Capabilities and Human Rights
Capabilities and Human Rights
Martha C. Nussbaum 0
0 Thi s Article is brought to you for free and open access by FLASH: The F ordham Law Archive of Scholarship and History. It has been accepted for inclusion in Fordham Law Review by an authorized editor of FLASH: The F ordham Law Archive of Scholarship and History. For more information , please contact
Recommended Citation Martha C. Nussbaum, Capabilities and Human Rights, 66 Fordham L. Rev. 273 (1997). Available at: http://ir.lawnet.fordham.edu/flr/vol66/iss2/2
HEN governments and international agencies talk about
people's basic political and economic entitlements, they regularly
use the language of rights. When constitutions are written in the
modem era, and their framers wish to identify a group of particularly
urgent interests that deserve special protection, once again it is the
language of rights that is regularly preferred.
The language of rights has a moral resonance that makes it hard to
avoid in contemporary political discourse. But it is certainly not on
account of its theoretical and conceptual clarity that it has been
preferred. There are many different ways of thinking about what a right
is, and many different definitions of "human rights."' For example,
rights are often spoken of as entitlements that belong to all human
beings simply because they are human, or as especially urgent
interests of human beings as human beings that deserve protection
regardless of where people are situated.2 Within this tradition there are
differences. The dominant tradition has typically grounded rights in
the possession of rationality and language, thus implying that
nonhuman animals do not have them, and that mentally impaired humans
may not have them.3 Some philosophers have maintained that
sentience, instead, should be the basis of rights; thus, all animals would be
rights-bearers.4 In contrast to this entire group of natural-rights
theorists, there are also thinkers who treat all rights as artifacts of state
action.5 The latter position would seem to imply that there are no
* Ernst Freund Professor of Law and Ethics: Law School, Philosophy
Department, and Divinity School, The University of Chicago.
1. For one excellent recent account, with discussions of other views, see Alan
Gewirth, The Community of Rights (1996).
2. For just one example, this is the view of Thomas Paine. See Thomas Paine,
Rights of Man-Common Sense 80-85 (Alfred A. Knopf 1994) (quoting and
discussing the French Declaration of the Rights of Man and of Citizens); id. at 114 (insisting
that rights, so conceived, should be the foundation of a nation's prosperity). Such
views ultimately derive from ancient Greek and Roman Stoic views of natural law.
The Latin word ius can be translated either as "right" or as "law," Grotius already
discussed the manifold applications of ius. See Hugo Grotius, De lure Belli Ac Pacis
(On the Law of War and Peace) (P.C. Molhuysen, A.W. Sijthoff 1919) (1625).
3. The most influential exemplar of such a view, followed by most later theorists,
is Cicero. See M. Tulli Ciceronis, De Officiis (On Duties), bk. 1,paras. 11-14 (Oxford
Univ. Press 1994) (distinguishing humans from beasts by reference to rationality and
language); id. paras. 20-41 (deriving duties from this).
4. See Peter Singer, Animal Liberation (2d ed. 1990).
5. This view is most influentially found in Kant. See Immanuel Kant, The
Metaphysics of Morals,in Kant: Political Writings 132-35 (Hans Reiss ed. & H.B. Nisbet
trans., Cambridge Univ. Press 2d enlarged ed. 1991) (1798) (defining right and the
theory of right with reference to law and the state).
human rights where there is no state to recognize them. Such an
approach appears to the holders of the former view to do away with the
very point of rights language, which is to point to the fact that human
beings are entitled to certain types of treatment whether or not the
state in which they happen to live recognizes this fact.
There are many other complex unresolved theoretical questions
about rights. One of them is the question whether the individual is
the only bearer of rights, or whether rights belong, as well, to other
entities, such as families, ethnic, religious, and linguistic groups, and
nations. Another is whether rights are to be regarded as
side-constraints on goal-seeking action, or as parts of a goal that is to be
promoted.6 Still another unresolved question is whether rights-thought
of as justified entitlements-are correlated with duties. If A has a
right to S, then it would appear there must be someone who has a duty
to provide S to A. But it is not always clear who has these
dutiesespecially when we think of rights in the international context. Again,
it is also unclear whether all duties are correlated with rights. One
might hold, for example, that we have a duty not to cause pain to
animals without holding that animals have rights-if, for example, one
accepted one of the classic accounts of the basis of rights that makes
reference to the abilities of speech and reason as the foundation, and
yet still believed that we have other strong reasons not to cause
Finally, there are difficult theoretical questions about what rights
are to be understood as rights to. When we speak of human rights, do
we mean, primarily, a right to be treated in certain ways? A right to a
certain level of achieved well-being? A right to certain resources with
which one may pursue one's life plan? A right to certain
opportunities and capacities with which one may, in turn, make choices
regarding one's life plan? Political philosophers who debate the nature of
equality standardly tackle a related question head on, asking whether
the equality most relevant to political distribution should be
understood, primarily, as equality of well-being, or equality of resources, or
equality of opportunity, or equality of capabilities.' The language of
6. An influential example of the first approach is in Robert Nozick, Anarchy,
State, and Utopia 26-53 (1974) (Chapter 3: Moral Constraints and the State), arguing
that rights supply moral constraints on state action. See also Samuel Scheffler, The
Rejection of Consequentialism (rev. ed. 1994) (developing a theory of rights as side
constraints). For the second approach, see, for example, Amartya Sen, Rights as
Goals, in Equality and Discrimination: Essays in Freedom and Justice (Stephen
Guest & Alan Milne eds., 1985) [hereinafter Rights as Goals], developing an account
of rights as among the goals of public action.
