Human rights occupy a privileged position within contemporary politics. They are widely taken to constitute perhaps the most fundamental standards for evaluating the conduct of states with respect to persons residing within their borders. They are enshrined in numerous international documents, national constitutions, and treaties; and those that have been incorporated into international law are monitored and enforced by numerous international and regional institutional bodies. Human rights have been invoked to justify popular revolt, secession, large-scale political reform, as well as forms of international action ranging from the imposition of conditions on foreign assistance and loans to economic sanctions (as in South Africa and Burma) and military intervention (as in Kosovo and East Timor). Michael Ignatieff has gone so far as to claim that human rights have become “the major article of faith of a secular culture that fears it believes in nothing else,” and one might add that they are articles of faith of many non-secular cultures, too.Footnote ¹

Yet despite the widespread influence of human rights discourse, it remains unclear precisely what human rights are. The problem is not simply that there is a lack of clarity about the content and grounds of human rights claims—what particular human rights there are, and why we have the particular human rights we do. Rather, it is that the very idea or concept of a human right remains obscure. It is far from obvious, in other words, what we are even saying when we say that there is a “human right” to education and health, for example, or a “human right” not to be tortured or made to engage in forced labor.

Part of what we are saying, of course, is that there is a right—specifically, a claim-right—to these things.Footnote ² It is part of the concept of a claim-right that such rights impose duties on others; and this distinguishes human rights from aspirations, goals, and many other values.Footnote ³ But what kinds of claims are human rights claims, and what sorts of duties do they impose? Clearly, they are not simply legal claims and duties—say, claims and duties that are enshrined within international law—for we may intelligibly ask of anything that is called a human right within international law whether it is a genuine human right.Footnote ⁴ Rather, they are moral claims and duties. But this cannot be the whole story, either. For not all moral rights are human rights. Most of us think that there is a moral right that promisers keep promises they have made to us. But no one seriously thinks that promisees have a “human right” to have promises kept, or that breaking a promise—even a very important promise on which much depends—constitutes a human rights violation.

What is needed, then, is an account of human rights that can illuminate what is special about them. Two important recent books by distinguished human rights theorists attempt to answer this challenge. James Griffin argues that human rights are moral rights whose protection is required to preserve our “capacity to choose and pursue our conception of a worthwhile life” (p. 45). Charles Beitz argues that human rights are moral rights that play a certain kind of role within an emerging global practice. While both Griffin's “personhood account” and Beitz's “practice-based account” of human rights contain important insights, we shall argue that neither is entirely satisfactory. We conclude with a suggestion for what a more adequate account might look like.

The Personhood Account

Griffin's aim is to develop an account of the concept of a human right that remedies the “damaging indeterminateness of sense” that has plagued it since its inception (p. 93).Footnote ⁵ Griffin starts by noting two elements of human rights that are traceable back to the Enlightenment (p. 2). The first is the idea of rights that we have “simply in virtue of being human” (p. 13). The second is the idea of rights that are needed to protect the “dignity of the human person.” Griffin takes these elements as starting points for the idea of human rights, but deems them to be insufficient. The first leaves open the question of what it is about our humanity that is supposed to be important (p. 15). The second is problematic because the concept of the dignity of the human person is arguably even less clear than that of a human right (p. 5).

Griffin tries to remedy these deficiencies by offering an account of human rights that is based on a particular interpretation of the idea of the dignity of the human person, one that captures what is distinctive about our humanity and why it deserves protection. The key is that human rights are moral rights that are necessary to protect what Griffin calls our normative agency. By this he means the distinctive human capacity to form and pursue a conception of a worthwhile life. He writes:

Three things are required in order for our existence as normative agents to be sustained. First, we must have autonomy (pp. 33, 149–58). To be autonomous is “to choose one's own path through life—that is, not be dominated or controlled by someone or something else” (p. 33). One must be able to make up one's own mind, and thereby “exercise one's capacity to distinguish true values from false, good reasons from bad” (p. 150). Second, one must have liberty (pp. 159–75). That is, “others must also not forcibly stop one from pursuing what one sees as a worthwhile life” (p. 37). Our normative agency consists “not only in deciding for oneself what is worth doing, but also in doing it.” We lack liberty insofar as others interfere with our “carrying out our decisions” (pp. 150–51). Finally, we must have a certain “minimum provision” of welfare (pp. 33, 176–87). This involves having sufficient education, information, resources, and capabilities in order for one's choices to be “real” (p. 33).

