The Legitimacy of International Institutions


Thomas Christiano

The Challenge for an Account of Legitimacy

The analysis of legitimacy of international institutions presents a distinct challenge because we do not yet have an entirely clear picture of the nature of international society and the position of international institutions within that society. The institutions are not very much like state institutions or voluntary associations. At the same time international society shares features of both of these kinds of association. This is an important observation because these are the classical kinds of association to which ideas of legitimacy have traditionally been applied. And so the notion of legitimacy that we apply in the case of international law and institutions will have some features of both of these.

It is worth our while to have some further observations about the international system before we proceed. One, we need to have on the table the list of entities that are relevant to the characterization of international society. First, the most important entities are the states that make up the international system. Second are the international institutions that states have created but that in some cases exhibit some independence from the states that have created them. We will discuss these in some detail. Third are nongovernmental organizations that play a role in international society. Fourth are multinational corporations. Fifth are individual persons. All five of these elements are directed by international law and institutions, although states are still the principal subjects. Individuals are only marginally directly subjected to international law; mostly they are indirectly subjected to it.

Two, the system is highly fragmentary. Distinct institutions govern functionally distinct as well as regionally distinct activities. There are regional economic organizations such as the European Union. The World Trade Organization has a kind of legislative function over its members’ trade relations. Global environmental treaties such as the Montreal Protocol are designed to regulate some specific global environmental matters. The United Nations Security Council has assumed some quasi-legislative functions since the end of the Cold War, but these are still fairly limited (Szasz 2002). The Security Council also exercises a kind of enforcement authority by permitting states to act against other states that threaten international peace and security. Other institutions exercise a kind of administrative authority. They make rules within the framework of a treaty that establishes the institution and that mostly help implement the more general treaty law. The Montreal Protocol has provisions for rule making and rule alteration that are along these lines (Kingsbury 2007). Some international institutions exercise judicial authority as well. The dispute-settlement mechanism of the World Trade Organization has the authority to settle disputes between states when a member state complains that another member state has violated its agreements. The International Court of Justice adjudicates as well but only for those states that accept jurisdiction in the case at hand. And there are international lending agencies, which are lenders of last resort to states in economic crisis.

In addition, there are a variety of institutions that create what is called “soft law.” This is law that is not binding on the state parties, though there can be pressure on states to conform to this law. Declarations and other acts of the United Nations General Assembly are generally construed as nonbinding. The rule making of many different institutions also has soft-law status.

What can make these institutions legitimate? That is, how might these institutions have the powers to create binding obligations for states and/or individuals? The traditional answer to this question is that international law is binding on states primarily because and only because states consent to it. The Statute of the International Court of Justice lists in addition that the sources of international law include custom, the general principles of international law and the opinions of eminent jurists. But the two main materials are treaties and customary international law.

In this essay, I will attempt to lay out the idea of state consent as the ground of the legitimacy of international institutions, its bases and its limits, as well as some proposed ideas about how to extend and modify it. In the end I will argue that state consent, suitably modified, must play a central role within a larger system of international decisionmaking. I will start with some observations about legitimacy and some of the reasons for which consent has been taken seriously as a basis of legitimacy. I will then articulate the doctrine of state consent. Then I will go through a series of objections to the doctrine that can be met by the doctrine, in some cases only with important modifications. Finally, I will discuss a series of objections that can only be met by supplementing state consent with two other important requirements on the system of international decision-making: a constitutional system with enabling and limiting components and a set of internal standards for particular international institutions.


Let us start with some of the elements of the idea of legitimate authority. I will understand legitimate authority as involving at least a moral power to impose a set of rules by which the others must regulate their conduct. These rules can be straightforward duties, they can be power-conferring rules (or rules that specify how persons may bind themselves and others) and they can be rules conferring permissions. Usually legitimate authority involves the power to impose all three of these rules but the imposition of duties on others is the most central case and the hardest case of legitimate authority because it involves the imposition of burdens on persons or groups. Possession of a moral power is a minimal sense in which a person or group can be said to have a right to rule. It is not the only thing that can be meant by this; sometimes the right to rule involves in addition a claim on the part of the ruler to the obedience of the ruled so that the obedience is owed to the ruler. Though this latter relation holds in some important cases, it is not necessary for legitimate authority.

The exercise of the moral power grounds the duty in the subject. In this way the subject has a content-independent reason for complying with the directive of the ruler. He has the reason to comply because the ruler has exercised the moral power. Hence he has a reason to obey.

The moral function of legitimacy is to have a public moral standard for making decisions when there is considerable disagreement among the subjects. Subjects have moral reasons to go along with the legitimate authority even when they disagree with it. This enables societies to pursue basic moral purposes in a coordinated fashion.

How does a person or group of persons acquire such a power? There are, in contemporary thought, three main types of answer to this question. They needn’t be mutually exclusive. The first answer is that a person or group of persons acquires a moral power to impose duties on others only if the others have consented to the group possessing the moral power. The consenters thereby acquire content-independent reasons to go along with the directives. The second answer is that the moral power is somehow possessed collectively by a democratic assembly, in which case the subjects must comply with the democratically made decisions. The third kind of reason is an instrumental one. If a purported authority is reasonably just and its directives enable subjects to act better in accordance with reason, then the purported authority has a moral power to impose rules on its subjects.