7. See Amartya Sen, Equalityof What?, I The Tanner Lectures on Human Values
195 (Sterling M. McMurrin ed., 1980), reprintedin Choice, Welfare and Measurement
353 (1982) [hereinafter Equality of What?] (arguing that the most relevant type of
equality for political purposes is equality of capability); see also Amartya Sen,
Inequality Reexamined passim (1992) [hereinafter Inequality Reexamined] (making the
same case in more detail); Richard J. Arneson, Equality and Equal Opportunityfor
rights to some extent cuts across this debate and obscures the issues
that have been articulated.
Thus, one might conclude that the language of rights is not
especially informative, despite its uplifting character, unless its users link
their references to rights to a theory that answers at least some of
these questions.8 It is for this reason, among others, that a different
language has begun to take hold in talk about people's basic
entitlements. This is the language of capabilities and human functioning.
Since 1993, the Human Development Reports of the United Nations
Development Programme 9 ("UNDP") have assessed the quality of
life in the nations of the world using the concept of people's
capabilities, or their abilities to do and to be certain things deemed valuable."0
Under the influence of economist/philosopher Amartya Sen, they
have chosen that conceptual framework as basic to inter-country
comparisons and to the articulation of goals for public policy.
Along with Sen, I have been one of the people who have pioneered
what is now called the "capabilities approach," defending its
importance in international debates about welfare and quality of life. My
own use of this language was originally independent, and reflected the
fact that Aristotle used a notion of human capability (Greek dunamis)
and functioning (Greek energeia) in order to articulate some of the
goals of good political organization. 1 But the projects soon became
fused: I increasingly articulated the Aristotelian idea of capability in
Welfare, 56 Phil. Stud. 77 (1989) (defending equality of opportunity for welfare); G.A.
Cohen, On the Currency of EgalitarianJustice, 99 Ethics 906, 920-21 (1989) (arguing
that the right thing to equalize is "access to advantage"); Ronald Dworkin, What Is
Equality? Part1: Equality of Welfare, 10 Phil. & Pub. Aff. 185 (1981) (discussing
distributional equality); Ronald Dworkin, What Is Equality? Part 2: Equality of
Resources, 10 Phil. & Pub. Aff. 283 (1981) (arguing that the right thing to equalize are
resources, and defining a suitable conception of equality of resources); John E.
Roemer, Equality of Resources Implies Equality of Welfare, 101 04. Econ. 751 (1986)
(arguing that, suitably understood, equality of resources implies equality of welfare).
8. See, e.g., Bernard Williams, The Standardof Living: Interestsand Capabilities,
in The Standard of Living 94, 100 (Geoffrey Hawthorn ed., 1987) (arguing for an
approach to basic human rights through basic capabilities).
9. See, e.g., United Nations Development Progamme, Human Development
Report 1996; United Nations Development Progamme, Human Development Report
1993 [hereinafter Human Development Report 1993].
10. The reports' primary measure of quality of life is the -human development
index" ("HDI"). Human Development Report 1993, supra note 9, at 10. HDI is a
composite of three basic components of human development: longevity (measured by
life expectancy), knowledge (measured by a combination of adult literacy and mean
years of schooling), and standard of living (measured by income relative to the
poverty level). Id-at 100. For a standard definition of capabilities, see Amartya Sen,
Capability and Well-Being, in The Quality of Life 30-31 (Martha Nussbaum &
Amartya Sen eds., 1993), explaining the choice of the term and its relationship to
other basic concepts.
11. See Martha C. Nussbaum, Nature,Function,and Capability: Aristotle on
Political Distribution,in [Supplementary Volume] Oxford Studies in Ancient Philosophy
145 (Julia Annas & Robert H. Grimm eds., 1988) [hereinafter Nature, Function,and
terms pertinent to the contemporary debate, 2 while Sen increasingly
emphasized the ancient roots of his idea. 3 In a variety of contexts, we
argued that the capabilities approach was a valuable theoretical
framework for public policy, especially in the international
development context. 14 We commended it to both theoreticians and
practitioners as offering certain advantages over approaches that focus on
opulence-GNP per capita, or welfare-construed in terms of utility
or desire-satisfaction, or even the distribution of basic resources. 15
Both Sen and I stated from the start that the capabilities approach
needs to be combined with a focus on rights. Sen wrote about rights
as central goals of public policy throughout the period during which
he developed the approach. 16 I stressed from the start that Aristotle's
theory was grossly defective because it lacked a theory of the basic
human rights, especially rights to be free from government
interference in certain areas of choice. 7 More recently, responding to
com12. See Martha Nussbaum, Aristotelian Social Democracy, in Liberalism and the
Good 203 (R. Bruce Douglass et al. eds., 1990) [hereinafter Aristotelian Social
Democracy]; Martha C. Nussbaum, Aristotle on Human Nature and the Foundations of
Ethics, in World, Mind, and Ethics: Essays on the Ethical Philosophy of Bernard
(J.E.J. Altham & Ross Harrison eds., 1995)
[hereinafter Human Nature];
Martha C. Nussbaum, Human Capabilities,Female Human Beings, in Women,
Culture, and Development 61 (M. Nussbaum & J. Glover eds., 1995) [hereinafter Human
Capabilities];Martha Nussbaum, Non-Relative Virtues: An Aristotelian Approach, in
The Quality of Life, supranote 10, at 242; Martha C. Nussbaum, Human Functioning
and Social Justice: In Defense of AristotelianEssentialism, 20 Pol. Theory 202 (1992)
[hereinafter Human Functioning];Martha C. Nussbaum, The Good as Discipline, The
Good as Freedom, in The Ethics of Consumption and Global Stewardship 312 (D.
Crocker & T. Linden eds., forthcoming 1998) (manuscript on file with the Fordham
Law Review) [hereinafter The Good as Discipline, The Good as Freedom]; Martha C.
Nussbaum, Sex and Social Justice (forthcoming 1998) (Chapter 1: Women and
Cultural Universals) (manuscript on file with the FordhamLaw Review) [hereinafter
Women and Cultural Universals].