According to Griffin, human rights are moral rights that serve to protect these three aspects of our normative agency. Within the category of human rights, there is an important distinction between those that are truly universal and those that only arise given certain social conditions. Truly universal human rights are those that are required in order to protect our normative agency under (more or less) any conceivable social conditions. For example, the human right not to be tortured seems to fall into this category since it is plausibly required in order to protect our autonomy and liberty whether we inhabit a modern state or a state of nature. Others—such as the right to freedom of the press and the right to democratic government—may be necessary to protect normative agency only under certain circumstances (pp. 50, 251–54). It is not impossible to imagine circumstances under which our normative agency could be protected without such rights. Griffin refers to the considerations that, when conjoined with the values of personhood, yield a list of genuine human rights in any particular context as “practicalities” (p. 37).

Griffin's personhood account offers a straightforward answer to the question of what is special or distinctive about human rights. Human rights are special inasmuch as they protect and preserve our normative agency. They are “protections not of the fully flourishing life but only of the more austere life of a normative agent” (pp. 37, 53). Many moral rights that protect important interests are not required in order to protect our normative agency. It hardly threatens one's status as a normative agent to have someone break a promise she has made to us, or to have a romantic partner treat one in a cold or indifferent manner. Normative agency seems to be the crucial thing that distinguishes us from other animals. We are, as Griffin puts it, “self deciders,” and this “is part of the dignity of human standing” (pp. 150, 32). As such, normative agency is something that is of particular value to us as persons, irrespective of our particular conceptions of worthwhile living. This helps to explain why human rights have been thought to be universal and why they have been thought to generate such stringent and far-reaching duties.

Griffin's account also yields plausible substantive results across a wide range of cases. Some of the richest parts of the book are those in which Griffin considers whether particular human rights are genuine by exploring whether they can be derived from his trio of highest-level human rights conjoined with practicalities (chapters 11–14). For example, he argues that the right to health is a plausible human right, but only so long as its scope is restricted to protect those aspects of our health that are crucial to our lives as normative agents (p. 101). We do not have a human right, however, to the “highest attainable level” of human health, as asserted in Article 12 of the International Covenant on Economic, Social and Cultural Rights and in the constitution of the World Health Organization.

Griffin also provides an account of welfare rights that avoids the stark minimalism of libertarianism without succumbing to the temptation to call all valuable social goals human rights. On the personhood account, we have a human right to the provision of goods necessary for our lives as normative agents, but not a human right to anything like the level of material well-being that is characteristic of most developed societies (pp. 181–85). This is not to say that such an elevated standard of living is not required as a matter of justice, or that we do not have some other moral claim to it, but only that we have no human right to it. As Griffin notes, there are many values other than human rights, and it is a mistake to try to turn everything that is of value into a human right (pp. 41–43).

Despite its appeal, the personhood account is vulnerable to serious objections. We shall mention two. The first is that it is unclear that the personhood account can make sense of the full range of human rights. Consider, for example, rights against racial discrimination. It is not obvious that a system of racial discrimination or apartheid does anything to undermine normative agency.Footnote ⁶ To be sure, such discrimination will often significantly restrict the options of people belonging to certain groups and undermine their ability to achieve their goals. But Griffin is quite explicit that this is not enough in order for there to be a human rights violation. He writes, “Human rights, I propose, are rights to what allows one merely to act as a normative agent, not as a normative agent, as Raz would have it, with ‘a good chance . . . of achieving one's goals.’”Footnote ⁷

Griffin might defend the personhood account against this criticism in either of two ways. First, he could claim that social arrangements allowing racial discrimination would be unfair or unjust, but deny that they constitute human rights violations in particular. This first strategy seems plainly unsatisfactory. Targeting social arrangements that have involved discrimination, and the norms they support, has been a fundamental aim of the human rights movement.Footnote ⁸ It should count heavily against any account of human rights if it cannot make room for antidiscrimination rights.

