SEN, A. Elements of A Theory of Human Rights

AMARTYA SEN Elements of a Theory of
Human Rights
I. The Need for a Theory

Few concepts are as frequently invoked in contemporary political dis-
cussions as human rights. There is something deeply attractive in the
idea that every person anywhere in the world, irrespective of citizenship
or territorial legislation, has some basic rights, which others should
respect. The moral appeal of human rights has been used for a variety of
purposes, from resisting torture and arbitrary incarceration to demand-
ing the end of hunger and of medical neglect.1
At the same time, the central idea of human rights as something that
people have, and have even without any specific legislation, is seen by
many as foundationally dubious and lacking in cogency. A recurrent
question is, Where do these rights come from? It is not usually disputed
that the invoking of human rights can be politically powerful. Rather,
the worries relate to what is taken to be the “softness” (some would
say “mushiness”) of the conceptual grounding of human rights. Many
philosophers and legal theorists see the rhetoric of human rights as just
An earlier version of this article served as my Gilbert Murray Lecture (“Why Invent
Human Rights?”) given in Oxford on 14 November 2002. For helpful suggestions, I am
particularly grateful to the Editors of Philosophy & Public Affairs, and also to Catherine
Barnard, Rosanne Flynn, Sakiko Fukuda Parr, Ivan Hare, Will Kymlicka, Jo Miles, Martha
Nussbaum, Onora O’Neill, Siddiq Osmani, Mary Robinson, Emma Rothschild, Thomas
Scanlon, Arjun Sengupta, Frances Stewart, Rosemary Thorpe, and Rosie Vaughan.
1. See International Human Rights in Context: Law, Politics and Morals, ed. Henry J.
Steiner and Philip Alston (New York: Oxford University Press, 2000); Richard Falk, Human
Rights Horizons: The Pursuit of Justice in a Globalizing World (New York: Routledge, 2000);
Jack Donnelly, Universal Human Rights in Theory and Practice, 2nd ed. (Ithaca: Cornell
University Press, 2003). See also Micheline R. Ishay, The Human Rights Reader: Major Polit-
ical Writings, Essays, Speeches, and Documents from the Bible to the Present (New York:
Routledge, 1997).
©  by Blackwell Publishing, Inc. Philosophy & Public Affairs , no. 
316 Philosophy & Public Affairs
loose talk—perhaps kindly and well meaning forms of locution—but

loose talk nevertheless.
The contrast between the widespread use of the idea of human rights
and the intellectual skepticism about its conceptual soundness is not
new. The U.S. Declaration of Independence, in 1776, took it to be “self-
evident” that everyone is “endowed by their Creator with certain inalien-
able rights,” and thirteen years later, the French declaration of “the rights
of man” asserted that “men are born and remain free and equal in rights.”
But it did not take Jeremy Bentham long, in his Anarchical Fallacies
written during 1791 and 1792 (aimed against the French “rights of man”),
to propose the total dismissal of all such claims. Bentham insisted that
“natural rights is simple nonsense: natural and imprescriptible rights (an
American phrase), rhetorical nonsense, nonsense upon stilts.”2 That sus-
picion remains very alive today, and despite persistent use of the idea
of human rights in practical affairs, there are many who see the idea of
human rights as no more than “bawling upon paper,” to use another of
Bentham’s barbed portrayals of natural right claims.
The dismissal of human rights is often comprehensive and is aimed
against any belief in the existence of rights that people can have uncon-
ditionally, simply by virtue of their humanity (rather than having them
contingently, on the basis of specific qualifications, such as citizenship
or legal entitlements). Some critics, however, propose a discriminating
rejection: they accept the general idea of human rights but exclude, from
the acceptable list, specific classes of proposed rights, in particular the
so-called economic and social rights, or welfare rights. These rights,
which are sometimes referred to as second generation rights, such as a
common entitlement to subsistence or to medical care, have mostly
been added relatively recently to earlier enunciations of human rights,
thereby vastly expanding the claimed domain of human rights.3 These
2. Jeremy Bentham, Anarchical Fallacies; Being an Examination of the Declaration of

Rights Issued during the French Revolution (1792); republished in The Works of Jeremy
Bentham, vol. II, ed. J. Bowring (Edinburgh: William Tait, 1843), p. 501.
3. See Ivan Hare, “Social Rights as Foundational Human Rights,” in Social and Labour
Rights in Global Context, ed. Bob Hepple (Cambridge: Cambridge University Press, 2002),
and William F. Felice, The Global New Deal: Economic and Social Human Rights in World
Politics (Lanham: Rowman & Littlefield, 2003). See also Cass R. Sunstein, After the Rights
Revolution: Reconceiving the Regulatory State (Cambridge, Mass.: Harvard University Press,
1990), and Thomas W. Pogge, World Poverty and Human Rights: Cosmopolitan Responsi-
bilities and Reforms (London: Polity Press, 2002).
317 Elements of a Theory of Human Rights
additions have certainly taken the contemporary literature on human

rights well beyond the eighteenth-century declarations that concen-
trated on a narrower class of “rights of man,” including such demands
as personal liberty and political freedom. These newer inclusions have
been subjected to more specialized skepticism, with the critics focusing
on their feasibility problems and their dependence on specific social
institutions that may or may not exist.4
Human rights activists are often quite impatient with such critiques.
The invoking of human rights tends to come mostly from those who are
concerned with changing the world rather than interpreting it (to use a
classic distinction made famous, oddly enough, by that overarching the-
orist, Karl Marx). It is not hard to understand their unwillingness to
spend time trying to provide conceptual justification, given the great
urgency to respond to terrible deprivations around the world. This pro-
active stance has had its practical rewards, since it has allowed immedi-
ate use of the colossal appeal of the idea of human rights to confront
intense oppression or great misery, without having to wait for the theo-
retical air to clear. However, the conceptual doubts must also be satis-
factorily addressed, if the idea of human rights is to command reasoned
loyalty and to establish a secure intellectual standing. It is critically
important to see the relationship between the force and appeal of
human rights, on the one hand, and their reasoned justification and
scrutinized use, on the other.
There is, thus, need for some theory and also for some defense of any
proposed theory. The object of this article is to do just that, and to con-
sider, in that context, the justification of the general idea of human rights
and also of the includability of economic and social rights within the
broad class of human rights. For such a theory to be viable it is neces-
sary to clarify what kind of a claim is made by a declaration of human
rights, and how such a claim can be defended, and furthermore how the
diverse criticisms of the coherence, cogency and legitimacy of human
4. The reasoning behind such rejection has been powerfully presented by Maurice
Cranston, “Are There Any Human Rights?” Daedalus (1983): 1–17, and Onora O’Neill,
Towards Justice and Virtue (Cambridge: Cambridge University Press, 1996). See also the
critique of Michael Ignatieff, supporting some claims to human rights while strongly dis-
puting others, in Human Rights as Politics and Idolatry (Princeton: Princeton University
Press, 2001).
rights (including economic and social rights) can be adequately

addressed. That is the aim of this article.
However, before going into this investigation, I should make a clarifi-
catory point. The rhetoric of human rights is sometimes applied to par-
ticular legislations inspired by the idea of human rights. There is clearly
no great difficulty in seeing the obvious judicial status of these already
legalized entitlements. No matter what they are called (“human rights
laws” or any other appellation), they stand shoulder to shoulder with
other established legislations. The present inquiry on the foundations
and cogency of human rights does not have any direct bearing on the
obvious legal status of these “human rights laws,” once they have been
properly legislated. As far as these laws are concerned, the relevance, if
any, of this study would lie, rather, in the motivation that leads to the
enacting of such laws, which builds on the pre-legislative standing of
these claims.
Indeed, a great many acts of legislation and legal conventions (such
as the “European Convention for the Protection of Human Rights and
Fundamental Freedoms”) have been clearly inspired by a belief in some
pre-existing rights of all human beings. This applies even to the adop-
tion of the U.S. Constitution, including the Bill of Rights, linked to the
normative vision of the U.S. Declaration of Independence (as was noted
earlier). The difficult questions regarding the status and standing of
human rights arise in the domain of ideas, before such legalization
occurs. We also have to examine whether legislation is the pre-eminent,
or even a necessary, route through which human rights can be pursued.
II. Questions to Be Answered

A theory of human rights must address the following questions in
particular:
(1) What kind of a statement does a declaration of human rights
make?
(2) What makes human rights important?
(3) What duties and obligations do human rights generate?
(4) Through what forms of actions can human rights be promoted,
and in particular whether legislation must be the principal, or even a
necessary, means of implementation of human rights?
(5) Can economic and social rights (the so-called second generation
rights) be reasonably included among human rights?
(6) Last but not least, how can proposals of human rights be
defended or challenged, and how should their claim to a universal status
be assessed, especially in a world with much cultural variation and
widely diverse practice?
These questions are addressed sequentially in what follows. However,

since this is not a detective story, I am perhaps allowed to give away a
sketch of the proposed answers, with the hope that this might help in
following this long and not entirely uncomplicated article (even though
there is some risk of oversimplification involved in any summary
formulation).
(1) Human rights can be seen as primarily ethical demands. They
are not principally “legal,” “proto-legal” or “ideal-legal” commands. Even
though human rights can, and often do, inspire legislation, this is a
further fact, rather than a constitutive characteristic of human rights.
(2) The importance of human rights relates to the significance of the
freedoms that form the subject matter of these rights. Both the oppor-
tunity aspect and the process aspect of freedoms can figure in human
rights. To qualify as the basis of human rights, the freedoms to be
defended or advanced must satisfy some “threshold conditions” of (i)
special importance and (ii) social influenceability.
(3) Human rights generate reasons for action for agents who are in
a position to help in the promoting or safeguarding of the underlying
freedoms. The induced obligations primarily involve the duty to give rea-
sonable consideration to the reasons for action and their practical impli-
cations, taking into account the relevant parameters of the individual
case. The reasons for action can support both “perfect” obligations as
well as “imperfect” ones, which are less precisely characterized. Even
though they differ in content, imperfect obligations are correlative with
human rights in much the same way as perfect obligations are. In par-
ticular, the acceptance of imperfect obligations goes beyond volun-
teered charity or elective virtues.
(4) The implementation of human rights can go well beyond legis-
lation, and a theory of human rights cannot be sensibly confined within
the juridical model in which it is frequently incarcerated. For example,
public recognition and agitation (including the monitoring of violations)
can be part of the obligations—often imperfect—generated by the

acknowledgment of human rights. Also, some recognized human rights
are not ideally legislated, but are better promoted through other means,
including public discussion, appraisal and advocacy (a basic point
that would have come as no surprise to Mary Wollstonecraft, whose A
Vindication of the Rights of Woman: with Strictures on Political and
Moral Subjects was published in 1792).
(5) Human rights can include significant and influenceable eco-
nomic and social freedoms. If they cannot be realized because of inad-
equate institutionalization, then, to work for institutional expansion or
reform can be a part of the obligations generated by the recognition of
these rights. The current unrealizability of any accepted human right,
which can be promoted through institutional or political change, does
not, by itself, convert that claim into a non-right.
(6) The universality of human rights relates to the idea of surviv-
ability in unobstructed discussion—open to participation by persons
across national boundaries. Partisanship is avoided not so much by
taking either a conjunction, or an intersection, of the views respectively
held by dominant voices in different societies across the world (includ-
ing very repressive ones), but through an interactive process, in particu-
lar by examining what would survive in public discussion, given a
reasonably free flow of information and uncurbed opportunity to
discuss differing points of view. Adam Smith’s insistence that ethical
scrutiny requires examining moral beliefs from, inter alia, “a certain dis-
tance” has a direct bearing on the connection of human rights to global
public reasoning.
III. Human Rights: Ethics and Law