13. See, for example, Inequality Reexamined, supranote 7, which also contains his
most recent formulation of the approach.
14. A good summary of our approaches, and the similarities and differences
between Sen's and my views, is in David A. Crocker, Functioningand Capability: The
Foundations of Sen's and Nussbaum's Development Ethic, 20 Pol. Theory 584 (1992)
[hereinafter Functioningand Capability: Part1], and David A. Crocker, Functioning
and Capability: The Foundationsof Sen's and Nussbaum's Development Ethic, Part2,
in Women, Culture, and Development, supra note 12, at 153 [hereinafter Functioning
and Capability: Part2].
15. See Amartya Sen, Capability and Well-Being, in The Quality of Life, supra
note 10, at 30; Amartya Sen, Commodities and Capabilities (1985); Equality of What?,
supra note 7; Amartya Sen, Gender Inequality and Theories of Justice, in Women,
Culture, and Development, supra note 12, at 259 [hereinafter Gender Inequality];
Amartya Sen, Well-Being, Agency and Freedom: The Dewey Lectures 1984, 82 J. Phil.
169 (1985) [hereinafter Well-Being].
16. See Amartya Sen, Rights and Capabilities,in Morality and Objectivity: A
Tribute to J.L. Mackie 130 (T. Honderich ed., 1985), reprintedin Amartya Sen, Resources,
Values and Development 307-24 (1984) [hereinafter Rights and Capabilities];Rights as
Goals, supra note 6; Amartya Sen, Rights and Agency, 11 Phil. & Pub. Aff. 3 (1982)
[hereinafter Rights and Agency].
17. See Aristotelian Social Democracy, supra note 12, at 239.
munitarian critics of rights-based reasoning and to international
discussions that denigrate rights in favor of material well-being, both
Sen and I have even more strongly emphasized the importance of
rights to our own capabilities approach. We stressed the various roles
liberty plays within our respective theories and emphasized the
closeness of our approach to liberal theories such as that of John Rawls."8
Moreover, rights play an increasingly large role inside the account
of what the most important capabilities are. Unlike Sen, who prefers
to allow the account of the basic capabilities to remain largely implicit
in his statements, I have produced an explicit account of the most
central capabilities that should be the goal of public policy. The list is
continually being revised and adjusted, in accordance with my
methodological commitment to cross-cultural deliberation and criticism.
But another source of change has been an increasing determination to
bring the list down to earth, so to speak, making the "thick vague
conception of the good"' 9 a little less vague, so that it can do real
work guiding public policy. At this point, the aim is to come up with
the type of specification of a basic capability that could figure in a
constitution,2 or perform, apart from that, the role of a constitutional
In the process, I have increasingly used the language of rights, or
the related language of liberty and freedom, in fleshing out the
account of the basic capabilities. Thus, in Human Capabilities,I speak
of "legal guarantees of freedom of expression ...and of freedom of
religious exercise" 21 as aspects of the general capability to use one's
mind and one's senses in a way directed by one's own practical reason.
I also speak of "guarantees of non-interference with certain choices
that are especially personal and definitive of selfhood," and of "the
freedoms of assembly and political speech." 2' In a forthcoming paper,
I actually use the language of rights itself in articulating the capability
to seek employment outside the home, and several of the other
important capabilities.3 In part, this is a rhetorical choice, bringing the list
of capabilities into relation with international human rights
instruments that have a related content. But in part it also reflects a
theoretical decision to emphasize the affiliations of the approach with
liberal rights-based theories, in an era of widespread reaction against
the Enlightenment and its heritage.24
But there are still some large questions to be answered. The
relationship between the two concepts remains as yet underexplored.
Does the capabilities view supplement a theory of rights, or is it
intended to be a particular way of capturing what a theory of rights
captures? Is there any tension between a focus on capabilities and a
focus on rights? Are the two approaches competitors? On the other
hand, is there any reason why a capabilities theorist should welcome
the language of rights-that is, is there anything in the view itself that
leads naturally in the direction of recognizing rights? Would a
natural-law Catholic theorist who used an Aristotelian language of
capability and functioning, but rejected liberal rights-based language, be
making a conceptual error? 5 Does the capabilities view help us to
answer any of the difficult questions that I sketched above, which have
preoccupied theorists of rights? Does the capabilities view incline us
to opt for any particular set of answers to the various questions about
rights, or any particular conception of rights? For example, is Sen
justified in thinking that the capabilities view supports a conception of
rights as goals, rather than as side-constraints?2 6 Finally, is there any
reason, other than a merely rhetorical one, why we should continue to
use the language of rights in addition to the language of capabilities?
In short, the conceptual relationship needs further scrutiny.2 7
Commenting on Sen's Tanner Lectures in 1987, Bernard Williams
expressed sympathy with the capabilities approach, but called for a
I am not very happy myself with taking rights as the starting point.
The notion of a basic human right seems to me obscure enough, and
I would rather come at it from the perspective of basic human
capabilities. I would prefer capabilities to do the work, and if we are
going to have a language or rhetoric of rights, to have it delivered
from them, rather than the other way around. But I think that there
remains an unsolved p8roblem: how we should see the relations
between these concepts.
This paper is a contribution to that project. I shall not be able to
answer all the outstanding questions, and I shall certainly not be able to
offer a theory of rights that solves all the problems I outlined. But I
hope to illuminate some of the issues that must be faced when one
does attempt to connect the two ideas, some of the options one has,
some of the problems that arise, and some of the positive dividends
one may reap.
I shall begin by describing the capabilities approach and the
motivations for its introduction: what it was trying to do in political
philosophy, how it commended itself by contrast to other standard ways of
thinking about entitlements. Then I shall briefly clarify the
connection between the capabilities approach and liberal theories of justice.
