitical) types of practical authority, such as parents or bosses. The boss in the workplace (or a parent in a family) does not act in the name of the employee (or the child), and his or her decisions are (at least typically) not attributable to his or her employees (or children).Footnote 4 One central indication supporting this view is the sharp and principled distinction between private citizens and public officials. What makes public officials public is the fact that they speak in the name of (rather than for) the public, and their decisions are thereby authoritative. Hence, an understanding of what makes public officials and institutions genuinely public provides the key to the legitimacy of political authority.
The argument proceeds through the following stages. Section I examines different types of practical authority. It shows that political authority is distinctive in that it rests on a certain understanding of representation. Unlike other types of authority, political authority does not only promote interests; instead, it (ideally) speaks and acts in the name of its subjects.
Section II considers what political representation consists of. We divide this section into two. We first ask what it means to represent an individual, and then extend the analysis to the case of representing a group of people. This account characterizes the concept of what we call perspectivism, which is a way of making decisions as if they were made from the perspective of those who are subject to them. Section II also explains the precise sense in which political authority is public and why it might count as legitimate.
Finally, Section III points out some essential contrasts and similarities between the public conception of political authority and competing accounts grounded in reason, consent, and procedure (especially democratic decision-making procedures). The discussion explains the sense in which considerations of reason, consent, and procedural fairness can be accommodated by the proposed framework.
I Practical Authority, Political Authority, and Hierarchy
We characterize the difference between political and nonpolitical practical authority in terms of hierarchy. Doing so requires unpacking the general category called practical authority by differentiating between its most prominent instantiations.
Practical authorities can be characterized as relationships between persons. They all feature a person or an entity having a certain power over others with respect to certain matters. The power at issue entitles its holder to change, by way of speaking and/or acting, the normative situation of others. Most typically, it involves the making of binding decisions concerning what those who are subject to the authority ought (or ought not) to do. At times, the power of making these decisions renders more determinate existing reasons that would apply to its subjects independently of the decision. For instance, a police officer may require drivers to slow down in the face of a sudden, imminent storm. At other times, however, this power to make binding decisions includes the creation of reasons that would not have existed independently of the authority’s directive. For instance, an employer may demand that his or her employee complete a task by tomorrow evening. In what follows, we assume that both versions are plausible accounts of practical authority’s normative power.Footnote 5
The power of practical authorities to make binding decisions typically manifests itself in one of two ways. Most types of practical authorities, including bosses at the workplace or parents in the family, gain their power by virtue of promoting certain interests: the interests of the authority itself, the interests of those who are subject to the authority, or the interests of third parties. Political authority promotes interests as well, but, contrary to these other instances of practical authorities, it does so by speaking in the name of those who are subject to it. Let us explain.
Some practical authorities gain authoritative powers over others in order to promote the interests of the authority itself. Thus, a sole proprietor of a business decides for her or his workers what they may or may not do. There are limits to this power, to be sure, but the ultimate purpose of establishing the authority is to promote the interests of the proprietor, for example, her or his material gains. Second, some other practical authorities exercise their decision-making authority in order to promote the interests of their subjects. Parents’ authority over their children fits this description. At other times, deciding for their subjects is an integral aspect of the role of a professional authority. For instance, a lifeguard instructing someone to refrain from dive-jumping into a pool of shallow water is speaking and acting to promote that person’s own interests. Third, and finally, practical authorities often promote the interests of third parties. For instance, an entity such as Amnesty International is a corporation that wields the (circumscribed) authority to decide what its subsidiaries, workers, and agents should do in response to certain human rights violations. Such an authority is not claiming to promote the interests of those who hold the authority or those who are subject to the power of the authority but to promote the interests of victims of human rights violations. It instructs its employees how to act in order to further the interests of victims whom it decides to help.
Despite all the differences, these three variants of authority relationship are all designed to promote interests or, more generally, valuable ends in a certain way: The authority decides for its subjects what actions to take in order to promote the interests at issue. It is a top-down interaction in the sense that the decision does not emanate from the subject and so cannot be attributed to him or her. More specifically, it makes no essential reference to what the subject wants or judges to be just or fair, or who she or he is. Consequently, the resulting decisions are typically attributed exclusively to the authority itself. Simply put, practical authorities of the sorts we have just discussed do not exercise their powers in the name of their subjects.
What renders such practical authorities legitimate? The answer cannot rest solely on the fact that the authority promotes valuable interests of the authority itself, of those who are subject to the authority, or of some third parties. While promoting such interests is desirable, doing so may not (at least typically) justify the subjugation of individuals to authoritative directives issued by others. There are, however, circumstances when a person has such a duty, or at least when the authority has the power, to inflict a sanction on a person for failing to act in certain ways. We discuss here two such circumstances: vulnerability and consent.
Concerning vulnerability, promoting the well-being, say, of children may justify the circumscribed authority of their parents precisely because of children’s physical and emotional vulnerability.Footnote 6 Further, the exercise of parental authority often does not violate children’s autonomy as children are not fully autonomous agents. At other times, vulnerability does not emanate from the subject’s own peculiar condition, such as childhood, but rather from extrinsic circumstances that place him or her in a condition of imminent danger. For example, a plane crash features an emergency situation that renders its surviving passengers unusually vulnerable; it can therefore justify the practical authority of a highly experienced and ingenious flight attendant to direct the rescue operation for the survivors.Footnote 7
Concerning consent, committing oneself to an authority’s rule can render this authority legitimate.Footnote 8 Doing so need not necessarily rest on the belief (or be motivated by the belief) that the authoritative directives promote the interests of the consenting persons. Consent may also be motivated by extrinsic considerations, for example, salaries and other benefits. Consensual authority is typically terminable at will at any point. For instance, a worker can end his or her subjection to the employer’s rule simply by opting out (say, by way of resigning). Consent, moreover, cannot render a person liable to specifically perform his or her promise to work for the relevant organization.