What kind of an assertion does a declaration of human rights make? I
would submit that proclamations of human rights are to be seen as artic-
ulations of ethical demands. They are, in this respect, comparable with
pronouncements in utilitarian ethics, even though their respective sub-
stantive contents are, obviously, very different. Like other ethical claims
that demand acceptance, there is an implicit presumption in making
pronouncements on human rights that the underlying ethical claims will
survive open and informed scrutiny. Indeed, the invoking of such an
interactive process of critical scrutiny, open to information (including

that about other societies) as well as to arguments coming from far as
well as near, is a central feature of the theory of human rights proposed
here. It differs both (i) from trying to justify the ethics of human rights
in terms of shared—and already established—universal values (the
uncomplicated “non-partisan” view), and (ii) from abdicating any claim
of adherence to universal values (and in this sense, eschewing any claim
to being “non-partisan”) in favor of a particular political conception that
is suitable to the contemporary world.5
These issues, which relate to the foundational discipline of ethical cri-
tique, will be examined later, in Section IX, in response to question (6).
But the point to note for the moment, in answer to the first question, is
that pronouncements of human rights are quintessentially ethical artic-
ulations, and they are not, in particular, putative legal claims, despite
considerable confusion on this point, generated not least by Jeremy
Bentham, the obsessive slayer of what he took to be legal pretensions. (I
shall return later in this section to the nature of the misapprehension
involved.)
A pronouncement of human rights includes an assertion of the
importance of the corresponding freedoms—the freedoms that are iden-
tified and privileged in the formulation of the rights in question—and is
indeed motivated by that importance. For example, the human right of
not being tortured springs from the importance of freedom from torture
for all. But it includes, furthermore, an affirmation of the need for others
to consider what they can reasonably do to secure the freedom from
torture for any person. For a would-be torturer, the demand is obviously
quite straightforward, to wit, to refrain and desist. The demand takes the
clear form of what Immanuel Kant called a perfect obligation.6 However,
for others too (that is, those other than the would-be torturers) there are
responsibilities, even though they are less specific and come in the
general form of “imperfect obligations” (to invoke another Kantian
5. There are different variants of these two contrasting positions, and also other alter-
natives that differ from both, which are helpfully discussed and distinguished in Charles
Beitz, “Human Rights as a Common Concern,” American Political Science Review 95 (June
2001): 269–82.
6. Immanuel Kant, Critique of Practical Reason (1788), trans. L. W. Beck (New York:
Bobbs-Merrill, 1956).
concept).7 The perfectly specified demand not to torture anyone is sup-

plemented by the more general, and less exactly specified, requirement
to consider the ways and means through which torture can be prevented
and then to decide what one should, thus, reasonably do. The relations
between human rights, freedoms, and obligations will be further inves-
tigated in Sections IV through VI.
Even though recognitions of human rights (with their associated
claims and obligations) are ethical affirmations, they need not, by them-
selves, deliver a complete blueprint for evaluative assessment. An agree-
ment on human rights does involve a firm commitment, to wit, to give
reasonable consideration to the duties that follow from that ethical
endorsement. But even with agreement on these affirmations, there can
still be serious debates, particularly in the case of imperfect obligations,
on (i) the ways in which the attention that is owed to human rights
should be best paid, (ii) how the different types of human rights should
be weighed against each other and their respective demands integrated
together, (iii) how the claims of human rights should be consolidated
with other evaluative concerns that may also deserve ethical attention,
and so on.8 A theory of human rights can leave room for further discus-
sions, disputations and arguments. The approach of open public
reasoning, which is central to the understanding of human rights as
proposed here, can definitively settle some disputes about coverage
and content (including the identification of some clearly sustainable
rights and others that would be hard to sustain), but may have to
leave others, at least tentatively, unsettled.9 The admissibility of a
7. I have discussed, in an earlier paper, the relevance of the Kantian distinction

between “perfect” and “imperfect” obligations even for a largely consequentialist frame-
work. “Consequential Evaluation and Practical Reason,” Journal of Philosophy 97 (2000):
477–502.
8. Some of the central issues are discussed by John Mackie, “Can There Be a Rights-
based Moral Theory?” Studies in Ethical Theory: Midwest Studies in Philosophy 3, ed. Peter
A. French, et al. (Morris : University of Minnesota, 1978). Reprinted in Theories of Rights,
ed. Jeremy Waldron (Oxford: Oxford University Press, 1984), pp. 168–81.
9. This applies not only to the persistence of interpersonal disagreements, but also to
specific areas of unresolved disputes within one person’s own reasoned assessment. An
adequate theory of rationality has to make room for such “incompleteness” of assessment.
The general issue of admissibility of incompleteness is discussed in my Collective Choice
and Social Welfare (San Francisco: Holden-Day, 1970; republished, Amsterdam: North-
Holland, 1979); “Maximization and the Act of Choice,” Econometrica 65 (1997): 745–80,
domain of continued dispute is no embarrassment to a theory of human

rights.10
In practical applications of human rights, such debates are, of course,
quite common and entirely customary, particularly among human rights
activists. What is being argued here is that the possibility of such
debates—without losing the basic recognition of the importance of
human rights—is not just a feature of what can be called human rights
practice, they are actually part of the general discipline of human rights
including the underlying theory (rather than being an embarrassment to
that discipline). An acknowledgment of the necessity to pay ethical
attention to human rights, far from obliterating the need for such delib-
eration, actually invites it. A theory of human rights can, therefore, allow
considerable internal variations, without losing the commonality of the
agreed principle of attaching substantial importance to human rights
(and to the corresponding freedoms and obligations) and of being com-
mitted to considering seriously how that importance should be appro-
priately reflected.
Variability of this kind is not only not an embarrassment, it tends to
be standardly present in all general theories of substantive ethics.
Indeed, a similar diversity can be found within utility-centered ethics,
even though this feature of that large ethical discipline often receives
little or no recognition. In the case of utility-based reasoning, variations
can arise not only from the different ways in which utilities can be inter-
preted (as pleasures, fulfillment of desires, or realization of choices),11 nor
only from the acknowledged heterogeneity of utilities themselves (well
reprinted in Rationality and Freedom (Cambridge, Mass.: Harvard University Press, 2002);
and also “Incompleteness and Reasoned Choice,” Synthese (forthcoming 2004). See also
Isaac Levi, Hard Choices: Decision Making under Unresolved Conflict (Cambridge:
Cambridge University Press, 1986), and Hilary Putnam, “Über die Rationalität von
Präferenzen,” Allgemeine Zeitschrift für Philosophie 21 (1996): 204–28, English version, “On
the Rationality of Preferences,” in his The Collapse of the Fact / Value Dichotomy and Other
Essays (Cambridge, Mass.: Harvard University Press, 2002).
10. Also, as Jeremy Waldron has argued, disagreement about rights “is a sign—the best
possible sign in modern circumstances—that people take rights seriously.” See Law and
Disagreement (Oxford: Oxford University Press, 2001), p. 311.
11. See Jeremy Bentham, An Introduction to the Principles of Morals and Legislation
(London: Payne, 1789; republished, Oxford: Clarendon Press); Henry Sidgwick, The Method
of Ethics (London: Macmillan, 1874); A. C. Pigou, The Economics of Welfare (London:
MacMillan, 1920); Frank P. Ramsey, Foundations: Essays in Philosophy, Logic, Mathematics
recognized by both Aristotle and John Stuart Mill).12 They can also arise
from the diversity of ways in which utilities can be used, whether by mere
addition, or by multiplication (after suitable normalization), or through
the addition of concave transformations of utility functions, all of which
have been proposed and pursued, within the discipline of utility-based
evaluation.13 Further, the discipline of interpersonal comparison of util-
ities may itself allow alternative procedures of quantification of utilities
and go comfortably with accommodating permissible variations within
specified classes of “partial comparability.”14 The existence of different
ways of making use of utility-based reasoning and alternative utilitarian
procedures does not invalidate or even undermine the general approach
of utility-centered ethics. And, similarly, the ethics of human rights is
not nullified or thwarted by internal variations that it allows and
incorporates.
Thus, the analogy between articulations of human rights and utilitar-
ian pronouncements has considerable perspicacity, even though the
great founder of modern utilitarianism, Jeremy Bentham, managed to
miss that connection altogether in his classic hatchet job on natural
rights in general and on the “rights of man” in particular. Bentham took
and Economics (London: Routledge, 1978); Richard M. Hare, Freedom and Reason (Oxford:
Clarendon Press, 1963); J.C.B. Gosling, Pleasure and Desire (Oxford: Clarendon Press,
1969); Derek Parfit, Reasons and Persons (Oxford: Clarendon Press, 1984); R. E. Goodin,
“Laundering Preferences,” in The Foundations of Social Choice Theory, ed. Jon Elster and
Aanund Hylland (Cambridge: Cambridge University Press, 1986), pp. 75–101; James Griffin,
Well-being (Oxford: Clarendon Press, 1986); John Broome, Weighing Goods (Oxford:
Blackwell, 1991); among many other contributions.
12. See Aristotle, The Nicomachean Ethics, trans. David Ross, rev. ed. (Oxford: Claren-
don Press, 1980), and John Stuart Mill, Utilitarianism (London, 1861; republished London:
Collins/Fontana, 1962). I have discussed the issues involved in “Plural Utility,” Proceedings
of the Aristotelian Society 81 (1980–81): 193–215.
13. See John F. Nash, “The Bargaining Problem,” Econometrica 18 (1950): 155–62; John
C. Harsanyi, “Cardinal Welfare, Individualistic Ethics, and Interpersonal Comparisons of
Utility,” Journal of Political Economy 63 (1955): 309–21; James A. Mirrlees, “An Exploration
of the Theory of Optimum Income Taxation,” Review of Economic Studies 38 (1971): 175–208.
14. These issues are discussed in Amartya Sen, “Interpersonal Aggregation and Partial
Comparability,” Econometrica 38 (1970): 393–409, and Choice, Welfare and Measurement
(Oxford: Blackwell, 1982; republished, Cambridge, Mass.: Harvard University Press, 1997);
Charles Blackorby, “Degrees of Cardinality and Aggregate Partial Orderings,” Econometrica
43 (1975): 845–52; Ben J. Fine, “A Note on ‘Interpersonal Aggregation and Partial Compara-
bility’,” Econometrica 43 (1975): 169–72; Interpersonal Comparisons of Well-beings, ed. Jon
Elster and John Roemer (Cambridge: Cambridge University Press, 1991).
the appropriate comparison to be that between the legal significance,

respectively of: (1) declarations of human rights, and (2) actually legis-
lated rights. Not surprisingly, he found the former to be essentially
lacking in legal status in the way the latter, obviously enough, would
have. Bentham’s dismissal of human rights came, thus, with amazingly
swiftness.
Right, the substantive right, is the child of law; from real laws come
real rights; but from imaginary laws, from “law of nature” [can come
only] “imaginary rights.”15
It is easy to see that Bentham’s rejection of the idea of natural “rights