Finally, I shall turn to my central topic, the relationship between rights
I. THE CAPABILITIES APPROACH: MOTIVATION AND ARGUMENT
Why, then, should there be a theory of human capabilities? What
questions does it answer, and what is its practical point? Why should
an international agency such as the UNDP use a measure of quality of
life based on human capability and functioning, rather than other
more traditional measures: for example, those based on opulence,
utility, or a distribution of resources that satisfies some constraint,
whether it be a social minimum, or the Rawlsian Difference Principle,
or some more exacting egalitarian condition?
The account of human capabilities has been used as an answer to a
number of distinct questions, such as: What is the living standard?29
What is the quality of life?30 What is the relevant type of equality that
we should consider in political planning?3 It has also been closely
linked to discussion of a theory of justice, because such a theory has a
need for an account of what it is trying to achieve for people. I
believe that the most illuminating way of thinking about the capabilities
approach is that it is an account of the space within which we make
comparisons between individuals and across nations as to how well
they are doing. This idea is closely linked with the idea of a theory of
justice, since one crucial aim of a theory of justice typically is to
promote some desired state of people; and in Aristotelian Social
Democracy I linked it very closely to an account of the proper goal of
government, to bring all citizens up to a certain basic minimum level
28. Williams, supra note 8, at 100.
29. Id. at 100-02 (discussing Sen's proposal that the living standard should be
defined in terms of capabilities).
30. See The Quality of Life, supra note 10.
31. See Inequality Reexamined, supra note 7.
of capability." But up to a point, the approach is logically
independent of a theory of justice, since a theory of justice may acknowledge
many constraints with regard to how far it is entitled to promote
people's well-being. For example, Robert Nozick could grant that
capabilities are the relevant space within which to make comparisons of
well-being, while denying that this has anything at all to do with a
theory of justice, since he rejects theories of justice based on a
"patterned end-state" conception, preferring to define justice solely in
terms of procedures and entitlements.33
The capabilities idea is also closely linked to a concern with
equality, in that Sen has always used it to argue that people are entitled to a
certain level of rough material and social equality. But, strictly
speaking, these two concerns of Sen's are logically independent. One might
agree that capabilities are the relevant space within which to compare
lives and nations, and yet hold that equality of capability is not the
appropriate goal. Capabilities inform us as to what type of equality
might be thought pertinent; they do not by themselves tell us whether
we should value an equal distribution or some other distribution.
As a theory of the relevant space within which to make
comparisons, the capabilities approach is best understood by contrasting it
with its rivals in the international development arena. The most
common method of measuring the quality of life in a nation and making
cross-national comparisons used to be simply to enumerate GNP per
capita. This crude method is reminiscent of the economics lesson
imagined by Charles Dickens in Hard Times, and used by Sen and me
to introduce our volume on The Quality of Life:
"And he said, Now this schoolroom is a Nation. And in this nation,
there are fifty millions of money. Isn't this a prosperous nation?
Girl number twenty, isn't this a prosperous nation, and a'n't you in
a thriving state?"
"What did you say?" asked Louisa.
"Miss Louisa, I said I didn't know. I thought I couldn't know
whether it was a prosperous nation or not, and whether I was in a
thriving state or not, unless I knew who had got the money, and
whether any of it was mine. But that had nothing to do with it. It
was not in the figures at all," said Sissy, wiping her eyes.
"That was a great mistake of yours," observed Louisa. 34
In short, the crude approach does not even tell us who has the money,
and thus typically gave high marks to nations such as South Africa,
which contained enormous inequalities. Still less does it provide any
information at all about elements of human life that might be thought
very important in defining its quality, but that are not always well
cor32. Supra note 12.
33. Nozick, supranote 6, at 150-64 (criticizing patterned end-state conceptions in
favor of procedural conceptions).
34. Charles Dickens, Hard Times 74-75 (Oxford Univ. Press 1989) (1854).
related with GNP per capita: educational opportunities, health care,
life expectancy, infant mortality, the presence or absence of political
liberties, the extent of racial or gender inequality.
Somewhat less crude is an economic approach that measures quality
of life in terms of utility, understood as the satisfaction of preference
or desire." This approach at least has the advantage of concerning
itself to some degree with distribution, in the sense that it does look at
how resources are or are not going to work to make people's lives
better. But it has severe shortcomings. First, there is the familiar
problem that utilitarianism tends to think of the social total, or
average, as an aggregate, neglecting the salience of the boundaries
between individual lives.36 As Rawls pointed out, this approach means
that utilitarianism can tolerate a result in which the total is good
enough, but where some individuals suffer extremely acute levels of
deprivation, whether of resources or of liberty.37 In that sense, it does
not tell Sissy "who has got the money and whether any of it is mine,"
any more than does the GNP-based approach. (Indeed, Sissy's
teacher was clearly a Benthamite Utilitarian.) Rawls was convinced
that the failure of utilitarianism to justify adequately strong
protections for the basic political liberties, given this propensity to
aggregate, was by itself sufficient reason to reject it.' Bernard Williams,
similarly, has considered utilitarianism's neglect of the "separateness
of persons" to be a cardinal failure, and a reason why the theory
cannot give an adequate account of social well-being.3 9
A second problem with utilitarianism is its commitment to the
commensurability of value, the concern to measure the good in terms of a
single metric and thus to deny that there are irreducibly plural goods
that figure in a human life." Both Sen and I have pursued this
question extensively, apart from our work on capabilities.41 But it has also
had importance in justifying the capabilities approach, since the
quality of life seems to consist of a plurality of distinct features-features
that cannot be simply reduced to quantities of one another. This
recognition limits the nature of the tradeoffs it will be feasible to make.42
But a third feature of utilitarianism has been even more central to
the capability critique. As Sen has repeatedly pointed out, people's
satisfactions are not very reliable indicators of their quality of life.