For this reason Griffin seems to favor adopting a second strategy—to try to show that his view can accommodate antidiscrimination rights after all. Responding to a version of this criticism formulated by Allen Buchanan, he writes:

Griffin is surely right that discrimination involves a highly objectionable form of unequal treatment. But the question that is at issue is whether such unequal treatment should be thought to undermine our normative agency. Griffin's answer is that various kinds of discrimination are likely to undermine our normative agency “because of their potentially destructive effect on an agent's self-image” (p. 42). As an empirical claim, however, this seems questionable. The effects upon the self-image of the members of particular groups of discriminatory social arrangements will seldom be so great as to call into question their status as normative agents. More important, this seems to be a rather flimsy basis upon which to claim that discrimination involves a violation of one's human rights. Imagine, for example, that the members of a particular group who are discriminated against happen to have an especially robust self-image—perhaps precisely because they have already suffered centuries of discrimination—such that they are able to maintain their positive self-image in the face of such arrangements. Surely, the admirable fortitude of such individuals should do nothing to undermine our sense that they are the victims of human rights abuses.

The other problem with the personhood account is that it fails to account for what we might call the political aspect of human rights. A number of theorists, including Beitz, have noted that talk of “human rights” only seems appropriate insofar as something like an organized political authority is somehow involved. Consider, for example, the right not to be tortured. To speak of a “human right” not to be tortured only seems to be apt if the right is correlated with a duty on the part of some organized political authority to take steps to protect members of society from torture. Thus, for a person's human right not to be tortured to be violated, it is not enough that the person is tortured—say, by a sadistic kidnapper acting on his own behalf. Rather, something like a state, or those acting on behalf of the state, must have engaged in or facilitated torture, or else failed to take appropriate steps to ensure that the torture did not take place, or that the torturer was brought to justice.Footnote ¹⁰

The problem for the personhood view should be clear. It seems that torture is equally destructive of our normative agency no matter who carries it out. Indeed, suppose that we are in a state of nature in which there are no agents claiming political authority whatsoever. This seems to make no difference to the destructive effect of torture on our normative agency. The personhood account therefore seems to lack the resources for making sense of the distinctively political nature of human rights—that is, the way in which human rights are constitutively tied to, and dependent upon, political entities of a certain kind.Footnote ¹¹ Let us now consider a very different account of human rights in which their political aspect is emphasized.

The Practice-Based Account

Whereas Griffin tries to understand human rights in terms of the role they play in protecting our normative agency, Beitz takes as his starting point the emerging global practice of human rights. According to Beitz, the practice of human rights plays a fundamental role in determining the concept of a human right. We must, Beitz writes, “tr[y] to grasp the concept of a human right by understanding the role this concept plays within the practice” (p. 8).

Beitz describes the “global practice” of human rights as follows:

Beitz claims that the practice began to take on something like its modern form in the aftermath of the Second World War, when the world's powers came together to enumerate a set of binding rules, framed in the language of human rights, concerning the responsibilities of states and the global community. A crucial step in this process was the adoption in 1948 of the nonbinding Universal Declaration of Human Rights (UDHR). This was followed by the ratification of a number of binding international treaties that served to interpret and supplement the rights adumbrated in the UDHR;Footnote ¹² and the creation of a number of international and regional institutional bodies whose role was to hold states accountable for failures to abide by these treaties.Footnote ¹³ The practice was further solidified and refined by individual states taking a wide variety of practical steps to help ensure compliance with human rights norms both within their own borders (creating domestic legislation, modifying their constitutions, adopting particular public policies and institutional arrangements) and outside their borders (employing financial or trade incentives, diplomacy, economic sanctions, and military force). The practice has also been shaped by numerous nongovernmental organizations, such as Human Rights Watch, Amnesty International, and Oxfam, whose activities have included drawing attention to human rights abuses, mobilizing public support, and lobbying national governments to engage in reforms that would better realize human rights.