of man” depends substantially on the rhetoric of privileged use of the
term of “rights,” seeing it in its specifically legal interpretation. However,
insofar as human rights are meant to be significant ethical claims, the
pointer to the fact that they do not by themselves have legal or institu-
tional force is obvious enough, but also quite irrelevant to the discipline
of human rights.16 The appropriate comparison is, surely, between:
(1) a utility-based ethics (championed by Bentham himself), which
sees intrinsic ethical importance in utilities but none in human rights or
human freedoms (any role that the latter can have in the utilitarian
system is, thus, entirely instrumental), and
(2) an ethics that makes room for the fundamental significance of
human rights (as the advocates of “rights of man” did), linked with a
diagnosis of the basic importance of human freedoms and the obliga-
tions generated by that diagnosis.17
15. Jeremy Bentham, Anarchical Fallacies, in Collected Works, vol. II, p. 523.
16. Accepting a general contrast between the respective categories of ethical assertions
and legal pronouncements does not, of course, deny the possibility that ethical views may
contribute to the interpretation and, thus, the substantive content of laws. The recogni-
tion of that possibility may go against a strictly positivist theory of law (on which see
Ronald Dworkin, A Matter of Principle [Cambridge, Mass.: Harvard University Press, 1985]).
This understanding does not, however, obliterate the motivational and substantive dis-
tinction between primarily ethical claims and principally legal proclamations.
17. The importance of rights and freedoms can, of course, be combined with incorpo-
rating the significance of utility or well-being in ethical reasoning, but if such a “combined”
system is to be pursued, some consistency problems will have to be faced in devising
a coherent and integrated social choice procedure; on this see Amartya Sen, “The
Just as utilitarian ethical reasoning takes the form of insisting that the
utilities of the relevant persons must be taken into account in deciding
on what should be done, the human rights approach demands that the
acknowledged human rights must be given ethical recognition (the form
and the informational basis of that recognition will be discussed further
in the next two sections). The relevant comparison lies in this contrast,
not in differentiating the legal force of legislated rights (for which
Bentham’s phrase “the child of law” is an appropriate description) from
the absence of any legal standing generated by an ethical recognition of
rights (without any legislation or legal reinterpretation). Indeed, even as
Bentham was busy in 1791 and 1792 writing down his dismissal of “rights
of man,” the reach and range of ethical interpretations of rights were
being powerfully explored by Thomas Paine’s Rights of Man, and by Mary
Wollstonecraft’s A Vindication of the Rights of Woman: with Strictures on
Political and Moral Subjects, both published during the period 1791 to
1792 (though neither seemed to arouse Bentham’s curiosity).18
An ethical understanding of human rights goes not only against
seeing them as legal demands (and against taking them to be, as in
Bentham’s view, legal pretensions), but also differs from a law-centered
approach to human rights that sees them as if they are basically grounds
for law, almost “laws in waiting.” Ethical and legal rights do, of course,
have motivational connections. In a rightly celebrated article “Are
There Any Natural Rights?” Herbert Hart has argued that people “speak
of their moral rights mainly when advocating their incorporation in a
legal system.” He added that the concept of a right “belongs to that
branch of morality which is specifically concerned to determine when
one person’s freedom may be limited by another’s and so to determine
what actions may appropriately be made the subject of coercive legal
Impossibility of a Paretian Liberal,” Journal of Political Economy 78 (1970): 152–57, and

Rationality and Freedom (Cambridge, Mass.: Harvard University Press, 2002), essays 12–14
and 20–22. See also Robert Nozick, Anarchy, State and Utopia (New York: Basic Books, 1974),
and the special number of Analyse & Kritik 18 (1996) on “the Liberal Paradox,” particularly
Kotaro Suzumura, “Welfare, Rights and Social Choice Procedures,” pp. 20–37.
18. Thomas Paine, The Rights of Man: Being an Answer to Mr. Burke’s Attack on the
French Revolution (1791); second part, Combining Principle and Practice (1792); repub-
lished, The Rights of Man (London: Dent, and New York: Dutton, 1906). Mary
Wollstonecraft, A Vindication of the Rights of Woman (1792); republished, The Rights of
Woman (London: Dent, and New York: Dutton, 1929).
rules.”19 Whereas Bentham saw rights as a “child of law,” Hart’s view takes
the form, in effect, of seeing some natural rights as parents of law: they
motivate and inspire specific legislations. Although Hart does not make
any reference whatever to human rights in his article, the reasoning
about the role of natural rights as inspiration for legislation can be seen
to apply to the concept of human rights as well.20
There can, in fact, be little doubt that the idea of moral rights can
serve, and has often served in practice, as the basis of new legislation. It
has frequently been utilized in this way, and this is indeed an important
use of human rights. That, for example, is precisely the way the diagno-
sis of inalienable rights was invoked in the U.S. Declaration of Indepen-
dence and reflected subsequently in the Bill of Rights, a route that has
been well-trodden in the legislative history of many countries in the
world.21 Providing inspiration for legislation is certainly one way in
which the ethical force of human rights has been constructively
deployed.
However, to acknowledge that such a connection exists is not the
same as taking the relevance of human rights to lie exclusively in deter-
mining what should “appropriately be made the subject of coercive legal
rules.” It is important to see that the idea of human rights can be, and is,
actually used in several other ways as well. Indeed, if human rights are
seen as powerful moral claims, indeed as “moral rights” (to use Hart’s
phrase), then surely we have reason for some catholicity in considering
different avenues for promoting these claims. (This question will be
pursued in Section VII.) The ways and means of advancing and imple-
menting human rights need not, thus, be confined only to making new
laws (even though sometimes legislation may indeed turn out to be the
right way to proceed). For example, monitoring and other activist
support, provided by such organizations as Human Rights Watch or
19. H. L. A. Hart, “Are There Any Natural Rights?” The Philosophical Review 64 (1955),
reprinted in Theories of Rights, ed. Jeremy Waldron (Oxford: Oxford University Press, 1984),
p. 79.
20. On this see Maurice Cranston, “Are There Any Human Rights?”
21. The framers of the Universal Declaration of Human Rights in 1948 hoped, in fact,
that this declaration would serve as a template for bills of rights in different nations, with
national courts taking a lead in their enforcement. See Mary Ann Glendon’s wonderful
account of that remarkable history, A World Made New: Eleanor Roosevelt and the Univer-
sal Declaration of Human Rights (New York: Random House, 2001).
Amnesty International or OXFAM or Médicins Sans Frontièrs, can them-

selves help to advance the effective reach of acknowledged human
rights.22 In many contexts, legislation may not, in fact, be involved.
IV. Rights, Freedoms and Social Influence

Why are human rights important? Since declarations of human rights are
ethical affirmations of the need to pay appropriate attention to the sig-
nificance of freedoms incorporated in the formulation of human rights
(as was discussed in the last section), an appropriate starting point must
be the importance of freedoms of human beings to be so recognized.
Note that while rights involve claims (specifically, claims on others who
are in a position to make a difference), freedoms, in contrast, are pri-
marily descriptive characteristics of the conditions of persons.23
By starting from the importance of freedoms as the appropriate
human condition on which to concentrate, rather than on utilities (as
Bentham did), we get a motivating reason not only for celebrating
our own rights and liberties, but also for our taking an interest in the
significant freedoms of others, not just in their pleasures and desire-
fulfillment (as under utilitarianism). Bentham’s insistence on choosing
utility as the basis of ethical evaluation can be contrasted with the
reasons for focusing instead on freedoms. I have discussed elsewhere
why those reasons are weighty and how the focus on freedoms can avoid
some of the major pitfalls of concentrating only on utility in the form of
pleasure or desire fulfillment. For example, the utilitarian calculus can
suffer from valuational distortions resulting from the neglect of sub-
stantive deprivation of those who are chronically disadvantaged but who
learn, by force of circumstances, to take pleasure in small mercies and
22. Since the Gilbert Murray Lecture given at Oxford in November 2002, in which this
article originated, was arranged by OXFAM (Gilbert Murray was one of OXFAM’s founders),
it was also a suitable occasion to discuss this broader connection of human rights with a
plurality of ways of pursuing them.
23. However, the ethical force of freedoms can help to generate claims on others. On
different aspects of the “entanglements” between descriptive and evaluative concerns, see
Hilary Putnam, The Collapse of the Fact / Value Dichotomy and Other Essays (Cambridge,
Mass.: Harvard University Press, 2002). See also William Van Orman Quine, “Two Dogmas,
Of Empiricism,” in his From a Logical Point of View (Cambridge, Mass.: Harvard University
Press, 1961), pp. 20–46, and Vivian Walsh, “Philosophy and Economics,” in The New Pal-
grave: A Dictionary of Economics, ed. John Eatwell, Murray Milgate and Peter Newman
(London: Macmillan, 1987), pp. 861–69.
get reconciled to cutting down their desires to “realistic proportions”

(thereby appearing to be not particularly deprived in the special metric
of pleasures or desire-fulfillment).24
Before going into the difficult issue of duties related to rights, which
will be examined in Section VI, some exploration of the connection
between rights and freedoms is necessary, to which I devote the rest of
this section as well as Section V. Freedoms can vary in importance and
also in terms of the extent to which they can be influenced by social help.
For a freedom to count as a part of the evaluative system of human
rights, it clearly must be important enough to justify requiring that
others should be ready to pay substantial attention to decide what they
can reasonably do to advance it. It also has to satisfy a condition of plau-
sibility that others could make a material difference through taking such
an interest.
There have to be some “threshold conditions” of (i) importance and
(ii) social influenceability for a freedom to figure within the interper-
sonal and interactive spectrum of human rights. Insofar as the idea of
human rights demands public discussion and engagement, which I
noted earlier and will further discuss in Section IX, the agreement that
would be sought is not only on whether some specific freedom of a
particular person has any ethical importance whatsoever (that condi-
tion can be easy to satisfy), but also whether its significance and its
influenceability meet the threshold conditions for inclusion among the
human rights on which the society should focus.
The threshold conditions may prevent, for a variety of reasons, par-
ticular freedoms from being an appropriate subject matter of human
rights. To illustrate, it is not hard to argue that some importance should
be attached to all four of the following freedoms:
(1) a person’s freedom not to be assaulted;
(2) her freedom to receive medical care for a serious health problem;
(3) her freedom not to be called up regularly by her neighbors whom
she detests;
(4) her freedom to achieve tranquillity.
24. The evaluative framework of substantive freedoms can provide a more robust
appreciation of a person’s inability to achieve what they have reason to value. On this see
my “Well-being, Agency and Freedom: The Dewey Lectures 1984,” Journal of Philosophy 82
(1985): 169–220; Inequality Reexamined; and Development as Freedom (New York: Knopf,
1999).
However, even though all four may be important in one way or another,
it is not altogether implausible to argue that the first (freedom not to be
assaulted) is a good subject matter for a human right, and so is the
second (freedom to receive necessary medical care),25 but the third
(freedom not to be called up by detested neighbors) is not, in general,
important enough to cross the threshold of social significance to qualify
as a human right. Also, the fourth, while quite possibly extremely impor-
tant for the person, is too inward-looking—and too hard to be influenced
by others—to be a good subject matter for human rights. The exclusion
of a “right to tranquillity” relates not to any skepticism about the pos-
sible importance of tranquillity and the significance of a person’s
being free to achieve it, but to the difficulty of guaranteeing it through
social help.
There can be fruitful debates on the thresholds and their use, and in
particular on whether a specific case of freedom meets the threshold
conditions or not. As was briefly discussed in Sections II and III (and will
be further examined in Section IX), such discussions are part of the dis-
cipline of human rights. The analyses of thresholds, related both to the
seriousness and to the social influenceability of particular freedoms,
cannot but have a significant place in the discipline of human rights.
V. Processes, Opportunities and Capabilities