Wealthy and privileged people get used to a high level of luxury, and
feel pain when they do not have delicacies that one may think they do
not really need. On the other hand, deprived people frequently adjust
their sights to the low level they know they can aspire to, and thus
actually experience satisfaction in connection with a very reduced
living standard. Sen gave a graphic example: In 1944, the year after the
Great Bengal Famine, the All-India Institute of Hygiene and Public
Health did a survey.43 Included in this survey were a large number of
bad, in all kinds of ways Tbhuet pnoostiotiroionusolfywiindotwersmins Ionfdhiaeailsthexsttraetmuse.l4y5
widows and widowers."
But in the survey, only 2.5 percent of widows, as against 48.5 percent
of widowers, reported that they were either ill or in indifferent
health. 46 And when the question was just about "indifferent health,"
as opposed to illness-for which we might suppose there are more
public and objective criteria-45.6 percent of widowers said their
health was "indifferent," as opposed to zero percent of the widows. 47
The likely explanation for this discrepency is that people who have
regularly been malnourished, who have in addition been told that they
are weak and made for suffering, and who, as widows, are told that
they are virtually dead and have no rights, will be unlikely to
recognize their fatigue and low energy as a sign of bodily disease; but not so
for males, who are brought up to have high expectations for their own
physical functioning. Sen concludes: "Quiet acceptance of
tion and bad fate affects the scale of dissatisfaction generated, and the
utilitarian calculus gives sanctity to that distortion."4"
This phenomenon of "adaptive preferences"-preferences that
adjust to the low level of functioning one can actually achieve-has by
now been much studied in the economic literature, 9 and is generally
recognized as a central problem, if one wants to use the utilitarian
calculus for any kind of normative purpose in guiding public policy."0
We are especially likely to encounter adaptive preferences when we
are studying groups that have been persistent victims of
discrimination, and who may as a result have internalized a conception of their
own unequal worth. It is certain to be true when we are concerned
with groups who have inadequate information about their situation,
their options, and the surrounding society-as is frequently the case,
for example, with women in developing countries. For these reasons,
then, the utility-based approach seems inadequate as a basis for
offering comparisons of quality of life.
Far more promising is an approach that looks at a group of basic
resources and then asks about their distribution, asking, in particular,
how well even the worst off citizens are doing with respect to the
items on the list. Such is the approach of John Rawls, who, in A
Theory of Justice and subsequent works, advanced a list of the "primary
goods" intended to be items that all rational individuals, regardless of
their more comprehensive plans of life, would desire as prerequisites
for carrying out those plans.5' These items include liberties,
opportunities, and powers, wealth and income, and the social basis of
selfrespect. More recently, Rawls has added freedom of movement and
the free choice of occupation.52 The idea is that we measure who is
better off and less well off by using such a list of primary resources;
49. See Jon Elster, Sour Grapes-Utilitarianismnand the Genesis of Wants, in
Utilitarianism and Beyond, supranote 36, at 219 (defining adaptive preferences and
arguing that their existence poses insuperable problems for utilitarianism); Amartya K.
Sen, Gender and Cooperative Conflicts, in Persistent Inequalities 123 (Irene Tinker
ed., 1990) (arguing that women frequently adjust their expectations to the low level of
well-being they can achieve, and that on this account a bargaining model of the family
is superior to a utilitarian account).
50. See Gary S. Becker, Nobel Lecture: The Econonic Wiay of Looking at
Behavior,in The Essence of Becker 633, 636-37 (Ram6n Febrero & Pedro S. Schwartz eds.,
1995) (arguing that the beliefs of employers, teachers, and others that minorities are
less productive can be self-fulfilling, causing minorities to underinvest in education
and work skills, thus becoming less productive than they would otherwise have been).
51. A Theory of Justice, supranote 18,at 62, 90-95, 396-97. More recently, Rawls
has qualified his view by stating that the primary goods are to be seen not as
allpurpose means, but as the needs of citizens understood from a political point of view,
in connection with the development and expression of their "moral powers." He has
stressed that the account of the moral powers--of forming and revising a life plan-is
itself an important part of the political theory of the good. See Political Liberalism,
supra note 18, at 178-90.
52. Political Liberalism, supra note 18, at 181.
that information is used, in turn, by the parties who are choosing
principles of justice. Notice that this list is heterogeneous. Some of its
items are capacities of persons such as liberties, opportunities, and
powers, and the social basis of self-respect is a complex property of
society's relation to persons, but income and wealth are pure
resources. And income and wealth frequently play a central role in the
measurement of who is better and worse off.5 3 Rawls was at pains,
moreover, to state that this list of "primary goods" is not a
comprehensive theory of what is good or valuable in life.54 For Rawls, the
attraction of operating with a list of resources is that it enables the
approach to steer clear of prescribing the basic values of human life,
which individuals must be able to select for themselves, in accordance
with their own more comprehensive religious or ethical conceptions.
Sen's basic argument against Rawls, for the past twenty years, has
been that the space of resources is inadequate as a space within which
to answer questions about who is better and who is worse off." The
inadequacy derives from the fact that individuals vary greatly in their
need for resources and in their ability to convert resources into
valuable functionings. Some of these differences are physical. Nutritional
needs vary with age, occupation, and sex. A pregnant or lactating
woman needs more nutrients than a non-pregnant woman. A child needs
more protein than an adult. A person whose limbs work well needs
few resources to be mobile, whereas a person with paralyzed limbs
needs many more resources to achieve the same level of mobility.
Many such variations escape our notice if we live in a prosperous
nation that can afford to bring all individuals to a high level of physical
attainment; in the developing world we must be highly alert to these
variations in need. Some of the variations, again, are social, and have
to do with traditional social hierarchies. If we wish to bring all citizens
of a nation to the same level of educational attainment, we will need
to devote more resources to those who encounter obstacles from
traditional hierarchy or prejudice. Thus, women's literacy will prove
more expensive than men's literacy in many parts of the world. This
means that if we operate only with an index of resources, we will
frequently reinforce inequalities that are highly relevant to well-being.