How is the practice supposed to help determine the concept of a human right? The key, according to Beitz, is that “questions about the nature and content of human rights [are understood] to refer to objects of the sort called ‘human rights’ in international practice” (p. 102). To be an object of the sort called a “human right” within the practice is to play a certain “functional role” (pp. 103, 136). According to Beitz, the practice of human rights is distinctive, not simply on account of the content of human rights norms but because of the shared understanding of the aims and purposes that human rights standards are supposed to serve (p. 103). Implicit within this shared understanding is a certain conception of the functional role that human rights are supposed to play within global political life (p. 8). This functional role can be understood, Beitz claims, by attending “to the practical inferences that would be drawn by competent participants in the practice from what they regard as valid claims of human rights” (p. 102). The functional role is then taken to be fundamental in fixing the concept of a human right. As he puts it, “we take the functional role of human rights in international discourse and practice as basic: it constrains our conception of human rights from the outset” (p. 103).

What is the character of this functional role? According to Beitz, it has three aspects. First, human rights “protect urgent individual interests against certain predictable dangers (‘standard threats’) to which they are vulnerable under typical circumstances of life in a modern world order composed of states.” Second, they “apply in the first instances to the political institutions of states.” Third, they “are matters of international concern. A government's failure to carry out its first-level responsibilities with respect to human rights may be a reason for action for appropriately placed and capable ‘second-level’ agents outside the state” (p. 109).

According to Beitz, then, human rights are just those rights that are well suited to fulfilling these three aspects of the functional role of the things called “human rights” within the global practice of human rights. There is no presumption that all (or only) the things that happen to be called “human rights” within the practice satisfy this criterion. Something might be called a “human right” within the practice and yet be poorly suited for these roles—in which “case” assertions of such rights would be misguided. And there may be human rights that are not yet recognized within existing practice. The practice is “essential to the nature of human rights” (pp. 103–104), but does not determine infallibly what human rights there are.

Like Griffin's personhood account, Beitz's practice-based account offers a clear and in many ways compelling answer to the question of what makes human rights special. They are special, first, because they necessarily and primarily involve states as the duty bearers (pp. 113–15). This aspect of human rights serves to distinguish them from certain other moral rights, such as the moral right not to be tortured, which may be violated irrespective of whether or not a state is involved (or whether or not states even exist). Second, human rights are special inasmuch as they are matters of international concern.Footnote ¹⁴ Beitz does not claim that human rights entail that other states have a duty to interfere when the rights are not respected. But he does hold that they have a reason to do so. This reason may be outweighed by other considerations, but under certain circumstances it may be decisive (pp. 115–17). Moreover, other states and members of the international community are permitted to “hold states accountable” for human rights violations and to “assist an individual state to satisfy human rights standards in cases in which the state itself lacks the capacity to do so” (p. 109). This feature of human rights makes them different in an important way from other moral rights, including rights that generate duties that apply principally to one's own state. Suppose that one has a moral right not to be taxed beyond a certain threshold. One could argue that this cannot plausibly be thought of as a human right, since raising taxes beyond the threshold is not a matter of international concern. Other states would not have a reason to interfere with such a fiscally overdemanding state.

The practice-based account also appears to avoid the problems with Griffin's personhood account. It is better able to account for certain problematic rights, such as rights against discrimination. Surely being discriminated against in various ways could be sufficiently serious so as to entail a duty on the part of the state to avoid discrimination and a reason for other states to take steps to ensure that it did so. It can also make sense of the political aspect of human rights, since it is states that are taken to be the primary bearers of the duties entailed by claims about human rights.

The practice-based account is, however, problematic in other respects. One problem is that tying the concept of a human right so closely to the emerging practice has a number of highly implausible substantive implications. Suppose, for example, that a nuclear war or some other calamitous event were to result in the destruction of the system of sovereign nation-states. Or suppose that a single superpower state were to overthrow the other states and forcibly annex their territory. Since the functional role of objects of the kind called “human rights” within our current practice makes essential reference to a plurality of nation-states, it would follow that there are no human rights in such a world.Footnote ¹⁵ For in such a world none of the things that we would think of as human rights are able to fulfill the relevant functional role. Now suppose, however, that the absence of (a plurality of) states encourages acts of genocide and wrongful killings on the part of those trying to ascend to a position of power. Surely, such acts would constitute human rights violations. Indeed, one of the many reasons we have to prevent such a situation is precisely because it would likely be disastrous with respect to human rights fulfillment.