I turn now to a closer scrutiny of the contents of freedom and its multi-
ple features. I have argued elsewhere that “opportunity” and “process”
are two aspects of freedom that require distinction, with the importance
of each deserving specific acknowledgment.26 An example can help to
bring out the separate (though not necessarily independent) relevance
of both substantive opportunities and freedom of processes.
Consider an adult person, let us call her Rima, who decides that she
would like to go out in the evening. To take care of some considerations
25. However, in the second case (that is, the entitlement to necessary medical care), we
shall have to discuss whether this type of a “welfare right,” or more generally, economic
and social rights, can be seen as human rights, and this examination will be taken up in
Section VIII.
26. See Rationality and Freedom (Cambridge, Mass.: Harvard University Press, 2002),
particularly my Arrow Lectures (“Freedom and Social Choice”) included there: essays 20
through 22.
that are not central to the issues involved here (but which could make
the discussion more complex), it is assumed that there are no particular
safety risks involved in her going out, and that she has critically reflected
on this decision and judged that going out would be the sensible, indeed
the ideal, thing to do. Now consider the threat of a violation of this
freedom if some authoritarian guardians of society decide that she must
not go out in the evening (“it is most unseemly”), and if they force her,
in one way or another, to stay indoors. To see that there are two distinct
issues involved in this one violation, consider an alternative case in
which the authoritarian bosses decide that she must—absolutely must—
go out (“you are expelled for the evening: just obey”). There is clearly a
violation of freedom here even though Rima is being forced to do exactly
what she would have chosen to do anyway, and this is readily seen when
we compare the two alternatives “choosing freely to go out” and “being
forced to go out.” The latter involves an immediate violation of the
process aspect of Rima’s freedom, since an action is being forced on her
(even though it is an action she would have freely chosen also).
The opportunity aspect may also be affected, since a plausible
accounting of opportunities can include having options and it can inter
alia include valuing free choice. However, the violation of the opportu-
nity aspect would be more substantial and manifest if she were not only
forced to do something chosen by another, but in fact, forced to do
something she herself would not otherwise choose to do. The compari-
son between “being forced to go out” (when she would have gone out
anyway, if free) and, say, “being forced to polish the shoes of others at
home” (not her favorite activity) brings out this contrast, which is pri-
marily one of the opportunity aspect, rather than the process aspect. In
being forced to stay home and polish the shoes of others, Rima loses
freedom in two different ways, related respectively to (1) being forced
with no freedom of choice, and (2) being obliged in particular to do
something she would not choose to do.27
Both processes and opportunities can figure in human rights. A denial
of “due process” in being, say, imprisoned without a proper trial can be
the subject matter of human rights (no matter what the outcome of the
27. More complex features of the opportunity aspect and the process aspect of
freedoms are also discussed in my Arrow Lectures (“Freedom and Social Choice”) in
Rationality and Freedom, essays 20 through 22.
fair trial might be), and so can be the denial of the opportunity of
medical treatment, or the opportunity of living without the danger of
being assaulted (going beyond the exact process through which these
opportunities are made real).
For the opportunity aspect of freedom, the idea of “capability” (that
is, the opportunity to achieve valuable combinations of human func-
tionings: what a person is able to do or be) can typically provide a helpful
approach.28 It allows us to distinguish appropriately between (1) what she
values doing or being, and (2) the means she has to achieve what
she values. By shifting attention, in particular, towards the former, the
capability-based approach resists an overconcentration on means (such
as incomes and primary goods) that can be found in some theories of
justice (for example, in the Rawlsian Difference Principle). The capabil-
ity approach can capture the fact that two persons can have very differ-
ent substantial opportunities even when they have exactly the same set
of means: for example, a disabled person can do far less than an able-
bodied person can, with exactly the same income and other “primary
goods.” The disabled person cannot, thus, be judged to be equally
advantaged—with the same substantive opportunities—as the person
without any physical handicap but with the same set of means (such as
income and wealth and other primary goods). The capability perspec-
tive concentrates on what actual opportunities a person has, not the
means over which she has command. More particularly, the capability
perspective allows us to take into account the parametric variability in
the relation between the means, on the one hand, and the actual oppor-
tunities, on the other.29
28. On the concept of capability, see my “Equality of What?” in Tanner Lectures on

Human Values, vol. I, ed. Sterling M. McMurrin (Cambridge: Cambridge University Press,
and Salt Lake City: University of Utah Press, 1980), pp. 197–220, and Commodities and Capa-
bilities (Amsterdam: North-Holland, 1985), and also, jointly edited with Martha Nussbaum,
The Quality of Life (Oxford: Clarendon Press, 1993). The approach is powerfully developed
and applied by Martha Nussbaum, Women and Human Development: The Capabilities
Approach (Cambridge: Cambridge University Press, 2000). See also the related theories of
substantial opportunities developed by Richard Arneson, “Equality and Equality of Oppor-
tunity for Welfare,” Philosophical Studies 56 (1989): 77–112; G. A. Cohen, “On the Currency
of Egalitarian Justice,” Ethics 99 (1989): 906–44; and John E. Roemer, Theories of Distribu-
tive Justice (Cambridge, Mass.: Harvard University Press, 1996), among others contributors.
29. The importance of this variability for a theory of justice is discussed in my “Justice:
Means versus Freedoms,” Philosophy & Public Affairs 19 (1990): 111–21. Differences in the
capability to function can arise even with the same set of personal means (such as primary
The capability perspective can also help in bringing out the need for
transparent valuational scrutiny of individual advantages and adversi-
ties, since the different functionings have to be assessed and weighted in
relation to each other, and the opportunities of having different combi-
nations of functionings also have to be evaluated.30 The richness of the
capability perspective broadly interpreted, thus, includes its insistence
on the need for open valuational scrutiny for making social judgments,
and in this sense, it fits in well with the importance of public reasoning.31
This openness of transparent valuation contrasts with burying the eval-
uative exercise in some mechanical, and valuationally opaque, con-
vention (for example, by taking market-evaluated income to be the
invariable standard of individual advantage, thereby giving implicit nor-
mative priority to institutionally determined market prices).
goods) for a variety of reasons, such as (1) personal heterogeneities (related, for example, to
disability, or proneness to illness), (2) environmental diversities (such as climatic condi-
tions, or varying threats from epidemic diseases or from local crime), (3) variations in
non-personal resources (such as the nature of public health care, or social cohesion), or (4)
different relative positions vis-à-vis others (well illustrated by Adam Smith’s discussion, in
the Wealth of Nations, of the fact that the clothing and other resources one needs “to
appear in public without shame” depends on what other people standardly wear and how
they typically live in that society).
30. The need for an explicit valuational exercise is, thus, seen as an advantage, rather
than a limitation of the capability approach. For arguments in different directions on this
issue, see Charles R. Beitz, “Amartya Sen’s Resources, Values and Development,” Econom-
ics and Philosophy 2 (1986): 282–90; Bernard Williams, “The Standard of Living: Interests
and Capabilities,” in Amartya Sen et al., The Standard of Living, ed. Geoffrey Hawthorn
(Cambridge: Cambridge University Press, 1987), pp. 94–102; Amartya Sen, Inequality
Reexamined, and “Capability and Well-being,” in The Quality of Life, ed. Nussbaum and
Sen, pp. 31–53.
31. The capability approach can allow considerable difference in application. For a
somewhat different perspective, see Martha Nussbaum, “Nature, Function, and Capabil-
ity: Aristotle on Political Distribution,” Oxford Studies in Ancient Philosophy, Supplemen-
tary Volume (1988), pp. 145–54, and Women and Human Development: The Capabilities
Approach. Nussbaum has discussed the importance of identifying an overarching “list of
capabilities,” with given priorities, in a more Aristotelian way. My own reluctance to join
the search for such a canonical list arises partly from my difficulty in seeing how the exact
lists and weights would be chosen without appropriate specification of the context of their
use (which could vary), but also from a disinclination to accept any substantive diminu-
tion of the domain of public reasoning. The framework of capabilities, as I see it, helps to
clarify and illuminate the subject matter of public reasoning, which can involve epistemic
issues (including claims of objective importance) as well as ethical and political ones. It
does not—and cannot—displace the need for public reasoning.
There has, however, been some serious criticism of describing these

substantive opportunities—to live one kind of a life or another—as “free-
doms,” and it has been argued that this makes the idea of freedom too
inclusive. For example, in her illuminating and sympathetic critique of
my Development as Freedom, Susan Okin has presented arguments to
suggest that I tend “to overextend the concept of freedom.”32 She argues:
“It is hard to conceive of some human functionings, or the fulfillment of
some needs and wants, such as good health and nourishment, as free-
doms without stretching the term until it seems to refer to everything
that is of central value to human beings” (p. 292).
There is indeed scope for argument on how broadly the concept of
freedom should be used. But the particular example considered in Okin’s
counterargument is, I think, based on a misinterpretation of the idea
of freedom underlying the concept of capability. It has not been
suggested at all that a functioning (for example, being in good health or
being well-nourished) should be seen as freedom of any kind. Rather,
freedom, in the form of capability, concentrates on the opportunity to
achieve combinations of functionings (including, inter alia, the oppor-
tunity to be well-nourished or in good health, as in this particular case):
the person is free to use this opportunity or not. A capability reflects the
alternative combinations of functionings over which the person has
freedom of effective choice.
It is, therefore, not being suggested at all that being well-nourished or
in good health is to be seen as a freedom in itself.33 Capability, as a kind
of freedom, refers to the extent to which the person is able to choose
particular combinations of functionings (including, inter alia, such
things as being well-nourished), no matter what the person actually
decides to choose. Mahatma Gandhi famously did not use that
opportunity to be well-fed when he chose to fast, as a protest against the
32. Susan Okin, “Poverty, Well-being and Gender: What Counts, Who’s Heard?” Philos-
ophy & Public Affairs 31 (2003): 280–316. On related issues see also Joshua Cohen, “Review
of Sen’s Inequality Reexamined,” Journal of Philosophy 92 (1994): 275–88, esp. 278–80, and
G. A. Cohen, “Review: Amartya Sen’s Unequal World,” The New Left Review (1995): 117–29,
esp. 120–25.
33. I have discussed this issue in “Well-being, Agency and Freedom: The Dewey Lec-
tures 1984.” It is also important to examine how the concept of “freedom” links with a
broadly defined idea of “interest,” which underlies Joseph Raz’s reasoned diagnosis: “Rights
ground requirement for action in the interest of other beings.” See The Morality of Freedom
(Oxford: Clarendon Press, 1986), p. 180.
policies of the British Raj in India. In terms of the actual functioning of

being well-nourished, the fasting Gandhi did not differ from a starving
famine victim, but the freedoms and opportunities they respectively had
were quite different. The freedom to have any particular thing can be dis-
tinguished from actually having that thing. What a person is free to have,
not just what he or she actually has, is relevant, I have argued, to a theory
of justice.34 A similar point can be made about the relevance of substan-
tive freedoms in a theory of human rights.
The fact that many of the terrible deprivations in the world seem to
arise from a lack of freedom to avoid those deprivations (rather than
from choice, including choosing to be “indolent”: a classic issue in the
historical literature on poverty) is an important motivational reason to
emphasize the role of freedom. This led Marx to argue passionately for
the need to replace “the domination of circumstances and chance over
individuals by the domination of individuals over chance and circum-
stances.”35 The general idea of freedom, with its many distinct compo-
nents, seems particularly relevant to normative social choice theory, in
general, and to the theory of justice, in particular. The argument here is
that it can also figure powerfully in the normative foundations of human
rights.
To take a different type of example, consider the freedom of new
immigrants to West Europe or North America to conserve the ancestral
cultural customs and life-styles from their countries of origin. This
complex subject cannot be adequately assessed without distinguishing
between doing something and being free to do that thing. A strong argu-
ment can be constructed in favor of an immigrant’s having the freedom
to retain at least parts of her ancestral life-style, but this must not be seen
as an argument in favor of her pursuing her ancestral life-style whether
she chooses to do this or not. The central issue, in this argument, is the
freedom to choose how she should live, including the opportunity to
34. G. A. Cohen has presented arguments in favor of focusing on achieved function-

ings—related to his concept of “midfare”—rather than on capability; see his “On the
Currency of Egalitarian Justice,” and “Equality of What? On Welfare, Resources and
Capabilities,” in The Quality of Life, ed. Nussbaum and Sen, pp. 125–41. See also Richard
Arneson, “Equality and Equality of Opportunity for Welfare,” Philosophical Studies 56
(1989): 77–112.
35. Karl Marx, The German Ideology, with Friedrich Engels, in Karl Marx: Selected
Writings, ed. David McLellan (Oxford: Oxford University Press, 1977), p. 190.
pursue ancestral customs, and it cannot be turned into an argument for