An approach focusing on resources does not go deep enough to
diagnose obstacles that can be present even when resources seem to be
adequately spread around, causing individuals to fail to avail
themselves of opportunities that they in some sense have, such as free
public education, the the right to vote, or the right to work.
53. A Theory of Justice, supranote 18, at 97-98 (discussing different ways of
defining the least well off-both favored approaches focus on income and wealth as
54. Political Liberalism, supra note 18, at 187-88.
55. See Equality of What?, supra note 7, at 364-67; Gender Inequality, supra note
15, at 263-66.
For this reason, we argue that the most appropriate space for
comparisons is the space of capabilities. Instead of asking "How satisfied
is person A," or "How much in the way of resources does A
command," we ask the question: "What is A actually able to do and to
be?" In other words, about a variety of functions that would seem to
be of central importance to a human life, we ask: Is the person
capable of this, or not? This focus on capabilities, unlike the focus on
GNP, or on aggregate utility, looks at people one by one, insisting on
locating empowerment in this life and in that life, rather than in the
nation as a whole. Unlike the utilitarian focus on satisfactions, it
looks not at what people feel about what they do, but about what they
are actually able to do.56 Nor does it make any assumptions about the
commensurability of the different pursuits. Indeed, this view denies
that the most important functions are all commensurable in terms of a
single metric and it treats the diverse functions as all important, and
all irreducibly plural. 7 Finally, unlike the focus on resources, it is
concerned with what is actually going on in the life in question: not
how many resources are sitting around, but how they are actually
going to work in enabling people to function in a fully human way.58
II. THE CENTRAL HUMAN CAPABILITIES
Sen has focused on the general defense of the capability space, and
has not offered any official account of what the most central human
capabilities are, although in practice he has to some extent done so, by
focusing on some areas of human life and not others in constructing
the measures used in the Human Development Reports5.9 Again, his
recent book on India gives many concrete examples of the importance
and the interrelationships of various concrete human capabilities.' I,
by contrast, have focused on the task of producing such a working list,
describing a methodology by which we might both generate and
justify such a list6 and defending the whole project of giving such a list
against the objections of relativists and traditionalists. 62 The list is
supposed to be a focus for political planning, and it is supposed to
select those human capabilities that can be convincingly argued to be
of central importance in any human life, whatever else the person
pursues or chooses. The central capabilities are not just instrumental to
further pursuits: They are held to have value in themselves, in making
a life fully human. But they are held to have a particularly central
importance in everything else we plan and choose. In that sense,
central capabilities play a role similar to that played by primary goods in
Rawls's more recent account: They support our powers of practical
reason and choice, and have a special importance in making any
choice of a way of life possible. They thus have a special claim to be
supported for political purposes in societies that otherwise contain a
great diversity of views about the good. I do not think of the political
sphere in exactly the way that Rawls conceives it, since I do not make
the assumption that the nation-state should be the basic deliberative
unit,63 and the account is meant to have broad applicability to
crosscultural deliberations. Nonetheless, the basic point of the account is
the same: to put forward something that people from many different
traditions, with many different fuller conceptions of the good, can
agree on as the necessary basis for pursuing their good life.64
The list is an attempt to summarize the empirical findings of a broad
and ongoing cross-cultural inquiry. As such, it is open-ended and
humble; it can always be contested and remade. It does not claim to
read facts of "human nature" off of biological observation, although it
does of course take account of biology as a relatively constant element
in human experience. Nor does it deny that the items on the list are to
some extent differently constructed by different societies. Indeed,
part of the idea of the list is that its members can be more concretely
specified in accordance with local beliefs and circumstances. In that
sense, the consensus it hopes to evoke has many of the features of the
"overlapping consensus" described by Rawls.65
Here is the current version of the list, revised as a result of my
recent visits to development projects in India:6
62. See Human Capabilities,supra note 12, at 67-72, 93-95; Human Functioning,
supra note 12; Women and Cultural Universals, supra note 12, at 12-20.
63. For an excellent discussion of this question, and a critique of Rawis with which
I largely agree, see Thomas W. Pogge, Realizing Rawls 211-280 (1989).
64. See The Good as Discipline, The Good as Freedom, supra note 12, at 324,
where I have stressed this political-liberal role of the capabilities list more than in
65. Political Liberalism, supra note 18, passim.
66. The primary changes are a greater emphasis on bodily integrity, a focus on
dignity and non-humiliation, and an emphasis on control over one's environment.
Oddly, these features of human "self-sufficiency" are the ones most often criticized by
Western feminists as "male" and "Western"-one reason for their more muted role in
earlier versions of the list. See Martha C. Nussbaum, The Feminist Critique of
1. LmE. Being able to live to the end of a human life of normal
length; not dying prematurely, or before one's life is so reduced as
to be not worth living.
2. BODILY HEALTH. Being able to have good health, including
reproductive health; to be adequately nourished; to have adequate
3. BODILY INTEGRITY. Being able to move freely from place to
place; to be secure against violent assault, including sexual assault
and domestic violence; having opportunities for sexual satisfaction
and for choice in matters of reproduction.
4. SENSES, IMAGINATION, AND THOUGHT. Being able to use the
senses; being able to imagine, to think, and to reason-and to do
these things in a "truly human" way, a way informed and cultivated
by an adequate education, including, but by no means limited to,
literacy and basic mathematical and scientific training. Being able
to use imagination and thought in connection with experiencing and
producing expressive works and events of one's own choice,
religious, literary, musical, and so forth. Being able to use one's mind
in ways protected by guarantees of freedom of expression with
respect to both political and artistic speech and freedom of religious
exercise. Being able to have pleasurable experiences and to avoid
5. EMOTIONS. Being able to have attachments to things and
people outside ourselves; to love those who love and care for us, to
grieve at their absence; in general, to love, to grieve, to experience
longing, gratitude, and justified anger. Not having one's emotional
development blighted by fear and anxiety. Supporting this
capability means supporting forms of human association that can be shown
to be crucial in their development.