Political authority is different. Its legitimacy does not depend upon either preexisting vulnerability or consent (or both). To begin with, political authority is not limited to people experiencing some form of reduced cognitive or other agential underfunctioning, nor to emergency situations. Further, political authority is consent-independent at the formation stage as well as during the course of its operation. Thus, it can arise in the absence of (explicit or implicit) consent, and it can arise even when its subjects actively refuse to give their consent.Footnote 9 Further, subjugation to political authority is typically not renounceable at will. Unlike employment relations, subjects of political authority are often required to leave behind their place of residence and possibly even join – that is, become subject to the authority of – another political community.Footnote 10
Against this backdrop, we submit that the legitimacy of political authority is especially challenging. Lived experience and common sense suggest that public officials with lawmaking, law-enforcing, and related powers do not fit the consensual model of a business owner or the vulnerability model underlying parents’ or lifeguards’ authority. We argue that political authority gains its power not merely by virtue of the interests it serves and how well it serves them, but rather by virtue of its capacity (or standing) to speak and act in the name of those subject to it.
Consider the distinctive ways in which our language treats political authority vis-à-vis other types of authority, namely hierarchical authority relationships. Key to this difference lies in how we think and talk about representation. Legislators are often labeled representatives, and the lower house of the US Congress is called the House of Representatives. Further, political leaders often perceive their role in terms of representation. The legislature is not the only public institution that invokes the language of representation to render legitimate its authority. In important ways, both the executive and judicial branches of government draw on the language of representation.Footnote 11
The use of the term “representative” in this context is not only prevalent but also unique to political authority. While the term “representation” is often used in other contexts, such as with reference to lawyers who represent a client or agents who represent their principals, it is rarely if ever used to denote practical authorities other than in the case of political authority. Nonpolitical practical authorities are virtually never described as representatives. Employers, army commanders, parents, or schoolteachers – who are all paradigmatic cases of practical authorities – are not considered representatives.
II Nonhierarchical Authority
How could the relationship between authority and subject be nonhierarchical? We argue that political authority is one that speaks not for us, but rather in our name. One of the implications of speaking in our name is that the decisions made by the authority can be attributed to us. To unpack these claims, we divide this section into three. We first examine what it means to make decisions in the name of an individual. Second, we extend the analysis to the case of group representation, especially political representation. Lastly, we establish that speaking in the name of an agent may give rise to duties on the part of the agent.
A One-to-One Representation
We examine here two ways of accounting for nonhierarchical authority relations: identification and perspectivism. Identification denotes the attitude of a citizen who regards herself or himself as belonging to the state and, accordingly, as being accountable (in some sense) for its decisions. Perspectivism takes up the commitment and ability of the authority to look at the world from the perspective of the subject. We criticize the former and defend the latter. Recall that we begin with the simple case of a one-to-one authority relationship before we extend the analysis to the context of political representation addressing the question of how decisions made by representatives can be attributed to a plurality of people.
Identification provides a straightforward sense in which the authority/subject relationship can be made nonhierarchical. By identifying with the state or the city, subjects may also come to perceive the decisions made by state or city authorities as ultimately theirs.Footnote 12 On this view, what links the subjects to the authority’s decisions is a certain attitude, a feeling of trust in, or a sense of belonging to the particular entity on behalf of which the authority acts. A binding directive is not merely imposed upon us by a team, a city, a state, or some such entity. Instead, because the city (et alia) is ours, and because we equate the city with its municipal authorities acting on its behalf, the directives of the authorities at issue are essentially our collectively self-imposed directives.
However, identification should be rejected on several grounds. As Joseph Raz has rightly observed, subjects’ identification with an authority cannot account for the legitimacy of authority as it does not serve as a necessary condition for legitimacy. More specifically, it leaves out many subjects who are indifferent to, not to mention alienated from, the entity in authority.Footnote 13 For them, authority cannot be justified in terms of identification.
Not only does identification fail to provide a necessary condition for grounding nonhierarchical authority, but it is also an insufficient one. The latter failure is instructive as it will help identify what might ultimately justify nonhierarchical authority. Perspectivism, rather than identification, establishes the requisite linking of subjects to the authoritative decision-maker.
To see why identification is insufficient to account for authority, consider the case of people who fully identify with the city in which they reside. They think and talk about the city as “our city” and understand its ordinances as “our ordinances” even when they sometimes disagree on the merits. That said, that attitude does not give rise to a duty on the part of the citizens to defer to the authority of the city officials. Nor does it give rise to attributability. Subjects who wholeheartedly identify with the purported institution have no necessary role in the making of binding decisions. The subjects’ perspectives as to what decisions to make are not (necessarily) reflected in these decisions. In other words, identification with the authoritative institution does not imply convergence between the directives issued by the authority and the perspectives of the subjects of the authority. While the subjects may feel empathy toward their city or state, their substantive judgments, preferences, or essential features need not affect, let alone dictate, the directives issued by the authority.Footnote 14 To this extent, the subjects may identify with the authority or the directives issued by the authority, but the directives issued by the authority are not their (the subjects’) directives, and so these directives cannot be attributed to them.