her specifically pursuing those customs in particular, irrespective of the
alternatives she has and the choices she would make.36 The importance
of capability, reflecting opportunities, is central to this distinction.37
I have been concentrating on what the capability perspective can do
for a theory of justice or of human rights in the immediately preceding
discussion, but I now to turn to what it cannot do. Although the idea of
capability has considerable merit in the assessment of the opportunity
aspect of freedom, it cannot possibly deal adequately with the process
aspect of freedom, since capabilities are characteristics of individual
advantages, and they fall short of telling us enough about the fairness or
equity of the processes involved, or about the freedom of citizens to
invoke and utilize procedures that are equitable.
Let me illustrate the contrast of perspectives with a somewhat harsh
example. It is now fairly well established that given symmetric care,
women tend to live longer than men. If one were concerned only with
capabilities (and nothing else), and in particular with equality of the
capability to live long, it would have been possible to construct an argu-
ment for giving men more medical attention than women to counteract
the natural masculine handicap. But giving women less medical atten-
tion than men for the same health problems would clearly violate an
important requirement of process equity, and it seems reasonable to
argue, in cases of this kind, that demands of equity in process freedom
could sensibly override a single-minded concentration on the opportu-
nity aspect of freedom (and the requirements of capability equality in
particular). While it is important to emphasize the relevance of the capa-
bility perspective in judging people’s substantive opportunities (particu-
36. Though this is not the occasion to provide a critical assessment of “multicultural-
ism” as a social policy, it is perhaps worth noting here that there is a big difference between
(1) valuing multiculturalism because of the way, and to the extent that, it enhances the free-
doms of the people involved to choose to live as they would like (and have reason to like);
and (2) valuing cultural diversity per se, which focuses on the descriptive characteristics of
a societal pattern, rather than on the freedoms of the people involved.
37. Capability is also central to the relationship between multiculturalism and gender
equity. The important question that Susan Okin asks in her joint book, Is Multiculturalism
Bad for Women?, ed. J. Cohen, M. Howard and M. C. Nussbaum (Princeton, N.J.: Prince-
ton University Press, 1999), turns, to a great extent, on possible tensions between multi-
culturalism and the freedom of individual persons (in this case, women) within a
community to freely consider and choose how they would live.
larly in comparison with alternative approaches that focus on incomes,

or primary goods, or resources), that point does not, in any way, go
against seeing the simultaneous relevance also of the process aspect of
freedom in a theory of human rights, or, for that matter, in a theory of
justice.
Related to this issue, I should perhaps take the opportunity here to
correct a misinterpretation of the place of the capability perspective in
a theory of justice. A theory of justice, or more generally an adequate
theory of normative collective choice, has to be alive both to the fairness
of the processes involved and to the equity and efficiency of the sub-
stantive opportunities that people can enjoy.38 In dealing with the latter,
capability can indeed provide a very helpful perspective, in comparison
with, say, the Rawlsian concentration on “primary goods.” But capabil-
ity can hardly serve as the sole informational basis for the other consid-
erations, related to processes, that must also be accommodated in
normative collective choice theory.
Perhaps the point can be seen most easily by considering the differ-
ent components of Rawls’s theory of justice. His “first principle” of justice
involves the priority of liberty, and the first part of the “second princi-
ple” involves process fairness, through demanding that “positions and
offices be open to all.” Even though the concerns that lead Rawls to these
particular formulations can be dealt with in different ways, not only in
the way that Rawls himself addresses them, the force and cogency of
these Rawlsian concerns can neither be ignored nor be adequately
addressed through relying only the informational base of capabilities.39
In contrast, capability comes into its own in dealing with the remain-
der of the second principle, viz. “the Difference Principle” (with its con-
38. The plurality of concerns, involving processes as well as opportunities, which is

inescapably involved in normative collective choice (including theories of justice), is dis-
cussed in my Collective Choice and Social Welfare (1970) and “Well-being, Agency and
Freedom: The Dewey Lectures 1984,” Journal of Philosophy 82 (1985). Since I have seen it
asserted that I propound a “capability-based theory of justice,” I should make it absolutely
clear that this could be true only in the rather limited sense of naming something accord-
ing to a principal part of it (comparable with, say, using England for Great Britain, or
Holland for the Netherlands).
39. See my Collective Choice and Social Welfare, particularly chapters 5 through 9, and
“Well-being, Agency and Freedom: The Dewey Lectures 1984.” The issues involved are more
fully addressed in my forthcoming book Freedom and Justice, to be published by Harvard
University Press.
centration on “primary goods”).40 The territory that Rawls reserved for

the accounting of primary goods, as used in his Difference Principle,
would indeed be, I argue, better served by the capability perspective.
That does not, however, obliterate in any way the relevance of the rest of
the territory of justice, in which process considerations, including liberty
and procedural equity, figure. The same plurality of informational base
links with the multiplicity of considerations that can be invoked in a
theory of human rights. Capabilities and the opportunity aspect of
freedom, important as they are, have to be supplemented by considera-
tions of fair processes and the lack of violation of the individual’s right
to invoke and utilize them.
VI. Duties, Reasonable Consideration and Imperfect Obligations

I turn now from rights to correlative duties. We can, again, proceed from
the importance of freedoms and their different aspects. Since freedoms
are important, people have reason to ask what they should do to help
each other in defending or promoting their respective freedoms. Since
violation, or non-realization, of the freedoms underlying significant
rights are, in this evaluative system, bad things to happen, even others
who are not themselves responsible for causing the violation have a good
reason to consider what they should do to help.41 Nevertheless, the move
from a reason for action to help another person, which is easy to see in
a consequence-sensitive ethical system, to an actual duty to give
40. It was indeed in the context of identifying an inadequacy in the Rawlsian focus on
primary goods in the Difference Principle, for judging distributional equity, that the use of
the capability perspective was proposed in my 1979 Tanner Lectures, published as “Equal-
ity of What?” (1980). In judging distributional equity, the capability perspective also has, I
believe, advantages over the concentration on what Ronald Dworkin calls “resources” in
“What Is Equality? Part 2: Equality of Resources,” Philosophy & Public Affairs 10 (1981):
185–243. Dworkin has recently argued that on one interpretation, there is no substantial
difference between my focus on capability and his focus on resources, while on another
interpretation, he is just right and I am plain wrong (Sovereign Virtue: The Theory and Prac-
tice of Equality [Cambridge, Mass.: Harvard University Press, 2000]). I resist the tempta-
tion, which I must confess is fairly strong, to join that debate in this article.
41. The rationale and reach of a consequence-sensitive framework for this type of
ethical reasoning have been investigated in my essays “Rights and Agency,” Philosophy &
Public Affairs 11 (1982): 3–39, “Positional Objectivity,” Philosophy & Public Affairs 22 (1993):
126–45, and “Consequential Evaluation and Practical Reason,” Journal of Philosophy 97
(2000).
reasonable consideration to undertaking such an action might appear,

at least at first sight, to be a rather gigantic jump.
However, that sense of distance is largely illusory. The difference
would indeed involve an immense escalation if the duty in question were
not one of giving reasonable consideration to a possible action, but an
absolute obligation to undertake that action, no matter what other
values one has and what other commitments one has reason to consider.
But that way of seeing one’s duties—as compulsory action—is not only
at some distance from the acknowledgment of reasons for action, but it
also lacks cogency, and even internal coherence. There are many fine
deeds for each of which a reason for action exists, but it would typically
be impossible to carry out the totality of all those deeds. There is a need
for the assessment of priorities and for discrimination in the way the
obligation to give reasonable consideration may be followed up by
sensible choices of action.
To accept that one has a duty to give reasonable consideration to
many different types of actions is not an agreement to tie oneself up in
hopeless knots. And it is particularly important in the present context to
emphasize the converse: the determination not to get into a pandemo-
nium of practical reasoning is not a ground for denying that one does
have a duty to give reasonable consideration to what one can sensibly
do for the rights, and the underlying significant and influenceable free-
doms, of others. The demands of reasonable consideration would vary
with a great many parameters that may be relevant to a person’s practi-
cal reasoning.42 Even though the acknowledgment that certain freedoms
qualify as human rights already reflects an assessment of their general
importance and their possible influenceability (discussed in Section IV),
a person has to go beyond these pervasive features into more specific
circumstances in giving reasonable consideration to what he or she, in
particular, should do in a specific case.
The person has to judge, for example, how important the freedoms
and rights are in the case in question compared with other claims on the
person’s possible actions (involving other rights and freedoms, but also
42. Making adequate room for parametric variations is a general feature of rational
assessment, and not a characteristic only of ethical reasoning in particular. I have dis-
cussed this issue in Rationality and Freedom (Cambridge, Mass.: Harvard University Press,
2002), essays 1 through 5.
altogether different concerns that a person may, inter alia, sensibly

have). Furthermore, the person has to judge the extent to which he or
she can make a difference in this case, either acting alone or in con-
junction with others. It will be relevant also to consider what others can
be expected to do, and the appropriateness of how the required sup-
portive actions may be shared among possible agents. A great many
parametric considerations of these and other kinds will inescapably
figure in the reasoned evaluation of what a person should do, even after
the need to undertake such an evaluation has been fully accepted. Also,
since detailed reflection on what one should do is itself time consuming
(and cannot even be actually undertaken for all the ills of the world), the
duty of reasonable consideration will not, in a great many cases, trans-
late into an obligation to take on an elaborate scrutiny—only a willing-
ness to do just that, when it seems relevant and appropriate.
The recognition of obligations in relation to the rights and freedoms
of all human beings need not, thus, be translated into preposterously
demanding commands. And yet, despite the parametric variability of
the reach and force of reasonable consideration, the requirement to
give such consideration is not by any means vacuous. The basic general
obligation is that one must be willing to consider seriously what one
should reasonably do, taking note of the relevant parameters of the cases
involved. The necessity to ask that question (rather than proceeding on
the assumption that we owe nothing to others, unless we have actually
harmed them) can be the beginning of a more comprehensive line of
ethical reasoning.43 The territory of human rights firmly belongs there.
The reasoning cannot, however, end there. Given one’s limited abilities
and reach, and the need for priorities involving different types of oblig-
ations as well as the demands of other moral concerns, there are serious
exercises of practical reasoning to be undertaken, in which one’s various
obligations (including “imperfect obligations”) must figure, in an explicit
or implicit form.
The recognition of human rights is not an insistence that everyone
everywhere rises to help prevent every violation of every human right no
matter where it occurs. It is, rather, an acknowledgment that if one is in
a plausible position to do something effective in preventing the violation
43. The centrality of that general question is powerfully discussed by Thomas Scanlon,
What We Owe to Each Other (Cambridge, Mass.: Harvard University Press, 1998).
of such a right, then one does have an obligation to consider doing just
that. It is still possible that other obligations or non-obligational con-
cerns may overwhelm the reason for the particular action in question,
but that reason cannot be simply brushed away as being “none of one’s
business.” Loosely specified obligations must not be confused with no
obligations at all. Rather, they belong, as was mentioned earlier, to the
important category of duties that Immanuel Kant called “imperfect
obligations” (and to which he attached great importance).
It is to be noted that, in this understanding, imperfect obligations are
ethical requirements that stretch beyond the fully delineated duties, “the
perfect obligations,” that specific persons may have to perform particu-
lar acts. They involve the demand that serious consideration be given by
anyone in a position to provide reasonable help to the person whose
human right is threatened. These “imperfect obligations” firmly corre-
late, in the same way as fully specified “perfect obligations” do, with
the recognition of rights. The difference lies in the nature and form of
the obligations, not in the general correspondence between rights and
obligations, which apply in the same way to imperfect as well as perfect
obligations.
It may be useful to illustrate, with a concrete example, the distinction
between different kinds of obligations that, despite their differences in
content, relate in a similar way to human rights. Consider a real-life case
that occurred in Queens, New York, in 1964, when a woman, Kitty Gen-
ovese, was fatally assaulted in full view of many others watching the
event from their apartments, who did nothing to help her. It is plausible
to argue that three terrible things happened here, which are distinct but
interrelated:
(1) the woman’s freedom—and right—not to be assaulted and killed
was violated (this is clearly the principal nastiness in this case);
(2) the murderer violated the immunity that anyone should have
against assault and killing (a violation of a “perfect obligation”); and
(3) the others who did nothing whatever to help the victim also
transgressed their general—and “imperfect”—obligation to seriously
consider providing the help which they could reasonably be expected to
provide.
These distinct failings bring out a complex pattern of rights–duties
correspondence in a structured ethics, which can help to explicate the
evaluative framework of human rights, which yields imperfect as well as