6. PRACTICAL REASON. Being able to form a conception of the
good and to engage in critical reflection about the planning of one's
life. This entails protection for the liberty of conscience and
A. FRIENDSHIP. Being able to live for and to others, to recognize
and show concern for other human beings, to engage in various
forms of social interaction; to be able to imagine the situation of
another and to have compassion for that situation; to have the
capability for both justice and friendship. Protecting this capability
means, once again, protecting institutions that constitute such forms
of affiliation, and also protecting the freedoms of assembly and
B. REsPEcr. Having the social bases of self-respect and
non-humiliation; being able to be treated as a dignified being whose worth
is equal to that of others. This entails provisions of
non-discrimination on the basis of race, sex, ethnicity, caste, religion, and national
alism, The Lindley Lecture, Univ. of Kansas (1997), also in Sex and Social Justice
(Martha C. Nussbaum ed., forthcoming 1998).
8. OTHER SPECIES. Being able to live with concern for and in rela
tion to animals, plants, and the world of nature.
9. PLAY. Being able to laugh, to play, and to enjoy recreational
10. CONTROL OVER ONE'S ENVIRONMENT.
A. POLITICAL. Being able to participate effectively in political
choices that govern one's life; having the right of political
participation, protections of free speech and association.
B. MATERIAL. Being able to hold property (both land and
movable goods); having the right to employment; having freedom from
unwarranted search and seizure.
The list is, emphatically, a list of separate and indispensible
components. We cannot satisfy the need for one of them by giving a larger
amount of another. All are of central importance and all are distinct
in quality. Practical reason and affiliation, I argue elsewhere, are of
special importance because they both organize and suffuse all the
other capabilities, making their pursuit truly human.67 The individual
importance of each component limits the trade-offs that it will be
reasonable to make, and thus limits the applicability of quantitative
costbenefit analysis. At the same time, the items on the list are related to
one another in many complex ways. One of the most effective ways of
promoting women's control over their environment, and their
effective right of political participation, is to promote women's literacy.
Women who can seek employment outside the home have more
resources in protecting their bodily integrity from assaults within it.
CAPABILITY OF GOAL
I have spoken of both functioning and capability. How are they
related? Understanding this relationship is crucial in defining the
relation of the "capabilities approach" to both liberalism and views of
human rights. For if we were to take functioning itself as the goal of
public policy, the liberal would rightly judge that we were precluding
many choices that citizens may make in accordance with their own
conceptions of the good, and perhaps violating their rights. A deeply
religious person may prefer not to be well-nourished, but instead
prefer to engage in strenuous fasting. Whether for religious or for other
reasons, a person may prefer a celibate life to one containing sexual
expression. A person may prefer to work with an intense dedication
that precludes recreation and play. Am I declaring, by my very use of
the list, that these are not fully human or flourishing lives? And am I
instructing government to nudge or push people into functioning of
the requisite sort, no matter what they prefer?
67. See Aristotelian Social Democracy,supra note 12, at 226-28; Human Nature,
supra note 12, at 102-20.
It is important that the answer to these questions is no. Capability,
not functioning, is the political goal. Capability must be the goal
because of the great importance the capabilities approach attaches to
practical reason, as a good that both suffuses all the other functions,
making them human rather than animal,' s and figures itself as a
central function on the list. It is perfectly true that functionings, not
simply capabilities, are what render a life fully human: If there were no
functioning of any kind in a life, we could hardly applaud it, no matter
what opportunities it contained. Nonetheless, for political purposes it
is appropriate for us to strive for capabilities, and those alone.
Citizens must be left free to determine their course after they have the
capabilities. The person with plenty of food may always choose to
fast, but there is a great difference between fasting and starving, and it
is this difference that we wish to capture. Again, the person who has
normal opportunities for sexual satisfaction can always choose a life of
celibacy, and we say nothing against this. What I speak against, for
example, is the practice of female genital mutilation, which deprives
individuals of the opportunity to choose sexual functioning, and
indeed, the opportunity to choose celibacy as well. 69 A person who has
opportunities for play can always choose a workaholic life. Again,
there is a great difference between that chosen life and a life
constrained by insufficient maximum-hour protections and/or the "double
day" that make women unable to play in many parts of the world.
I can make the issue clearer, and also prepare for discussion of the
relationship between capabilities and rights, by pointing out that there
are three different types of capabilities that figure in my analysis.70
First, there are what I call basic capabilities: the innate equipment of
individuals that is the necessary basis for developing the more
advanced capability. Most infants have from birth the basic capability
for practical reason and imagination, though they cannot exercise such
functions without a lot more development and education. Second,
there are internal capabilities: that is, states of the person herself that
are, so far as the person herself is concerned, sufficient conditions for
the exercise of the requisite functions. A woman who has not suffered
genital mutilation has the internalcapability for sexual pleasure; most
adult human beings everywhere have the internal capability to use
speech and thought in accordance with their own conscience. Finally,
there are combined capabilities7, which I define as internal
capabilities combined with suitable external conditions for the exercise of the
function. A woman who is not mutilated but is secluded and
forbidden to leave the house has internal but not combined capabilities for
sexual expression-and work, and political participation. Citizens of
repressive non-democratic regimes have the internal but not the
combined capability to exercise thought and speech in accordance with
their conscience. The aim of public policy is the production of
combined capabilities. This idea means promoting the states of the person
by providing the necessary education and care, as well as preparing
the environment so that it is favorable for the exercise of practical
reason and the other major functions.72
This explanation of the types of capability clarifies my position. I
am not saying that public policy should rest content with
internalcapabilities,but remain indifferent to the struggles of individuals who have
to try to exercise these capabilities in a hostile environment. In that
sense, my approach is highly attentive to the goal of functioning, and
instructs governments to keep functioning always in view. On the
other hand, I am not pushing individuals into the function: once the
stage is fully set, the choice is up to them.