Identification fails because it can only capture the authority side of the equation. It cannot account for the other side of the equation, namely, what the authority must do (and possibly also believes to be doing) in order for its decisions to count as being made in the name of the subjects. It imposes no constraints as to the content of the directives made by the authority. It does not presuppose that those directives must ultimately emanate from the subjects of the authority, their judgments, preferences, or, most abstractly, their perspectives. The mere fact of viewing the entity in authority or the unit it governs as ours does not make its decisions ours.
To understand and justify nonhierarchical authority, we need to look at the other side of the equation, namely the authority itself and the ways it makes decisions. Our starting point is that political authority looks at the world from the perspective of the citizen. When this condition obtains, the citizen can truthfully say that he or she is being represented by the authority. Hence, one can look at the relationship between the authority and the citizens in two complementary ways: The authority looks at the world from the perspective of the citizen and, consequently, the citizen is represented by the authority. A citizen is represented by the authority when the authority satisfies the perspectival condition: when it looks at the world from her or his perspective, rather than making decisions for her or him.
Satisfying the perspectival condition explains why the decisions of the authority can be attributed to the citizen as, in reality, it is the represented who made the decision. Attributability implies that under the appropriate conditions a citizen can be held accountable for the decision as if it is hers or his, although she or he has not made it. Attributability does not entail that a citizen can be blamed or prosecuted for the acts of her or his government. It does entail, however, that she or he ought to take some responsibility for the decision and that she or he cannot remain indifferent to it. We leave the analysis of responsibility and its normative implications to another occasion.Footnote 15
The perspectival condition, as we have said, is designed to render the relationship between authority and its subject nonhierarchical. Consequently, the connection between identification and attribution is turned on its head: Instead of expecting the subject to identify with the authority, the authority is required to take the perspective of the subject, which involves deference to this perspective, in order for the authority’s binding decisions to be attributable to this subject.
The real question is what “looking at the world from the perspective of the represented” might mean so that the representative can be properly described as deciding in the name of the represented. When could a decision made by a political authority count as that of the citizens – that is, a decision which the citizens can properly regard as theirs? Is it at all possible? If so, what does it require? Or, put it in terms of representation, what should an authority do in order to be considered as being representative of the citizens?
We address this question by using the notion of perspective. The term is used here as an abstract construct that can take different meanings. The question of what it takes for an authority to look at the world from the perspective of its subjects can have different answers because it depends upon the preliminary question of what captures a normatively significant feature of the subject, which can be foisted upon him or her and yet be attributed to him or her. Singling out such a feature may vary depending on the subject matter of the decision and on any number of factors and contexts. Hence, there could be many ways adequately to represent a person, and it is not our ambition to provide a comprehensive account of them all. We believe that there is a range of meanings that can apply to different institutional contexts and different societies. Moreover, there are more and less appropriate accounts of representation. There are also accounts of representation that are erroneous.
In this respect, our discussion does not aim to develop a substantive theory of representation but rather a theoretical framework that creates normative space for different substantive accounts. Each account develops a conception of representation that explains why directives issued by state officials might be attributed to their subjects.Footnote 16 Accordingly, we do not aim to defend or criticize such accounts. Instead, we focus our discussion on those varieties of representation that form a robust connection between the decision made by the representative and the perspective of the person represented.Footnote 17 We first explain what perspective-taking consists of, and proceed to distinguish between conceptions of representation based on their ability to meet, or frustrate, this requirement.
Taking the perspective of another person comes down to a combination of a certain attitude of deference and a commitment to making binding decisions accordingly. The representative is required to acquire a deferential stance toward the represented person. It involves the willingness to substitute the former’s judgments and/or worldviews and/or essential features with those of the latter. Such a deferential stance is a form of recognizing the actual features of the represented person’s self as having a controlling influence on the deliberations of the representative.Footnote 18
In addition to a deferential attitude, representation requires adopting a decision-making process capable of accurately identifying and articulating the point of view of the represented person. Such a commitment would typically direct a representative to the actual features of the represented person’s self. Her or his task is to decide on the basis of such features so as to make it as if the represented person is himself or herself making the decision. To the extent that the authority succeeds in representing, the subjects can justifiably claim that they are the genuine authors of the resulting decisions.
To make this account more concrete, we distinguish between two conceptions of representation that have been particularly influential in the history of political thought: the agency view and the essentialist view.Footnote 19 Under the agency view, to take the perspective of the represented requires tracking their judgments, choices, preferences, or some such features. Under the essentialist view, the representative makes judgments that accord with the essential or natural features or identity of the represented. Under this view, the citizen shares with other citizens features that constitute her or his identity, and those should be reflected in politics, in particular in the binding decisions of the state. We regard both conceptions as plausible articulations of representation and do not judge which is better. We take each conception in turn.
Under the agency view, a representative speaks and acts in another person’s name by making decisions that reflect the latter’s judgments, choices, and preferences on the matter at hand. Often this has immediate institutional implications. For instance, the authority of representative democracy is sometimes justified on account of its ability – which may partially turn on the commitment of its representatives – to be guided by the choices of the citizens.
Being guided by the represented decisions, judgments, or preferences is never a mechanical task. It requires the representative to engage in interpretation. The interpretive task can be challenging, indeed. To illustrate, assume that Alan represents Daniel, who is a supporter of capital punishment. Daniel – the represented – grounds his support of capital punishment in considerations of deterrence alone. He believes that capital punishment deters and, therefore, should be implemented by the state. It turns out, however, that capital punishment does not deter. What should the representative Alan do, given the tension between Daniel’s conviction and the truth of the matter? Under one view, Alan should voice Daniel’s view in support of capital punishment. On reflection, however, Alan, who is acutely aware of Daniel’s rationale for supporting capital punishment, may better represent Daniel by voting against capital punishment as only such a decision on his part takes seriously Daniel’s conditional support for capital punishment (that capital punishment is justified only because, and only insofar as, it deters).