perfect obligations.44
The presumed precision of legal rights is often contrasted with
inescapable ambiguities in the ethical claims of human rights. This con-
trast, however, is not in itself a great embarrassment for ethical claims,
including those of imperfect obligations, since a framework of norma-
tive reasoning can sensibly allow variations that cannot be easily accom-
modated in fully specified legal requirements. As Aristotle remarked in
the Nicomachean Ethics, we have “to look for precision in each class of
things just so far as the nature of the subject admits.”45
As it happens, however, in the laws of some countries, there is even a
legal demand, which can hardly have extreme precision, for providing
reasonable help to third parties. For example, in France there is provi-
sion for “criminal liability of omissions” in the failure to provide reason-
able help to others suffering from particular types of transgressions. Not
surprisingly, ambiguities in the application of such laws have proved to
be quite large and have been the subject of considerable legal discussion
in recent years.46 The ambiguity of duties of this type, whether in ethics
or in law, would be difficult to escape if third-party obligations of others
in general are given some room, and this cannot be avoided for an
adequate theory of human rights.
VII. Recognition, Agitation and Legislation

While the preceding analysis has been concerned with giving reasonable
consideration to actions in general that people can undertake in defend-
ing or advancing the human rights of others, it is the legislation of
human rights, along with their institutionalization, that has tended to
44. In this analysis I do not go into the distinction between agent-specific and agent-
neutral moral evaluations. The present line of characterization can be further extended
through making room for “position specific” assessments, in ways that I have tried to inves-
tigate in “Rights and Agency,” and “Positional Objectivity.”
45. The admissibility of inescapable ambiguities within a framework of rational assess-
ment is discussed in my “Internal Consistency of Choice,” Econometrica 61 (1993): 495–521,
and “Maximization and the Act of Choice,” Econometrica 65 (1997): 745–79, both reprinted
in Rationality and Freedom. See also Inequality Reexamined, pp. 46–49, 131–35.
46. See, for example, Andrew Ashworth and Eva Steiner, “Criminal Omissions and
Public Duties: The French Experience,” Legal Studies 10 (1990): 153–64; Glanville Williams,
“Criminal Omissions: The Conventional View,” Law Quarterly Review 107 (1991): 86–98.
receive the lion’s share of attention in the theoretical literature in this

field. It is this legislative outlook that has also been firmly incorporated
in much of the institutional understanding of human rights. However,
while legislation is an important domain of public action, there are
other ways and means which are also important and often effective in
advancing the cause of recognized human rights.
First, under what can be called the “recognition route” (to be distin-
guished from the “legislative route”), there is acknowledgment but not
necessarily any legalization or institutional enforcement of a class of
claims that are seen as fundamental human rights.47 The Universal
Declaration of Human Rights, sponsored by the United Nations in 1948,
which was perhaps the most important move that promoted global
activities on human rights in the last century, falls solidly into this cate-
gory (even though, as was discussed earlier, the framers of the Declara-
tion had also hoped that it would lead to specific bills of rights in
different countries). Subsequently, there has been a sequence of other
international declarations, often through the United Nations, giving
recognition, rather than a legal and coercive status, to various general
demands, for example the “Declaration on the Right to Development,”
signed in 1986.48 This approach is motivated by the idea that the ethical
force of human rights is made more powerful in practice through giving
it social recognition and an acknowledged status, even when no enforce-
ment is instituted.
A second line of advance goes beyond recognition to active agitation.
There can be organized advocacy urging compliance with certain basic
claims of all human beings that are seen as human rights, and there can
47. As Charles Beitz has pointed out, human rights play “the role of a moral touch-
stone—a standard of assessment and criticism for domestic institutions, a standard of
aspiration for their reform, and increasingly a standard of evaluation for the policies and
practices of international economic and political organizations.” See “Human Rights as a
Common Concern,” p. 269.
48. Analyses of the content of the right to development have been presented in United
Nations Development Programme, Human Development Report  (New York: United
Nations, 2000); S. R. Osmani, “Human Rights to Food, Health, and Education,” mimeo-
graphed, UNDP and the University of Ulster, 2000; Arjun Sengupta, “Development Policy
and the Right to Development,” Frontline, February 7–March 2, 2001; Arjun Sengupta,
Asborn Eide, Stephen Marks, and Bård Anders Andreassen, “The Right to Development and
Human Rights in Development,” presented at the Nobel Symposium in Oslo on Right to
Development, October 2003.
also be monitoring of violations of these rights and attempts to gener-

ate effective social pressure. The global NGOs have increasingly been
involved in advancing human rights, through public discussion and
support, on the one hand, and publicizing and criticizing violations, on
the other. These efforts have come not only from dedicated human rights
organizations, such as Human Rights Watch and Amnesty International,
but also from broader organizations, such as OXFAM, Médicins Sans
Frontièrs, the Red Cross, Save the Children, and Action Aid. The rights
invoked in this “agitation route” may or may not have any legal status in
the country in question, but advocacy and support are not necessarily
rendered useless by the absence of legal backing.49 Furthermore, even
when some identified human rights have legal status, good enforcement
of the relevant legislation may also call for public activism, which is to
be distinguished from the process of legislation itself.
The third approach is, of course, that of “legislation.” As was discussed
in Section III, even though the ethics of human rights must not be seen
merely as “parents” of “human rights laws,” it is certainly the case that
many such legislations have been encouraged or inspired by considera-
tions of human rights. Many actual laws have been enacted by individ-
ual states, or by associations of states, which gave legal force to certain
rights seen as basic human rights. For example, the European Court of
Human Rights, established in 1950 following the European Convention,
can consider cases brought by individuals from the signatory states
against violations of human rights. This has been supplemented by the
Human Rights Act of 1998, aimed at incorporating the main provisions
of the European Convention into domestic law, with an overseeing role
of the European Court to see “just satisfaction” of these provisions in
49. It is also worth noting that even when the agents involved in activist promotion of
human rights do not have any special legal status, they can still make a difference to polit-
ical, social and administrative practice through the use of existing laws, combined with
seeking public disclosures and critical debates. For example, unlike the Indian and South
African Human Rights Commissions, which are recognized in the respective national laws,
the Pakistan Human Rights Commission is basically just an NGO, and yet under the vision-
ary and courageous leadership of Asma Jahangir, I. A. Rehman, and others, it has been
remarkably effective in identifying and resisting violations of human rights, and in defend-
ing vulnerable persons, including religious minorities and ill-treated women. For a good
discussion of some of these supportive activities, see The State of Human Rights, 
(Lahore: Human Rights Commission of Pakistan, 2002).
domestic judgments. Many other examples can be given from different

parts of the world. The “legislative route” has had much active use.
There is an interesting question about the appropriate domain of the
legislative route. It would be a mistake, I would argue, to presume in
general that if a human right is important, then it must be ideal to leg-
islate it into a precisely specified legal right. For example, recognizing
and defending a wife’s moral right to be consulted in family decisions,
even in a traditionally sexist society, may well be extremely important,
and can plausibly satisfy the threshold conditions needed to qualify as
a human right.50 And yet the advocates of this human right, who empha-
size, correctly, its far-reaching ethical and political relevance, can quite
possibly agree that it is not sensible to make this human right into, in
Herbert Hart’s language, a “coercive legal rule” (perhaps with the result
that a husband would be taken in custody if he were to fail to consult his
wife). The necessary change would have to be brought about in other
ways. Because of the importance of communication, advocacy, exposure
and informed public discussion, human rights can have influence
without necessarily depending on coercive legislation.
Similarly, the moral or political entitlement, which can easily be seen
as a human right, of a somewhat slow speaker not to be snubbed in an
open public meeting by a rudely articulate sprinter may well be impor-
tant both for the self-respect of the leisurely speaker and for public good,
but it is not likely to be a good subject for punitive legislation. The pro-
tection of that human right would have to be sought elsewhere. The
effectiveness of the human rights perspective does not rest on seeing
them invariably as putative proposals for legislation.51
VIII. Economic and Social Rights

I turn now to criticisms that have been particularly aimed against
extending the idea of human rights to include economic and social
rights, such as the right not to be hungry, or the right to basic education
or to medical attention. Even though these rights did not figure in the
50. The importance and social reach of woman’s participation in family decisions is dis-
cussed in my Development as Freedom, ch. 8, “Women’s Agency and Social Change.”
51. For an early advocacy of a much broader approach, see Mary Wollstonecraft, A Vin-
dication of the Rights of Woman: with Strictures on Political and Moral Subjects (1792).
classic presentations of rights of human beings in, say, the U.S. Declara-
tion of Independence, or French “rights of man,” they are very much a
part of the contemporary domain of what Cass Sunstein calls the “rights
revolution.”52 The legitimacy of including these claims within the general
class of human rights has been challenged through two specific lines of
reproach, which I shall call, respectively, the institutionalization critique
and the feasibility critique.
The institutionalization critique, which is aimed particularly at eco-
nomic and social rights, relates to the general issue of the exact corre-
spondence between authentic rights and precisely formulated correlate
duties. Such a correspondence, it is argued, would exist only when a right
is institutionalized. Onora O’Neill has presented this line of criticism
with force:
Unfortunately much writing and rhetoric on rights heedlessly pro-

claims universal rights to goods and services, and in particular
“welfare rights,” as well as to other social, economic and cultural
rights that are prominent in international Charters and Declarations,
without showing what connects each presumed right-holder to some
specific obligation-bearer(s), which leaves the content of these sup-
posed rights wholly obscure. . . . Some advocates of universal eco-
nomic, social and cultural rights go no further than to emphasize that
they can be institutionalized, which is true. But the point of difference
is that they must be institutionalized: if they are not there is no right.53
In responding to this significant criticism, we have to invoke the

understanding, already discussed, that obligations can be both perfect
and imperfect. Even the classical “first generational” rights, like freedom
from assault, can be seen as yielding imperfect obligations on others, as
was illustrated with the example of the case of assault on Kitty Genovese
in public view in New York. Depending on institutional possibilities, eco-
nomic and social rights may similarly call for both perfect and imperfect
obligations. There is a large area of fruitful public discussion and possi-
bly effective pressure, concerning what the society and the state, even
52. Sunstein, After the Rights Revolution: Reconceiving the Regulatory State.
53. Onora O’Neill, Towards Justice and Virtue, pp. 131–32. See also her Bounds of Justice
(Cambridge: Cambridge University Press, 2000).
an impoverished one, can do to prevent violations of certain basic eco-

nomic or social rights (associated with, say, the prevalence of famines,
or chronic undernourishment, or absence of medical care).
Indeed, the supportive activities of social organizations are often
aimed precisely at institutional change, and these activities can be seen
as part of imperfect obligations that individuals and groups have in a
society where basic human rights are violated. Onora O’Neill is right to
emphasize the importance of institutions for the realization of “welfare
rights” (and even for economic and social rights in general), but the
ethical significance of these rights provide good grounds for seeking
realization through institutional expansion and reform. This can be
helped through a variety of approaches, including demanding and agi-
tating for appropriate legislation, and the supplementation of legal
demands by political recognition and social monitoring. To deny the
ethical status of these claims would be to ignore the reasoning that moti-
vates these constructive activities.
The feasibility critique proceeds from the argument that even with the
best of efforts, it may not be feasible to arrange the realization of many
of the alleged economic and social rights for all. This would have been
only an empirical observation (of some interest of its own), but it is made
into an allegedly powerful criticism of the acceptance of these claimed
rights on the basis of the presumption, largely undefended, that recog-
nized human rights must, of necessity, be wholly accomplishable. If this
presumption were accepted that would have the effect of immediately
putting many so-called economic and social rights outside the domain
of possible human rights, especially in the poorer societies.
Maurice Cranston puts the argument thus:
The traditional political and civil rights are not difficult to institute.
For the most part, they require governments, and other people gen-
erally, to leave a man alone. . . . The problems posed by claims to eco-
nomic and social rights, however, are of another order altogether. How
can governments of those parts of Asia, Africa, and South America,
where industrialization has hardly begun, be reasonably called upon
to provide social security and holidays with pay for millions of people
who inhabit those places and multiply so swiftly?54
54. Cranston, “Are There Any Human Rights?” p. 13.