The consent and democratic conceptions of legitimacy are distinctive in two ways. One, they establish, in addition to a moral power in the decision-maker, a claim right in persons to whom the compliance is owed. Consent establishes a claim right either in the power holder or in other persons subjected to the power holder (as in the case of classical consent theory) to the compliance of the consenter. The democratic conception establishes a claim right in the democratic assembly to the compliance of every member in the society. Two, the consent and democratic conceptions are distinctively public standards of legitimacy in the sense that each of these standards is one that subjects can see to be in effect for the most part. In this respect the instrumental basis of authority is a weaker basis. It cannot serve the public function of legitimacy as well as the consent or democratic principles, since the authority is based in the quality of the outcomes, which are precisely the subject of disagreement among the members. For this reason, it is worthwhile trying to figure out whether institutions that purport to have authority can have it on the basis of consent or democratic principles. These will be the focus of this essay, though I do not wish to suggest that the instrumental conception of authority is of no significance. One last remark: I will not focus here on ideas about global democracy because I do not think that we are even close to realizing the kinds of institutions that would be necessary for global democracy (Christiano 2010). Instead, we will see how democratic ideas can inform and modify the traditional idea of legitimacy in international institutions: through state consent.


Consent gives a fairly straightforward answer to the question, how does one person acquire the moral power to impose duties on another? Consent does this in part by having the person impose the duty on himself, with the concurrence of the power holder, as one does with a promise. The latter is a fairly uncontroversial form of imposition of duty. Furthermore, it involves a duty of the right type. When I promise to do x, I have a duty to do x merely on the ground that I have promised. The promise is a content-independent reason for doing x. And if I promise to obey a ruler, that gives me a contentindependent reason to comply with the ruler’s directives (Simmons 2001).

Usually consent must be given voluntarily and the person or group to whom consent has been given must not have defrauded the consenter. Furthermore, consent must be accompanied by some additional conditions under which the consent creates the moral power. For example, the consent must be given by a minimally competent person and it must not be given to obviously and seriously immoral commands. We shall consider some other conditions in what follows.

There are three traditional grounds for consent as a requirement on legitimate authority. The first is that normally a person is morally at liberty to do or not to do the kinds of things that eventually come under the scope of an authority. She may do it or not and act on the basis of her own moral judgment about the quality of the actions at issue. Consent to the authority is what suspends the moral liberty in the sense that the consenter must now comply with the authority’s directives, merely because the authority has said so. Consent suspends this moral liberty in a way that is compatible with the freedom of the consenter. Hence the requirement of consent protects the freedom of persons. The second ground is that the requirement of consent protects the moral equality among persons. Unlike the relation between parent and child, the relation between two adults is normally a relation among equals, which only consent can change. But consent does this by preserving the basic equality. Finally, the requirement of consent and the possibility of refusal as well as of terminating consent creates a relationship of accountability between the purported authority and the subject, since the subject can change the relationship at will. In sum, the requirement of and power to consent enable me to advance my interests and concerns as a free and equal person by binding myself to others.

The most straightforward application of the consent basis for legitimate authority is voluntary association. When I join a club, I thereby acquire a reason to go along with its rules and I accept the normative power of the duly constituted authority in that group to change the rules. When my consent to this authority is required and I am able to consent, my freedom and equality are preserved even when I bind myself to the voluntary association.

Some have argued that consent is the basis of the legitimate authority of the state (Locke 1986/1690), but this is a weakly grounded application of the consent idea. The reason why the requirement of consent is not very strongly grounded in this context is that it doesn’t protect the moral liberty of action or the equality of persons, nor does it secure the accountability of power to the consenter. Possession of the moral liberty usually involves the permission to act in a certain way or not and it involves the permission to act in accordance with one’s own judgment about how to act. These permissions clearly hold in the normal case in one’s relations to voluntary associations. Furthermore, the fact that I must consent to membership in the club and can exit it makes the club accountable to me to some extent.

But the permissions associated with moral liberty do not hold so clearly in the case of a reasonably just state. Here, in order to coordinate with others on a reasonably just course of action, one must follow the directives of the state officials—so moral liberty seems not to be very extensive with respect to the directives of the state. Thus the requirement of consent that is grounded in the protection of these permissions seems not to be present to any great degree (Raz 1986). Furthermore, consenting and refusal of consent seem to make for little accountability of the state to the person, except in unusual cases. My consent or lack of consent communicate very little to the state about my attitudes towards the state and impose little or no penalty on the state. Finally, in a democratic state, my equality with others is preserved in the democratic process of collective decision-making. I do not say that consent can play no role here; I want to say that in the case of a reasonably just state, consent is at best required only at the margins.

State Consent and Legitimacy

As I noted above, international institutions are neither state-like institutions nor voluntary associations, so the application of the consent idea to them poses distinct challenges. State consent is a large part of the traditional account of the legal obligations of international law. The Statute of the International Court of Justice lists treaties, custom and the general principles of law as the principal sources of international law (United Nations 1945, Article 38). The obligation of a treaty is grounded in the consent of the state parties to the treaty. The interpretations of treaties are usually based on the negotiations that preceded the treaty as well as the subsequent behaviors and statements of the state parties to the treaty. Furthermore, the Vienna Convention on the Law of Treaties asserts conditions for the validity of state consent that are very much like the usual conditions on valid individual consent to contracts (United Nations 1969, Articles 48–52). Most treaties, in addition, specify that it is permissible to exit the treaty in question (Helfer 2005). And states may tailor the provisions of a treaty to their particular interests by attaching reservations and understandings to their ratification of the treaty. Finally, treaties are agreed to one by one in a relatively fragmentary system. These features suggest a fair degree of accountability of international law to states through the process of state consent. And states have been traditionally conceived as morally at liberty to choose whether to join a treaty or not.

To be sure, I have been speaking in the previous paragraph of legal obligation, but given the power of promising to create obligations, it is reasonable to think that the consent of states under the right conditions and within reasonable moral bounds morally obligates the states.

There are four serious objections to the idea that state consent can ground obligations to international law.

Democratic State Consent