Either of these decisions can count as representation as both purport to take the perspective of the represented as the guiding principle. In the language of James Madison, the representatives have “to refine and enlarge the public views, by passing them through the medium of a chosen body of citizens.”Footnote 20 This observation also implies that representation, as we understand it, does not necessarily entail majoritarianism (although this may often be the best institutional mechanism for reaching representative decisions). Alan may better represent Daniel’s perspective under such circumstances by voting against the raw preference of Daniel, namely, against capital punishment and, consequently, Daniel may oppose Alan’s authority despite, or even precisely because of, his being a faithful representative of Daniel’s position.Footnote 21
The mere fact that making either one of these decisions could count as representation does not entail that they are equally compelling. As noted earlier, it is the task of political theory to provide a substantive account of representation. Such a theory should determine what counts as speaking and acting in the name of the represented in general and in particular cases.
Under the essentialist view, the authority defers not to the judgments (et alia) of the represented, but rather to the innate and acquired essential features of the represented. At times, some natural traits (such as gender) may fit this category. Thus, it is sometimes said that only women can represent women on some issues, not necessarily because they hold preferences or judgments in common, but because they somehow make decisions that reflect a set of attributes often associated with femininity.Footnote 22 Other features may be political, pertaining to our nationality, culture, or some other practices that give meaning to our lives. They also may be geographical, as when a person is defined by an area of residence, say a resident of the Sahara or midland Siberia. One’s identity as a working-class person may count as a significant feature, say when a state sets out to determine rules of fair employment in the private sector. These and other features, to a significant degree, shape who we are. Representing a person may involve making decisions that are appropriately reflective of his or her significant features or, at least, which are grounded in the lived experience of the represented. Perhaps this can explain why some countries require newcomers to gain basic familiarity with the local language and culture in order to acquire citizenship. Note that under this view, the judgments or preferences of the person who is the subject of authority are not conclusive. The authority defers to characteristics deeply ingrained in the subject or that constitute parts of the subject’s identity, even if the subject is not fully aware of these or fails to care for them.
Thus, perspective-taking depends on a certain idea about what features of the represented person are normatively significant and precisely when (i.e., with respect to what matters) they are so. In some cases, representation requires making decisions based on what the represented person wants or judges to be right or just, in which case the decisions made by the person in authority stand in for – namely, voice – the preferences or judgments of the authority’s subject. At times, voicing preferences may be straightforward. But at other times, as when the subject remains inarticulate or underarticulate, it might involve an interpretive effort on the part of the representative. The interpretive task is familiar in many different contexts, as when judges construe an ambiguous statutory provision by asking what the intentions or purposes of its legislators were. Alternatively, representation can focus on who the represented person “really” is, in which case the person in authority must make decisions that best reflect certain defining traits of the authority’s subject. In both cases, representation structures a nonhierarchical relationship between authority and subject; the former does not decide for the latter, but rather speaks and acts in the name of the latter.
Note, however, that the agency-based and the essentialist view require that the subjects form opinions, acquire preferences (under the agency view), or, under the essentialist view, that they share certain traits or conditions that determine their identity. In both cases, the subjects are actors who either form opinions or preferences or, alternatively, share certain sensibilities and features that form their identity. We will later see that this has critical institutional implications. For instance, it may provide some support for democratic decision-making processes, although, under this view, democracy has instrumental, rather than intrinsic, value. Democracy enables people to form preferences or opinions or authentic identities and convey those publicly, thereby facilitating representation.
Perspectivism as characterized here may give rise to a forceful objection, namely, the risk of totalitarianism. The representative can order people to act, and can claim to be acting in their name. This may remind us of the notorious claim made by Rousseau that he who refuses to obey the general will should be compelled to do so and thereby forced to become free.
Note, however, that the risk of abuse leading up to totalitarianism is endemic to many theories of political authority. While a dictator may falsely argue that she or he “represents” the people, she or he may also claim that she or he can better identify what reasons for action her or his subjects have and, therefore, that she or he can guide subjects to act in accordance with the demands of reason. On our view, the success of a claim for political authority rests on the fact that the authority succeeds in representing its subjects. Success in such an endeavor may require certain institutional structures, such as democratic elections (in the case of political representation) and other accountability mechanisms. Therefore, such institutions are means that may help tackle abuse and bring about the desired end, namely, representation. But the mere fact that claims of representation are open to abuse does not imply that they are not ultimately what account for the legitimacy of the authority.
The preceding discussion developed the notion of representation as perspective-taking by reference to a simple case of a one-to-one authority relationship. The authority speaks and acts in the subject’s name if, and only if, the former makes decisions that reflect what the subject wants or judges to be just or who she or he is. Political representation concerns speaking and acting in the name of all. It is one thing to consider representation in a one-to-one setting; quite another to talk about attributing the decisions made by a representative to a group. Such an extension requires certain adjustments that transform and weaken the deferential link between each individual and the authority. Strict deference to the perspective of a person may not be feasible in the case of a group. Yet, we maintain, a similar analysis could apply in the case of a group.
B Political Representation
The shift from one-to-one representation to political representation of a group gives rise to two basic questions, a substantive one and an institutional one. Substantively, what does representing a group mean?Footnote 23 Institutionally, how can it be made to work? We mainly focus here on the former because it raises fundamental normative questions, while the latter raises primarily pragmatic or empirical questions (which are discussed in Chapters 3–6).