In assessing this line of rejection, we have to ask: why should complete

feasibility be a condition of cogency of human rights when the objective
is to work towards enhancing their actual realization, if necessary
through expanding their feasibility? The understanding that some rights
are not fully realized, and may not even be fully realizable under present
circumstances, does not, in itself, entail anything like the conclusion that
these are, therefore, not rights at all.55 Rather, that understanding sug-
gests the need to work towards changing the prevailing circumstances
to make the unrealized rights realizable, and ultimately, realized.56
It is also worth noting in this context that the question of feasibility is
not confined to economic and social rights only; it is a much more wide-
spread problem. Even for liberties and autonomies, to guarantee that a
person is “left alone,” which Cranston seems to think is simple to guar-
antee, has never been particularly easy. That elementary fact, easily seen
always, cannot but be rather clearly recognized now, at least since Sep-
tember 11, 2001 (and more recent events). If the current feasibility of
guaranteeing complete and comprehensive fulfillment were made into
a necessary condition for the cogency of every right, then not only eco-
nomic and social rights, but also liberties, autonomies and even politi-
cal rights may well fall far short of cogency.
IX. The Reach of Public Reasoning

How can we judge the acceptability of claims to human rights and assess
the challenges they may face? How would such a disputation—or a
defense—proceed? I would argue that like the assessment of other
ethical claims, there must be some test of open and informed scrutiny,
55. For this reason, I would argue, it would be a misapplication to invoke the familiar
principle “ought implies can” to suggest that claims that are not yet fully realizable cannot
be taken to be rights at all. To see the ethical force of some claims is also a demand to con-
sider what one should do to make them realizable, for example through working for the
development of new institutions.
56. This corresponds to what Charles Beitz calls the “practical conception” of human
rights: “To say something is a human right is to say that social institutions that fail to
protect the right are defective” with the implication that “international efforts to aid or
promote reform are legitimate and in some cases may be morally required.” (“Human
Rights and the Law of Peoples,” in The Ethics of Assistance: Morality and the Distant Needy
ed. Deen Chatterjee [Cambridge: Cambridge University Press, 2004], p. 210. See also Henry
Shue, Basic Rights: Subsistence, Affluence, and U.S. Foreign Policy [Princeton: Princeton
University Press, 1980; 2nd ed., 1996].)
and it is to such a scrutiny that we have to look in order to proceed to a

disavowal or an affirmation. The status of these ethical claims must be
dependent ultimately on their survivability in unobstructed discus-
sion.57 In this sense, the viability of human rights is linked with what John
Rawls has called “public reasoning” and its role in “ethical objectivity.”58
Indeed, the connection between public reasoning and the formula-
tion and use of human rights is extremely important to understand. Any
general plausibility that these ethical claims, or their denials, have is
dependent, on this theory, on their survival and flourishing when they
encounter unobstructed discussion and scrutiny, along with adequately
wide informational availability. The force of a claim for a human right
would be seriously undermined if it were possible to show that they are
unlikely to survive open public scrutiny. But contrary to a commonly
offered reason for skepticism and rejection, the case for human rights
cannot be discarded simply by pointing to the fact (even when that is the
case) that in politically and socially repressive regimes, which do not
allow open public discussion, many of these human rights are not taken
seriously at all. Uncurbed critical scrutiny is essential for dismissal as
well as for defense. Even as far as use is concerned, the fact that moni-
toring of violations of human rights and the procedure of “naming and
shaming” can be so effective (at least in putting the violators on the
defensive) is some indication of the reach of public reasoning when
information becomes available and ethical arguments are allowed rather
than suppressed.
However, it is important not to confine the domain of public reason-
ing to a given society only, especially in the case of human rights, in view
of the inescapably non-parochial nature of these rights, which are meant
to apply to all human beings. This is in contrast with Rawls’s inclination,
57. Even though this requirement has a largely procedural form, the very insistence on
open public discussion from which no one is excluded involves an acceptance of equality,
which has substantive implications also for the content of the deliberation. On the sub-
stantive aspects of deliberative democracy, see Joshua Cohen, “Procedure and Substance
in Deliberative Democracy,” in Democracy and Difference: Contesting the Boundaries of the
Political, ed. Seyla Benhabib (Princeton: Princeton University Press, 1996), pp. 95–119.
58. John Rawls, A Theory of Justice (Cambridge, Mass.: Harvard University Press, 1971),
and Political Liberalism (New York: Columbia University Press, 1993), esp. pp. 110–13. On
related matters, see also Amy Guttman and Dennis Thompson, Democracy and Disagree-
ment (Cambridge, Mass.: Harvard University Press, 1996), and Democracy and Difference,
ed. Seyla Benhabib.
particularly in his later works, to limit such public confrontation within

the boundaries of each particular nation (or each “people,” as Rawls calls
this regional collectivity), for determining what would be just, at least in
domestic affairs.59 We can demand, on the contrary, that the discussion
include, even for domestic justice (if only to avoid parochial prejudices
and also to examine a broader range of counterarguments), views also
from “a certain distance.” The necessity of this was powerfully identified
by Adam Smith:
We can never survey our own sentiments and motives, we can never
form any judgment concerning them; unless we remove ourselves, as
it were, from our own natural station, and endeavour to view them as
at a certain distance from us. But we can do this in no other way than
by endeavouring to view them with the eyes of other people, or as
other people are likely to view them.60
The universalist nature of Adam Smith’s approach raises the question

whether distant people can, in fact, provide useful scrutiny of local
issues, given what are taken to be “uncrossable” barriers of culture. One
of Edmund Burke’s criticisms of the French declaration of the “rights of
man” and its universalist spirit was concerned with disputing the accept-
ability of that notion in other cultures. Burke argued that “the liberties
and the restrictions vary with times and circumstances, and admit of
infinite modifications, that cannot be settled upon any abstract rule.”61
The argument that, for this, or some similar, reason, the universality that
underlies the notion of human rights is profoundly mistaken can be
found in many other writings as well.
59. See particularly John Rawls, The Law of Peoples (Cambridge, Mass.: Harvard Uni-
versity Press, 1999). See also Rawls’s formulation of the original position in Political Liber-
alism, p. 12: “I assume that the basic structure is that of a closed society: that is, we are to
regard it as self-contained and as having no relations with other societies. . . . That a society
is closed is a considerable abstraction, justified only because it enables us to focus on
certain main questions free from distracting details.” If my reasoning is right, the Rawlsian
restrictions eliminate much more than the influence of “distracting details.”
60. Adam Smith, The Theory of Moral Sentiments (1759; rev. ed., 1790; republished,
Oxford: Clarendon Press, 1976), III, 1, 2, p. 110. The Smithian perspective on moral reason-
ing is pursued in my “Open and Closed Impartiality,” The Journal of Philosophy 99 (2002):
445–69.
61. Quoted in Steven Lukes, “Five Fables about Human Rights,” in The Human Rights
Reader, ed. Micheline R. Ishay (London: Routledge, 1997), p. 238.
For example, Rosa Luxemburg, a leading Marxist thinker and political

leader in the early twentieth century, invoked a similar line of criticism
of what she called the “metaphysical cliché of the type of ‘rights of man’
and ‘rights of the citizen’.”62 However, a scrutiny of Rosa Luxemburg’s real
concerns brings out the remarkable fact that she persistently invoked
universalist principles herself, as is quite standard in the Marxist tradi-
tion (consider: “from each according to his ability, to each according to
his needs”). Rather, Luxemburg was keen on emphasizing that the sub-
stantiation of these principles must depend on specific circumstances
that obtain. Shorn of the rhetoric, there is, in fact, no particular difficulty
in using basic universalist principles in general, while taking note of Lux-
emburg’s pointer to the relevance of local circumstances and regional
conditions in appropriately contingent, or parametric, specification of
the exact demands of human rights.
However, a belief in uncrossable barriers between the values of dif-
ferent cultures has surfaced and resurfaced repeatedly over the cen-
turies, and they are forcefully articulated today. The claim of magnificent
uniqueness, and often of superiority, has sometimes come from critics
of “Western values,” varying from champions of regional ethics (well
illustrated by the fuss in the 1990s about the peerless excellence of “Asian
values”), or religious or cultural separatists (with or without being
accompanied by fundamentalism of one kind or another). Sometimes,
however, the claim of uniqueness has come from Western particularists.
A good example is Samuel Huntington’s insistence that the “West was
West long before it was modern,” and his claim that “a sense of individ-
ualism and a tradition of individual rights and liberties” are “unique
among civilized societies.”63 Similarly, no less a historian of ideas than
Gertrude Himmelfarb has argued that ideas of “justice,” “right,” “reason”
and “love of humanity” are “predominantly, perhaps even uniquely,
Western values.”64
62. Rosa Luxemburg, “The National Question and Autonomy,” The Human Rights
Reader, p. 291.
63. Samuel P. Huntington, The Clash of Civilizations and the Remaking of World Order
(New York: Simon and Schuster, 1996).
64. Gertrude Himmelfarb, “The Illusions of Cosmopolitanism,” in For Love of Country:
Debating the Limits of Patriotism / Martha Nussbaum with Respondents, ed. Joshua Cohen
(Boston: Beacon Press, 1996), pp. 74–75.
I have discussed these diagnoses elsewhere.65 Contrary to cultural

