More precisely, rights are prima facie trumps. All things considered, rights may themselves be trumped by weighty other considerations. Claiming a right, however, in effect stops the conversation and both increases and shifts the burden of proof to those who would argue that this right in this particular case is itself appropriately trumped.1

Both rectitude and entitlement link right and obligation but in systematically different ways. Claims of rectitude (righteousness)—“That’s wrong,” “That’s not right,” “You really ought to do that”—focus on a standard of conduct and draw attention to the duty-bearer’s obligation under that standard. Rights claims, by contrast, focus on the right-holder and draw the duty-bearer’s attention to the right-holder’s special title to enjoy her right. Rights in this sense thus are sometimes called “subjective rights”; they focus on the subject (who holds them) rather than an “objective” standard to be followed or state of affairs to be realized.

Rights create—in an important sense are—a field of rule-governed interactions centered on, and under the control of, the right-holder. “A has a right to x (with respect to B)” specifies a right-holder (A), an object of the right (x), and a duty-bearer (B). It also outlines the relationships in which they stand. A is entitled to x (with respect to B), B stands under correlative obligations to A (with respect to x), and, should it be necessary, A may make special claims upon B to discharge those obligations.

Rights are not reducible to the correlative duties of those against whom they are held. If Anne has a right to x with respect to Bob, it is more than simply desirable, good, or even right that Anne enjoy x. She is entitled to it. Should Bob fail to discharge his obligations, besides acting improperly (i.e., violating standards of rectitude) and harming Anne, he violates her rights, making him subject to special remedial claims and sanctions.

Neither is having a right reducible to enjoying a benefit. Anne is not a passive beneficiary of Bob’s obligation. She is actively in charge of the relationship, as suggested by the language of “exercising” rights. She may assert her right to x. If he fails to discharge his obligation, Anne may press further claims against Bob, choose not to pursue the matter, or even excuse him, largely at her own discretion. Rights empower, not just benefit, those who hold them. Violations of rights are a particular kind of injustice with a distinctive force and remedial logic.

When we consider how rights work, though, one of the more striking facts is that we talk about rights only when they are at issue. If I walk into the supermarket and buy a loaf of bread, it would be odd to say that I had a right to my money, which I exchanged for a right to the bread. Only in unusual circumstances would we say that those who refrained from stealing my money or bread were respecting my rights. Rights are actually put to use, and thus important enough to talk about, only when they are at issue, when their enjoyment is questioned, threatened, or denied.

Three major forms of social interaction involving rights can be usefully distinguished.

Objective enjoyment must be the norm. For society, the costs associated with even active respect of a right must be the exception rather than the rule. Right-holders too would prefer not to have to exercise their rights. In an ideal world, rights would remain both out of sight and out of mind.

Nonetheless, the ability to claim rights, if necessary, distinguishes having a right from simply being the (rights-less) beneficiary of someone else’s obligation. Paradoxically, then, “having” a right is of most value precisely when one does not “have” (the object of) the right—that is, when active respect or objective enjoyment is not forthcoming. I call this the “possession paradox”: “having” and “not having” a right at the same time—possessing it but not enjoying it—with the “having” being particularly important precisely when one does not “have” it.

We thus should be careful not to confuse having a right with the respect it receives or the ease or frequency with which it is enforced. In a world of saints, rights would be widely respected, rarely asserted, and almost never enforced. In a Hobbesian state of nature, rights would never be respected. At best, disinterest or self-interest would lead duty-bearers to not deny the right-holder the object of her right. Only the accidental coincidence of interests (or self-help enforcement) would allow a right-holder to enjoy (the substance of) her right.

Differing circumstances of respect and enforcement tell us nothing about who has what rights. To have a right to x is to be specially entitled to x, whether the law that gave you a legal right is violated or not, whether the promise that gave rise to the contractual right is kept or not, whether others comply with the principles of righteousness that establish your moral right or not. I have a right to my car whether it sits in my driveway, is borrowed without my permission (for good reason or bad), is stolen but later recovered, or is stolen, never to be seen again by me (whether or not the thief is ever sought, apprehended, charged, tried, or convicted). Even if the violation ultimately goes unremedied and unpunished, the nature of the offense has been changed by my right.

Human rights are equal rights: one either is or is not a human being, and therefore has the same human rights as everyone else (or none at all). Human rights also are inalienable rights: one cannot stop being human, no matter how badly one behaves or how barbarously one is treated. And they are universal rights, in the sense that today we consider all members of the species Homo sapiens “human beings” and thus holders of human rights.

Much of this book explores the political implications of human rights being equal, inalienable, and universal. In this section I stress the implications of their being rights (in the sense discussed above) and their special role in enabling progressive political change.

For example, protection against arbitrary execution is an internationally recognized human right. The fact that people are not executed arbitrarily, however, may reflect nothing more than a government’s lack of desire. Even active protection may have nothing to do with a right (title) not to be executed. For example, rulers may act out of their sense of justice or follow a divine injunction that does not endow subjects with any rights. And even a right not to be arbitrarily executed may be a customary or statutory (rather than a human) right.

Such distinctions are more than scholastic niceties. Whether citizens have a right (title) shapes the nature of the injury they suffer and the forms of protection and remedy available to them. Denying someone something that it would be right for her to enjoy in a just world is very different from denying her something (even the same thing) that she is entitled (has a right) to enjoy. Furthermore, whether she has a human right or a legal right that has been contingently granted by the state dramatically alters both her relationship to the state and the character of her injury.

Armed with multiple claims, right-holders typically use the “lowest” right available. For example, in the United States, as in most countries, protection against racial discrimination in employment is available on several grounds. Depending on one’s employment agreement, a grievance may be all that is required, or a legal action based on the contract. If that fails (or is unavailable), one may be able to bring suit under a local ordinance or a state nondiscrimination statute. Federal statutes and the Constitution may offer remedies at still higher levels. In unusual cases, one may (be forced to) resort to international human rights claims. (In Europe, the European Court of Human Rights provides an intermediate stage between national and international law. See section 11.3.A.) In addition, a victim of discrimination may claim moral (rather than legal) rights—as well as appeal to non-rights-based considerations of justice or righteousness.

One can—and usually does—go very far before explicit appeals to human rights become necessary. The “higher” claims are always available; one still has those rights. In practice, though, they rarely are appealed to until lower-level remedies have been tried (if not exhausted). An appeal to human rights usually testifies to the absence of enforceable positive (legal) rights and suggests that everything else has been tried and failed, leaving one with nothing else (except perhaps violence).4 For example, homosexuals in the United States often claim a human right against discrimination because US courts have held that constitutional prohibitions of discrimination do not apply to sexual orientation. If rights are a sort of last resort, claimed only when things are not going well, human rights are a last resort in the realm of rights; no higher rights appeal is available.

Claims of human rights thus ultimately aim to be self-liquidating, giving the possession paradox a distinctive twist. Human rights claims characteristically seek to challenge or change existing institutions, practices, or norms—especially legal practices. Most often they seek to establish (or bring about more effective enforcement of) a parallel “lower” right. For example, claims of a human right to health care in the United States typically aim to create a legal right to health care. To the extent that such claims are politically effective, the need to make them in the future will be reduced or eliminated; the human rights claim will be replaced by a claim of ordinary legal rights.

A set of human rights thus can be seen as a standard of political legitimacy. The Universal Declaration of Human Rights, for example, presents itself as a “standard of achievement for all peoples and all nations.” To the extent that governments protect human rights, they are legitimate.

No less importantly, though, human rights authorize and empower citizens to act to vindicate their rights, to insist that these standards be realized, and to struggle t...