A one-to-one representation model by perspective-taking poses one major challenge: Precisely what features of the represented person’s self should guide the representative’s decision? Following the distinction introduced between agency and essentialist views of representation, these features should be associated with either what this person wants or judges to be true or fair, or who this person is and what his or her innate attributes are (or a certain combination of both). As we move from one-to-one to group representation, we believe the latter must also address a structurally similar challenge. At times, the significant features of the represented persons are equated with what they want – for instance, striking down laws as unconstitutional is sometimes (or on some views) explained in terms of judges determining what the will of the American people was at the formative period of constitution-making.Footnote 24 At other times, the pertinent features concern who the represented persons are – arguably, when President Biden defended his choice of an unprecedentedly diverse class of cabinet officials, he asserted that “[t]his cabinet will be more representative of the American people than any other cabinet in history.”Footnote 25 His assertion appealed to representation in terms of reflecting who the represented persons – “the American people” – are. Indeed, there is nothing in President Biden’s assertion to indicate that the selected officials have been picked on the ground that their binding decisions reflect what the American people prefer or judge to be true. It is beyond the scope of our argument – which is limited to defending a characterization of political authority in terms of representation – to defend any particular conception of representation. We just note that often both types of representation play a role in our understanding of what representation dictates.
The problem of representing a collective is an acute one. What happens when the relevant features which provide the basis for representation differ among individuals? What if, as is to be expected, the preferences, judgments, and identities of different individuals radically differ? To partially address this concern, our account of political representation is modest, holistic, and (partially) proceduralist. These three features combined together help (at least partially) to bridge the gap between the requirement of perspectivism and the realities of politics, in particular, the heterogeneity of modern society. More specifically they help us to explain why, despite the fact that different persons often have different preferences and hold different beliefs and values, they can nevertheless (at least often) be adequately represented as a group.
Our account is modest in the sense that perfect representation is often impossible, and therefore we typically should resort to achieving a modest goal, namely a sufficient degree of representation. It is holistic in the sense that it is sometimes the case that, overall, the state adequately represents citizens without it being the case that each and every decision of the government represents all citizens. It is (partially) proceduralist since individuals are asked not only to form a perspective on substantive questions but also on proceduralist questions, as they are asked to form opinions together. Procedure-related perspectives are ones that cope with conflicts about substantive questions. Let’s take each in turn.
1 Modesty
It may be impossible for either principled or pragmatic reasons to realize a very high degree of group representation. We suspect that the transition from representing an individual to representing a collective of persons requires lowering the bar of what counts as a legitimate authority such that even systems that are currently not fully – that is, perfectly – representative of each and every individual can be representative (and therefore legitimate) so long as they reach a certain threshold, namely, that they are sufficiently representative.
To be sure, a standard of “sufficiently representative” must meet a substantive threshold in order for representatives to speak and act in the name of the people they represent. Such a threshold should be responsive to the significance of the political decision at issue. For instance, it is one thing to amend the current statutory speed limit; quite another to introduce a major regime change, say, regarding the state’s commitment to religious neutrality. Moreover, such a standard must be sensitive to the particular role of the decision-maker. For instance, judges are tasked with identifying, interpreting, and applying the law. In most legal systems, possessing legal expertise, professional integrity, and commitment to the rule of law is not enough. It is also necessary that judges have the right political identity, namely, citizenship. The most desirable and just legal decision made by a complete alien is, nonetheless, his or her private judgment, as he or she is not a member of the political community subject to the decision.
2 Holism
Often, a citizen may complain that a particular decision does not represent her or him but, nevertheless, concede that, on the whole, she or he is adequately represented. Being politically represented does not require the near-perfect degree of convergence and precision that a one-to-one representation could achieve. More broadly, the government’s representation of all citizens implies that representation cannot be assessed on a case-by-case basis alone, but rather should concern itself with the system of governance as a whole. You might have good reasons to complain that your representative has failed to act on a particular matter; however, such a failing to do so does not imply that this representative, and the system of representative government to which she or he belongs, is overall no longer committed to speaking and acting in your name. This sentiment was evident in the title “Not in Our Name” given to an organization established by opponents of the US government’s war-on-terror policies enacted after the 9/11 attacks. Members of this group sought to disassociate themselves from certain decisions of their representatives without undermining the system of representation as a whole.
3 Proceduralism
Lastly, and perhaps most importantly, perspectives typically consist of not only substantive commitments but also proceduralist ones. Perspectivism requires making decisions that are grounded in what the subjects want and/or judge to be just or reflect their identity. People living in a society share values that bear on both the substantive and the procedural decision-making processes. Hence, perspectivism requires not only endorsing rules concerning substance but also rules concerning procedures. By procedures, we mean not merely processes of decision-making and their rules (say, a majority rule or a notice-and-comment process), but also commitments – such as free speech – that are meant to provide us all with meaningful opportunities to challenge decisions and call for alternative ones.
This facilitates the transition from one-to-one to political representation: It renders political authority compatible with the demands of perspectivism. A shared perspective about procedures (broadly conceived) can bridge the gap between what one believes ought ideally to be done and what one believes ought actually to be done, given unresolvable disagreements on a particular matter. It is typically easier to reach a degree of convergence on proceduralist issues than it is on substantive issues and, to the extent that a shared perspective with respect to procedure prevails, it may provide the prospects for establishing a legitimate political authority.
That said, two possibilities that render representation infeasible cannot be precluded. If either one applies, representation that gives rise to legitimate authority cannot obtain. First, it is possible that a particular community be so radically heterogeneous on questions of fundamental values as to render legitimate government impossible. To represent a political community, certain preconditions should be satisfied, including that the represented people share a common political project such that they can be collectively represented.