stereotypes, the histories of different countries in the world have shown
considerable variations over time as well as between different traditions
within the same country. The championing of open public discussion,
tolerating and encouraging different points of view, has a long history in
many countries in the world. Indeed, some of the earliest open general
meetings aimed specifically at settling disputes between different points
of view took place in India in the so-called Buddhist councils, the first of
which was held shortly after Gautama Buddha’s death twenty-five
hundred years ago. The grandest of these councils, the third, occurred
under the patronage of Emperor Ashoka in the third century bce. Ashoka
also tried to codify and propagate what must have been among the ear-
liest formulations of rules for public discussion, a kind of ancient version
of the nineteenth-century Robert’s Rules of Order. He demanded, for
example, “restraint in regard to speech, so that there should be no extol-
ment of one’s own sect or disparagement of other sects on inappropri-
ate occasions, and it should be moderate even in appropriate occasions.”
Even when engaged in arguing, “other sects should be duly honoured in
every way on all occasions.”
To consider another historical example, in early seventh-century
Japan, the Buddhist Prince Shotoku, who was regent to his mother,
Empress Suiko, produced the so-called constitution of seventeen arti-
cles, in 604 ad. The constitution insisted, much in the spirit of the Magna
Carta to be signed six centuries later in 1215 ad: “Decisions on important
matters should not be made by one person alone. They should be dis-
cussed with many.”
When, in the twelfth century, the Jewish philosopher Maimonedes
had to flee an intolerant Europe to try to safeguard his human right to
stick to his own religious beliefs and practice, he sought shelter in
Emperor Saladin’s Egypt (via Fez and Palestine), and found an honored
position in the court of this Muslim emperor. Several hundred years
later, when, in Agra, the Moghul emperor of India, Akbar, was arguing,
and legislating, on the government’s duty to uphold the right to religious
65. See my Development as Freedom, ch. 10. Also “Human Rights and Asian Values,” The
New Republic, July 14 and 21, 1997, pp. 33–40; “The Reach of Reason: East and West,”
The New York Review of Books, July 20, 2000, pp. 33–38; “Democracy and Its Global Roots,”
The New Republic, October, 2003, pp. 28–35.
freedom of all citizens, the European Inquisitions were still going on, and
Giardino Bruno was burnt at the stake in Rome, in 1600.
In his autobiography, Long Walk to Freedom, Nelson Mandela
describes how he learned about democracy and individual rights, as a
young boy, by seeing the proceedings of the local meetings held in the
regent’s house in Mqhekezweni:
Everyone who wanted to speak did so. It was democracy in its purest
form. There may have been a hierarchy of importance among the
speakers, but everyone was heard, chief and subject, warrior and
medicine man, shopkeeper and farmer, landowner and laborer.66
Not only are the differences on the subject of freedoms and rights that
actually exist between different societies often much exaggerated, but
also there is, typically, little note taken of substantial variations within
each local culture—over time and even at a point of time (in particular,
right now). What are taken to be “foreign” criticisms often correspond to
internal criticisms from non-mainstream groups. If, say, Iranian dissi-
dents are imprisoned by an authoritarian regime precisely because of
their heterodoxy, any suggestion that they should be seen as “ambas-
sadors of Western values” rather than as “Iranian dissidents” would only
add serious insult to manifest injury.
This issue is particularly important in determining what may be taken
to be culturally “partisan” in a world of many cultural differences.
Charles Beitz rejects, rightly, the plausibility of seeing the use of human
rights as emanating from a “supposedly symmetrical relationship to the
conception of political justice or legitimacy to be found in the world’s
cultures,” and he goes on to seek their justification in terms of “the role
they play in international relations.”67 But how should this “role” be
judged in terms of its acceptability, and in what sense should such an
evaluation be culturally “partisan”? If the reasoning presented here is
right, then we must distinguish between (1) the values that are domi-
nantly favored in a society (no matter how repressive it is), and (2) the
values that could be expected to gain wider adherence and support
when open discussion is allowed, when information about other
66. Nelson Mandela, Long Walk to Freedom (Boston: Little, Brown & Co, 1994), p. 21.
67. Beitz, “Human Rights as a Common Concern,” pp. 279–80.
societies becomes more freely available, and when disagreements with

the established views can be expressed and defended without suppres-
sion and fear.
Being “non-partisan” requires respecting the participation of people
from any corner of the earth, which is not the same thing as accepting
the prevailing priorities in existing societies when information is
extremely restricted and discussions and disagreements are not per-
mitted. Widespread acceptability, which must be distinguished from
pre-existing ubiquitous acceptance, is an important issue in any social
evaluation, even in dealing with the role that human rights play in inter-
national relations.
There does, of course, exist considerable variation in the balance of
manifest opinions and observed preconceptions in different countries
and different societies. These opinions and beliefs often reflect, as Adam
Smith noted in a powerfully illuminating analysis, strong influence of
existing practices in different parts of the world, along with a lack of
broader intellectual engagement. The need for open scrutiny, with unre-
strained access to information (including that about practices elsewhere
in the world and the experiences there), is particularly great because of
these connections. Which is precisely why Adam Smith’s insistence on
the necessity of viewing actions and practices from a “certain distance”
is so important for substantive ethics in general and the understanding
of human rights in particular.
In a chapter entitled “On the Influence of Custom and Fashion upon
the Sentiments of Moral Approbation and Disapprobation,” Smith illus-
trated his contention:
. . . the murder of new-born infants was a practice allowed of in

almost all the states of Greece, even among the polite and civilized
Athenians; and whenever the circumstances of the parent rendered it
inconvenient to bring up the child, to abandon it to hunger, or to wild
beasts, was regarded without blame or censure. . . . Uninterrupted
custom had by this time so thoroughly authorized the practice, that
not only the loose maxims of the world tolerated this barbarous pre-
rogative, but even the doctrine of philosophers, which ought to have
been more just and accurate, was led away by the established custom,
and upon this, as upon many other occasions, instead of censuring,
supported the horrible abuse, by far-fetched considerations of public
utility. Aristotle talks of it as of what the magistrates ought upon many
occasions to encourage. Plato is of the same opinion, and, with all that
love of mankind which seems to animate all his writings, no where
marks this practice with disapprobation.68
What are taken to be perfectly “normal” and “sensible” in an insulated

society may not be able to survive a broad-based and less limited exam-
ination once the parochial gut reactions are replaced by critical scrutiny,
including an awareness of variations of practices and norms across the
world.69
Scrutiny from a distance may have something to offer in the assess-
ment of practices as different from each other as the stoning of adulter-
ous women in Taliban’s Afghanistan and the abounding use of capital
punishment (sometimes with mass jubilation) in parts of the United
States. This is the kind of issue that made Smith insist that “the eyes of
the rest of mankind” must be invoked to understand whether “a pun-
ishment appears equitable.”70 Ultimately, the discipline of critical moral
scrutiny requires, among other things, “endeavouring to view [our sen-
timents and beliefs] with the eyes of other people, or as other people are
likely to view them.”71
The need for interactions across the borders can be as important in rich
societies as they are in poorer ones.72 The point to note here is not so much
whether we are permitted to make cross-boundary scrutiny, but that the
discipline of critical assessment of moral sentiments, no matter how
locally established they are, demands that such scrutiny be undertaken.
X. A Concluding Remark
I have tried to present, in this article, the elements of a theory of human
rights, which sees them as pronouncements in social ethics, sustainable
68. Adam Smith, The Theory of Moral Sentiments (rev. ed., 1790, V.2.15; republished,
Oxford: Clarendon Press, 1976), p. 210.
69. I have discussed this issue in “Open and Closed Impartiality,” Journal of Philosophy
99 (2002): 445–69.
70. Adam Smith, Lectures on Jurisprudence, ed. R. L. Meek, D. D. Raphael and P. G. Stein
(Oxford: Clarendon Press, 1978; reprinted, Indianapolis: Liberty Press, 1982), p. 104.
71. Smith, The Theory of Moral Sentiments, III, 1, 2, p. 110.
72. The treatment of prisoners held by the United States in the so-called war against
terrorism raises important issues of human rights, and the analysis of the prevailing prac-
tice can be helped by more wide-ranging public discussion and a fuller understanding of
the nature of global concerns on this issue.
by open public reasoning. They may or may not be reflected in a legal

framework through, say, specific “human rights legislation,” but there
are also other ways of implementing human rights (including public
recognition, agitation and monitoring).
Since the main themes developed in this article were specifically
noted in Section II, I shall not try to provide a further summary in this
concluding section. I should, however, emphasize that the understand-
ing and viability of human rights are, in this perspective, intimately
linked with the reach of public discussion, between persons and across
borders. The viability and universality of human rights are dependent on
their ability to survive open critical scrutiny in public reasoning. The
methodology of public scrutiny draws on Rawlsian understanding of
“objectivity” in ethics, but the impartiality that is needed cannot be con-
fined within the borders of a nation.
The fact that authoritarian orders are typically quite afraid of uncen-
sored news media and of uncurbed public discussion, which make them
resort often enough to suppression (including censorship, intimidation,
incarceration, and even execution), provides some indirect evidence
that the influence of public reasoning can indeed be quite large. That
influence also lies behind the effectiveness of the interactive ways and
means, including social recognition, informational monitoring and
public agitation, which human rights activists tend to use. There is cer-
tainly a need for a fuller understanding of the associative nature of the
acceptability of values, and this requires us to go well beyond lazy
reliance on the given mores of the dominant social groups in the respec-
tive societies.
To conclude, despite their practical preoccupations, human rights
activists have reason enough to pay attention to the skepticism that the
idea of human rights generates among many legal and political theorists.
These doubts have to be—and can be—addressed. But it is also impor-
tant to note that the conceptual understanding of human rights, in turn,
can benefit substantially from considering the reasoning that moves the
activists and the range and effectiveness of practical actions they under-
take, including recognition, monitoring and agitation, in addition to leg-
islation. Not only is conceptual clarity important for practice, the richness
of practice, I have argued, is also critically relevant for understanding the
concept and reach of human rights. There is, I must conclude, no great
deficit in the balance of trade between theory and practice.

SEN, A. Elements of A Theory of Human Rights

Uploaded by

Copyright:

Available Formats

SEN, A. Elements of A Theory of Human Rights

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

SEN, A. Elements of A Theory of Human Rights

Uploaded by

Copyright:

Available Formats

AMARTYA SEN Elements of a Theory of

I. The Need for a Theory

loose talk—perhaps kindly and well meaning forms of locution—but

2. Jeremy Bentham, Anarchical Fallacies; Being an Examination of the Declaration of

additions have certainly taken the contemporary literature on human

rights (including economic and social rights) can be adequately

II. Questions to Be Answered

These questions are addressed sequentially in what follows. However,

can be part of the obligations—often imperfect—generated by the

III. Human Rights: Ethics and Law

interactive process of critical scrutiny, open to information (including

concept).7 The perfectly specified demand not to torture anyone is sup-

7. I have discussed, in an earlier paper, the relevance of the Kantian distinction

domain of continued dispute is no embarrassment to a theory of human

the appropriate comparison to be that between the legal significance,

It is easy to see that Bentham’s rejection of the idea of natural “rights

Impossibility of a Paretian Liberal,” Journal of Political Economy 78 (1970): 152–57, and

Amnesty International or OXFAM or Médicins Sans Frontièrs, can them-

IV. Rights, Freedoms and Social Influence

get reconciled to cutting down their desires to “realistic proportions”

V. Processes, Opportunities and Capabilities

28. On the concept of capability, see my “Equality of What?” in Tanner Lectures on

There has, however, been some serious criticism of describing these

policies of the British Raj in India. In terms of the actual functioning of

34. G. A. Cohen has presented arguments in favor of focusing on achieved function-

pursue ancestral customs, and it cannot be turned into an argument for

larly in comparison with alternative approaches that focus on incomes,

38. The plurality of concerns, involving processes as well as opportunities, which is

centration on “primary goods”).40 The territory that Rawls reserved for

VI. Duties, Reasonable Consideration and Imperfect Obligations

reasonable consideration to undertaking such an action might appear,

altogether different concerns that a person may, inter alia, sensibly

evaluative framework of human rights, which yields imperfect as well as

VII. Recognition, Agitation and Legislation

receive the lion’s share of attention in the theoretical literature in this

also be monitoring of violations of these rights and attempts to gener-

domestic judgments. Many other examples can be given from different

VIII. Economic and Social Rights

Unfortunately much writing and rhetoric on rights heedlessly pro-

In responding to this significant criticism, we have to invoke the

an impoverished one, can do to prevent violations of certain basic eco-

54. Cranston, “Are There Any Human Rights?” p. 13.

In assessing this line of rejection, we have to ask: why should complete

IX. The Reach of Public Reasoning

and it is to such a scrutiny that we have to look in order to proceed to a

particularly in his later works, to limit such public confrontation within

The universalist nature of Adam Smith’s approach raises the question

For example, Rosa Luxemburg, a leading Marxist thinker and political

I have discussed these diagnoses elsewhere.65 Contrary to cultural

societies becomes more freely available, and when disagreements with

. . . the murder of new-born infants was a practice allowed of in

What are taken to be perfectly “normal” and “sensible” in an insulated

by open public reasoning. They may or may not be reflected in a legal

You might also like