Or consider the possibility of radical change to the practice of human rights itself. Take a real-world example that Beitz discusses at some length—the propensity of the practice of human rights to curtail state aggression and facilitate peace among the world's states. Beitz notes that there is some evidence to think that this was at least a subsidiary aim of the drafters of the UDHR. He also notes that to say that this is true of the existing practice is implausible, since “the empirical premise required for the argument that encouraging respect for human rights promotes international peace is notably insecure” (p. 132). But we can easily imagine that the future could take a different turn, such that the UDHR and subsequent treaties, while couched in the language of human rights, come to be seen and interpreted as molded in the service of curtailing state aggression and facilitating peace among the world's states. Presumably, the list of individual rights that would serve the function of curtailing state aggression and facilitating peace would look very different from what we would take to be a plausible list of human rights today. For example, it is not unlikely that rights to political participation or against discrimination on the basis of race and gender might be left off such a list. In affording practice the privileged role in determining the concept of a human right, it seems that the practice-based account makes us, in some measure, hostage to historical fortune.Footnote ¹⁶

There is also a second, more general problem with the practice-based account—namely, that it obscures the crucial distinction between human rights and the form in which they should be institutionalized. Beitz is surely correct that it is vitally important to pay attention to various kinds of contingent facts about the global order—the existence of territorially defined states, the absence of world government, and so on—when considering how to institutionalize human rights. Moreover, it seems important to pay attention to facts about the way human rights are currently institutionalized. After all, the fact that we have a global practice of human rights is a historically unprecedented and remarkable achievement; and it would be irresponsible to put it unnecessarily at risk.

But none of this entails that we should jettison the idea that human rights are prior to and independent of the way in which they ought to be institutionalized in current global practice. One way to bring that out is to think about the origin of the practice of human rights. What were the prime instigators supposed to be doing? The natural thing to say, of course, is that they were trying to institutionalize preexisting human rights (see also Griffin, p. 204). Of course, they doubtless had conflicting views about the content, grounds, and other aspects of human rights. But surely there was something that they were trying to institutionalize. They were trying to institutionalize human rights, and they were trying to do so in an appropriate way. But if the practice-based account is right, then there was nothing to institutionalize, appropriately or inappropriately. For there was no practice, and therefore no objects of the kind called “human rights” within the practice.Footnote ¹⁷

There are, to be sure, familiar social practices that are not dependent on norms that are prior to and independent of their proper institutionalization—indeed, for which the idea of the proper institutionalization of such norms makes little sense. For example, consider certain conventional social practices, such as the norms associated with dress and communal dining. Roughly, for a norm of this kind to exist it is sufficient for enough individuals to behave in a certain way, to regard behaving in that way in a certain positive light, and to take the fact that others generally behave in that way and regard behaving in that way in a positive light to be a reason to act in that way.Footnote ¹⁸ This seems correct, for example, of the conventional norm among Oxford dons that one must pass the port to the left.

But the practice of human rights does not seem to be like this at all. It does not seem accidental or arbitrary that the UDHR happens to contain the particular rights it does, that the main treaties contain the particular rights they do, that states have written into their constitutions the particular rights they have. The point is that the practice seems to be crucially dependent upon a notion of human rights that is prior to and independent of their proper institutionalization. In tying the concept of a human right so closely to the function that the things called “human rights” within the practice are supposed to be playing within global politics, Beitz's practice-based account obscures this point and gives a distorted account, both of human rights themselves and of the practice of human rights.

Toward a Structural Pluralist Account

We have examined two influential recent accounts of human rights and argued that neither offers a wholly satisfactory answer to the question of what is special about human rights. This raises the question, What would a more adequate account of human rights look like?