Second, it is possible that while most people are adequately represented, there exists a person or a group whose values or circumstances radically diverge so that they are not subject to political authority. That does not imply that they cannot justifiably be punished or even have no duties of obedience resulting from other considerations. For instance, their obedience may be dictated by respect for the fundamental rights or interests of others. But it means that the duties such citizens have, or that the powers that the state may use against them, differ (typically are weaker) from those that other citizens may have. Their duties are not attributable to the fact that the state represents them or that it constitutes a legitimate political authority with respect to them.Footnote 26 One may have certain duties of obedience to the state even when the state is not a legitimate political authority (or, at least, it is not a legitimate authority with respect to oneself).
Representation requires the representatives to endorse the perspective of the citizens. Perspective-taking can account for attributability. If representatives succeed in making a decision that can count as being made from the perspective of the represented persons, it can be attributed to the latter in the sense that the decision is made in their own name, that they are its real authors; the represented can therefore truthfully assert that the state is theirs as they are in charge of its rules.
Achieving this much requires adequate institutional mechanisms to facilitate representation. An election is a standard institutional mechanism that (if properly designed and subject to the considerations discussed in this section) can succeed in facilitating representation. Recent advances in political science further reinforce the point that this is essentially a question of institutional design – or, more specifically, one of the technologies of political representation – as it considers the most effective ways of linking the binding decisions of the representatives to the persons represented.Footnote 27 Decisions that are being made by the representatives gain legitimacy by taking the perspectives of the represented. Consequently, such decisions can also be properly attributed to the represented.
Perspectivism and attributability explain why we regard the relationship between political authority and its subjects as nonhierarchical. We now argue that it is by virtue of being represented that the authority is a public authority. This implies that the subjects, rather than their representatives, effectively get to decide, and are thereby accountable for these decisions. We also establish how this account explains the emergence of political obligations.
C Why Should We Care: Why Public? Why Political Authority?
Whereas an employer or a parent may have practical authority without being public, the state and other political institutions cannot because their authority hinges on the fact that they speak in the name of their citizens. But what does it mean to be public? What makes it the case that an institution counts as a public one? We now wish to explain the precise sense in which political authority is public.
The most basic and least controversial aspect of being a public entity concerns the absence of self-regard. A public entity cannot pursue its own ends because it ought not to have ends of its own. Only private persons are entitled to set ends and pursue them. Beyond this basic liberal observation, however, there can be any number of interpretations of what being public consists of.
One interpretation is that public institutions are defined by the purposes they pursue. According to Arthur Ripstein, “All that is required for the legislative will to be omnilateral is for the distinction between public and private purposes to apply to it in the right way.”Footnote 28 Another variation on this theme places a fiduciary structure on the standard of public purposes. It suggests that entities are public if they assume the role of fiduciaries in relation to their beneficiaries, namely us (the people). As fiduciaries, public entities are required to act solely in our interest.Footnote 29
That said, the public character of an institution is not exhausted by the purposes it advances, nor can it be exhausted by the fiduciary role it undertakes. It is one thing to say that a purpose is public in the sense that it promotes purposes that are conducive to the interests of the public. It is quite another to suggest that the entity pursuing this purpose is, therefore, a public entity. Similarly, it is one thing to determine that an entity owes a fiduciary duty to the people, quite another to deem it a public entity. After all, many nonprofit organizations pursue and declare themselves to be committed to pursuing public purposes. Moreover, they may do so by undertaking the fiduciary duty of loyalty and good faith to the people. However, so acting does not make them public entities in the same way in which state entities, to use a paradigmatic case of public entities, are deemed public. The authority of nongovernmental organizations over us is not political; it is typically grounded in consent or in some exceptional circumstances of vulnerability.
By contrast, our argument defends a nonhierarchical idea of the public. It implies that entities are public if and only if they speak and act in our name, rather than for us. Simply put, these entities are not our bosses, nor are they akin to fiduciaries who function as our money managers or guardians. Their decisions must reflect (in the right sense) the perspectives of those subject to them. On this view, to count as a genuinely public entity is to be identified not with a commitment to advance public purposes or to promote our interests, but rather with representing us all.
But can this understanding of the concept of the public give rise to obligations? Our answer rests on the fact that political representation can give rise to obligations when representation is accompanied by sufficiently substantial benefits. More specifically, representation weakens the claim of the subject that the duties imposed on her or him (for the sake of promoting the public good) unjustifiably violate her or his freedom and equal standing. If the person is adequately represented, compliance does not undermine her or his freedom as, in effect, the obligations imposed on her or him are not incompatible with her or his worldview or inner self. Nor does it undermine her or his equal standing as her or his worldview, or inner self, rather than the official in authority, determines these obligations.
Our argument provides a conditional reason to conform or comply with political authority. If one does not wish to live in a state of anarchy, then one has a compelling reason to join a political community. As the existence of this community and its well-being arguably depend on its having some measure of authority over its members, adopting the latter option involves submitting oneself to political authority. But this in itself is not enough as submission cannot be required merely to promote the public good. A nonhierarchical political authority predicated upon representation respects its subjects as both free and equal. First, it eliminates, or at least weakens, the complaint that the resort to force necessarily violates the freedom of the authority’s subjects. Second, it eliminates, or at least weakens, the inequality that is built into any hierarchical relationship between a practical authority and its subjects. Note that the proposed account is the only one we know to argue for truly egalitarian authority relations.Footnote 30 Further, it is the only account of authority that can make equality as nonhierarchy an integral part of its appeal.
These points imply that the authority can be justified in using force to compel its free and equal citizens to act in accordance with its orders and that its subjects have a duty to obey. Hence, we have a reason – that is, conditional on our having a prior reason to join a political community – to submit ourselves to freedom- and equality-respecting political authority, namely, a nonhierarchical public authority committed to representing us all.
III Three Challenges: Reason, Consent, and Procedural Fairness, and Why They Matter
This section examines three challenges to our account, each of which gives rise to a competing theory of authority. Under the first, our account of representation ignores the fact that authority needs to guide its subjects to act in accordance with reason. Under the second, our account is not sufficiently respectful of autonomy as it fails to account for the requirement of consent. Under the third, our account fails by not requiring fair or egalitarian decision-making procedures, that is, a democratic process. This section is designed to establish that none of these challenges succeeds. As a matter of fact, our account can accommodate the importance of each one of these considerations and yet establish that none can alone account for political authority.
A Representation and Reason
Political authority is often said to be grounded in its ability to guide people to act in accordance with reason. This is best articulated by the normal justification thesis proposed by Raz.Footnote 31
The appeal of the reason-based account rests on the fact that political discourse often justifies authority in terms of justice or reason. Politicians who wish to be elected argue that they, rather than their opponents, are likely to bring about justice or promote the public good, and they justify their authority on these grounds. Politicians are ultimately technocrats or moral technocrats; they are likely to bring about maximal convergence between action and reason. It might seem that representation-based authority cannot account for the centrality of reason and justice in political discourse. What happens if subjects’ worldviews are corrupt or unjust? Does it follow that a representation-based theory of legitimate political authority inevitably gives rise to injustice?
To address this objection, we distinguish between external and internal considerations. By external considerations, we mean justice-based (or, more generally, reason-based) limitations on the legitimacy of political authority. On this view, even ideal political representatives may not possess the legitimate authority to make binding decisions that flatly defy fundamental demands of justice. The more interesting part of our response concerns the justice considerations that should be deemed internal to legitimate political authority. We argue that the representation-based account can accommodate some demands of reason and justice. Such demands are integral to political representation. This is because reason (or justice) is regarded by most citizens as a constraint on, and, at times, even a goal of politics. To the extent that people care about justice (or reason, more generally), and that justice forms an essential part of who they are, representing them must take into account justice as a constraint. Justice is not merely an external constraint on representation, but rather also part of what representation itself dictates.Footnote 32
Note that implementing this internal consideration too stringently may result in endorsing a pure reason-based approach. Were people to care exclusively about justice or reason, representing them would require no more than doing what is just or what reason dictates. But we believe that genuine representation is not reducible to a reason-based account. To represent, the political authority must take seriously the actual values and convictions that people hold, which also include their concern for justice. Nevertheless, the people’s concern for justice does not imply that political authority can fail to account for less abstract concerns than the mere concern for justice (or reason) that the represented may have. Thus, it would be wrong for a representational authority to use reason alone without accounting for more concrete concerns that people have (as if it did so, it would fail to represent them).Footnote 33
Finally, we wish to draw attention to an important implication of our analysis that we believe might have significant implications for political philosophy. A reason-based theory focuses on cases in which resorting to reason dictates a particular decision. It implies that political decisions are grounded in reason. When reason does not dictate a particular answer, Raz’s normal justification thesis would be indifferent to what is being done. But too often in politics, decisions are grounded simply in the fact that they conform to our values or worldview. A government or a ruler who only guides us to act rightly or justly may not constitute an authority if their decisions fail to be ones that we can regard as our own. Political authority may lose its legitimacy if it acts unjustly, but it may not gain legitimacy in the first place without being representational.
B Representation and Consent
Traditionally, consent-based justifications of political authority have been the main competitor to reason-based justifications. On a standard account, the giving of consent is a form of allowing another person to occupy a position of authority over the consenting person.Footnote 34 By consent, we mean the actual, including tacit, consent of the governed, rather than either hypothetical or normative consent.Footnote 35 This is because both hypothetical and normative consent are essentially forms of reason-based justification. Consent-based arguments have been effectively criticized in the past, and we will not repeat these criticisms.Footnote 36 Our brief discussion of consent here is meant to show that our account addresses the concern that consent theories are designed to solve.
Consent-based theories aim to establish that authority is compatible with freedom. As, arguably, people consented to the authority, deference to the authority is their choice. Our account can also be regarded as one that reconciles authority with autonomy. We have argued that, unlike other forms of practical authority, political authority is nonhierarchical. Under this view, political authority gains legitimacy by representing us, that is, by endorsing and interpretatively implementing our worldview or reflecting our inner self. What counts as representing us is controversial, to be sure. But we argue that ultimately political authority can be justified only if it can be shown to be nonhierarchical.
One difference between consent-based theories and our account concerns the object of consent. The typical argument from consent suggests that the consent of the governed should be granted either to the basic political structure of society or to the person in authority. Thus, the object of consent is not the political decisions made by the governing institutions of society, but rather the institutions (or the procedures of decision-making).Footnote 37 By contrast, the nonhierarchical theory of political authority emphasizes the connection between the content of the decisions made by political authorities and the perspective of the authorities’ subjects. To represent us is to make decisions from our perspective, and doing so turns on establishing a sufficiently tight connection between the decisions’ substance and what we want, judge to be just, or who we are. As stated, our account does not ignore consent, but it does not give it an exclusive role in justifying legitimate authority. Consent can often indicate that the authority is genuinely representative. Further, genuine consent can often be achieved by faithful representation. There are, therefore, important connections between consent and representation but, nonetheless, they remain distinct concepts.