First, like Griffin's account (and unlike Beitz's), it will be practice-independent and substantive. It will be practice-independent inasmuch as it conceives of human rights as moral rights that are prior to and independent of their institutionalization within global practice. It will be substantive inasmuch as it conceives of human rights as moral rights that serve the role of protecting certain core interests that we all share to an adequate degree.

Second, unlike Griffin's account, it will be pluralist. That is to say that it will conceive of human rights as moral rights that serve to protect a broader range of human interests than merely an interest in being able to maintain and manifest our normative agency.Footnote ¹⁹ Exactly which interests belong on the list is a difficult matter, but it should be rich enough to account for paradigmatic human rights, such as those against racial and sexual discrimination. To this end, Martha Nussbaum suggests that we should include an interest in what she calls “affiliation,” which involves being “treated as a dignified being whose worth is equal to that of others.”Footnote ²⁰ James Nickel, another pluralist, posits an interest in avoiding “severely unfair treatment” as a ground of human rights.Footnote ²¹ And John Tasioulas talks of an interest in what he calls “living harmoniously with others,” which might also play this role.Footnote ²² At the same time, of course, the list should be restricted to our most urgent interests, so as not to invite profligacy in human rights claims.

Third, like Beitz's account (and unlike Griffin's), the account should be political in the sense that it will conceive of human rights as having distinctively “political” kinds of agents as the primary duties bearers, coupled with a prima facie right on the part of other well-placed agents (be they political agents, individuals, or whatever) to intervene insofar as these duties are not being met. This does not mean that the primary duty bearers must be states, as suggested by Beitz, Rawls, and others. Rather, the primary duty bearers must have certain more general features that we would recognize as distinctively political: they must claim for themselves a certain kind of authority, they must wield power and have the capacity to coerce those over whom they exercise authority on a large scale, and so on. In our world, the duty bearers of human rights are often states and agents of the state, but we can imagine other kinds of entities having such characteristics. A reformed European Union and other supranational agencies could, for example, become primary bearers of such duties in a world in which political authority is not concentrated at the level of the state.

Let us call this the “structural pluralist” account of human rights. According to the structural pluralist account, human rights are moral rights that protect certain urgent human interests and that are directed in the first instance to authoritative institutions and those agents that carry out official roles within them. What is special about human rights is that they are moral rights that combine these structural and substantive properties.

A structural pluralist account of human rights appears to combine the strengths of the personhood and the practice-based accounts while avoiding their weaknesses. We saw that the personhood account encountered difficulties both in accounting for the full range of human rights and in explaining the sense in which human rights are political. A structuralist pluralist account of human rights deals with the first problem by expanding the list of fundamental interests beyond a narrow concern with our normative agency. And it deals with the second problem by conceiving of human rights as having a distinctively political structure.

The problem with the practice-based account is that it ties human rights too closely to contingent features of the actual world and that it elides the important distinction between human rights and their proper institutionalization. The structural pluralist account avoids the first problem by conceiving of the political character of human rights more broadly than the practice-based account. It therefore need not say that a world without states is necessarily a world without human rights. (To be sure, it does say that a world without political entities of any kind is a world without human rights, but this seems to be the intuitively correct result.) This account avoids the second problem by holding that the practice of human rights is irrelevant to what a human right is. It therefore allows us to maintain a strict distinction between human rights and their appropriate institutionalization within global practice.

What we have offered here is only a very cursory sketch. Much more would have to be done in order to assess the viability of the structural pluralist account, and of course there are different ways in which it could be specified. For example, Thomas Pogge has offered a kind of structural pluralist account in our sense, but we need not follow him in claiming that human rights are claims on the design of social institutions. Pogge's view has the counterintuitive implication that if a person is tortured by a “rogue” government official in a society that generally is very effective at preventing torture, then that person's human rights are not violated, even though their moral rights are.Footnote ²³ In our view, we should instead conceive of human rights as claims directed in the first instance against agents of a certain kind—those holding political authority of some sort—that they conduct themselves in certain ways. Such an account, we believe, offers the best hope of explaining what is special about human rights.