C Representation and Fair Decision-Making Procedures
Can our account justify the concern for egalitarian or democratic procedures? Many theories of political authority identify legitimate authority with the fairness of the procedures leading up to the decisions of authorities. A particularly powerful variation on this theme is equating fair procedure with a democratic or majoritarian process. On this variation, the democratic process can establish, reflect, or sustain equality among its participant-citizens. By linking the democratic process to equality, the argument is that the legitimacy of the decisions made by democratic authority does not necessarily depend on the substance of the decisions made by democratic institutions (i.e., on their conformity with reason), but rather on the promise that citizens participate as equals (or, at least, have the power to participate as equals) in the making of political decisions.
There are different interpretations of what “participating as equals” might mean. Thomas Christiano emphasizes that the democratic procedure, properly constructed, reflects each citizen’s equal say in a collective decision-making process, by which he means equal say vis-à-vis each and every other participating citizen. Disobeying democratically made decisions is a certain form of disrespect, namely, publicly treating our fellow citizens as inferiors, as citizens who are not entitled to their equal say.Footnote 38 By contrast, Daniel Viehoff argues that democratic processes do the “central justificatory work”Footnote 39 in grounding democracy’s authority insofar as they implement an ideal of relational equality and, in particular, nonsubjection. On this account, citizens participate in democratic processes as equals when doing so does not render them vulnerable to the superior (economic, physical, et alia) powers of their peers. Niko Kolodny has developed a different variation on the relational equality theme, arguing that democracy’s authority necessarily lies in securing, for each and every citizen, an equal opportunity to influence political decisions.Footnote 40 Recently, Seana Shiffrin developed a communicative account of democracy under which democratic law is a means of communicating to each other our moral commitments to a broad range of values.Footnote 41
There are important differences between these four accounts – and other theories that explain political authority in terms of democracy – and our account.Footnote 42 First, although democratic authority is a subset of political authority, it is not clear to us that it has to be the only possible form of legitimate political authority. This suspicion suggests that the ultimate justification of legitimacy is not the egalitarian character of decision-making processes (which is realized, let’s assume, only in democratic systems). Rather, it may be the convergence between the perspective of the represented and the decisions made by the authority and, consequently, the possibility of attributing authoritative decisions to the real authors, namely those who are subject to them. Further, our account also suggests that, at times, democracy does not give rise to political authority as understood by us. Voting (as the example of Alan and Daniel indicates) is not necessarily the best indication of what one’s perspective dictates as voters may have preferences or make judgments that do not fully fit into their worldviews.
Second, whereas the democratic accounts under consideration require egalitarian procedures of decision-making, there is no requirement that the decision-maker itself stands in a nonhierarchical authority relationship to its subject-citizens. That is, giving citizens an equal say in casting their votes and running for office, establishing subjection-free processes of democratic participation, or making political institutions compatible with each citizen’s equal opportunity to influence political decisions may well facilitate citizen-to-citizen equality. But at the same time, these factors fall short in that they do not require the representative to adopt the perspectives of the citizens and, so, to decide in their name, rather than for them.Footnote 43 Democratic decision-making procedures that reflect a robust commitment to the free election of representatives are fully compatible with having their representatives make binding decisions for the citizens, that is, on the basis of what the representatives judge to be the right decisions. To put it provocatively, an egalitarian proceduralist democracy can, in reality, be a form of democratic dictatorship; there is nothing that guarantees that its decisions converge with, not to mention reflect, the perspectives of the represented and, consequently, nothing that guarantees that the decisions are ones that can properly be attributed to them.
Yet our theory of political authority requires some decision-making procedures that resemble contemporary practices of democratic states. To act in our name, persons in authority must have sufficient access to what persons want or judge to be just, or to who they are. A publicly recognizable procedure provides an institutional context in which all members of society have the opportunity to articulate their respective views and freely express who they are (say, by being entitled to pursue their heterodox conception of the good life). Thus, in our view, producing such input matters not necessarily because it gives effect to democratic and egalitarian values, but rather because it provides the requisite epistemic resources on which authorities must draw in order for them to be able to speak and act in our names.
By implication, equal opportunity of influence – the requirement characteristic of some proceduralist theories – may often be congenial to representation as understood by us. It may even be the best institutional means of realizing such representation, but ultimately representation sets a different standard for legitimacy than that. Democracy can, therefore, be understood as an institutional means for the facilitation of representation, rather than as the ultimate justification of political authority.Footnote 44
To sum up, while our account differs from reason-based, consent-based, and proceduralist theories of authority, it also accounts for the appeal of these theories. It shows that reason, consent, and fair procedure are not the foundational features characterizing political authority but rather merely derivative of another more fundamental concern: that the laws that govern us be our laws, that is, that we govern ourselves.
Conclusion
This chapter develops a public conception of political authority. Our account defends three observations. First, political authority is a sui generis case of practical authority in that, unlike most other cases of practical authority, it takes a nonhierarchical form. Second, political authority claims legitimacy by making decisions that properly represent the perspectives of the represented. And third, the decisions made by a political authority can be attributed to those represented.
These observations explain why citizens regard decisions made by public authorities as if they were their own decisions, and also why we may regard citizens as partially accountable for the decisions of the political authority. To become a political authority, an institution or an official needs to be faithful to the perspective of those who are subject to it. By so doing, those who are subject to its directives count as governing themselves. Consequently, the decisions being made by the authority are in reality authored by the people. This is ultimately what makes public officials public. They count as public when they speak in the name of us all. It makes possible the reconciliation of authority with both the freedom and the equality of those who are subject to its rule. It allows people to view the binding decisions of the state as their own making. The next chapter seeks to establish that the normativity of law should also be explained in a similar way.