The approach is therefore very close to Rawls's approach using the
notion of primary goods.73 We can see the list of capabilities as like a
long list of opportunities for life-functioning, such that it is always
rational to want them whatever else one wants. If one ends up having a
plan of life that does not make use of all of them, one has hardly been
harmed by having the chance to choose a life that does. Indeed, in the
cases of fasting and celibacy it is the very availability of the alternative
course that gives the choice its moral value. The primary difference
between this capabilities list and Rawls's list of primary goods is its
length and definiteness, and in particular its determination to include
the social basis of several goods that Rawls has called "natural goods,"
such as "health and vigor, intelligence and imagination."74 Since
Rawls has been willing to put the social basis of self-respect on his list,
it is not at all clear why he has not made the same move with
imagination and health." Rawls's evident concern is that no society can
guar71. In earlier papers I called these "external capabilities," see, for example,
Nature, Function,and Capability,supranote 11, at 164, but Crocker has suggested to me
that this suggests a misleading contrast with "internal."
72. This distinction is related to Rawls's distinction between social and natural
primary goods. A Theory of Justice, supranote 18, at 62. Whereas he holds that only
the social primary goods should be on the list, and not the natural (such as health and
imagination), we say that the social basis of the natural primary goods should most
emphatically be on the list.
73. Id. at 62, 90-95.
74. Id. at 62.
75. Rawls comments that "although their possession is influenced by the basic
structure, they are not so directly under its control." Id. This is of course true if we
are thinking of health, but if we think of the social basis of health, it is not true. It
antee health to its individuals-in that sense, saying that the goal is
full external capability may appear unreasonably idealistic. Some of
the capabilities, for example, some of the political liberties, can be
fully guaranteed by society, but many others involve an element of
chance and cannot be so guaranteed. My response to this concern is
that the list is a list of political goals that should be useful as a
benchmark for aspiration and comparison. Even though individuals with
adequate health support often fall ill, it still makes sense to compare
societies by asking about actual health-capabilities, since we assume
that the comparison will reflect the different inputs of human
planning, and can be adjusted to take account of more and less favorable
natural situations. Sometimes, however, it is easier to get information
on health achievements than on health capabilities; to some extent we
must work with the information we have, while not forgetting the
importance of the distinction.
In saying these things about the political goal, we focus on adults
who have full mental and moral powers-what Rawls calls "normal
cooperating member[s] of society."76 Children are different, since we
are trying to promote the development of adult capabilities. We may
in some cases be justified in requiring functioning of an immature
child, as with compulsory primary and secondary education, but we
must always justify coercive treatment of children with reference to
the adult-capability goal.
Earlier versions of the list appeared to diverge from the approach of
Rawlsian liberalism by not giving as large a place to the traditional
political rights and liberties-although the need to incorporate them
was stressed from the start.' This version of the list corrects that
defect of emphasis. These political liberties have a central importance in
rendering well-being human. A society that aims at well-being while
overriding these liberties has delivered to its members a merely
animal level of satisfaction.78 As Sen has recently written: "Political
rights are important not only for the fulfillment of needs, they are
crucial also for the formulation of needs. And this idea relates, in the
end, to the respect that we owe each other as fellow human beings. 79
This idea of freedoms as need has recently been echoed by Rawls:
primary goods specify what citizens' needs are from the point of view
of political justice.80
seems to me that the case for putting these items on the political list is just as strong as
the case for the social basis of self-respect. In The Priority of Right and Ideas of the
Good,17 Phil. & Pub. Aff. 251,257 (1988), Rawls suggests adding leisure time and the
absence of pain to the list. He makes the same suggestion in Political Liberalism,
supra note 18, at 181-82. For Rawls's current treatment of health, see id. at 184.
76. Political Liberalism, supra note 18, at 183.
77. See Aristotelian Social Democracy,supra note 12, at 239-40.
78. See Human Nature,supra note 12, at 110-120.
79. Freedoms and Needs, supra note 18, at 38.
80. Political Liberalism, supra note 18, at 187-88.
The capability view justifies its elaborate list by pointing out that
choice is not pure spontaneity, flourishing independently of material
and social conditions. If one cares about people's powers to choose a
conception of the good, then one must care about the rest of the form
of life that supports those powers, including its material conditions.
Thus, the approach claims that its more comprehensive concern with
flourishing is perfectly consistent with the impetus behind the
Rawlsian project. Rawls has always insisted that we are not to rest content
with merely formal equal liberty and opportunity, but that we must
pursue their fully equal worth by ensuring that unfavorable economic
and social circumstances do not prevent people from availing
themselves of liberties and opportunities that are formally open to them.81
The guiding thought behind this form of Aristotelianism is, at its
heart, a profoundly liberal idea,8' and one that lies at the heart of
Rawls's project as well: the idea of the citizen as a free and dignified
human being, a maker of choices.83 Politics here has an urgent role to
play, providing citizens with the tools that they need, both in order to
choose at all and in order to have a realistic option of exercising the
most valuable functions. The choice of whether and how to use the
tools, however, is left up to the citizens, in the conviction that this
choice is an essential aspect of respect for their freedom. They are
seen not as passive recipients of social patterning, but as dignified free
beings who shape their own lives.84
IV. RIGHTS AND CAPABILITIES: Two DIFFERENT RELATIONSHIPS
How, then, are capabilities related to human rights? We can see, by
this time, that there are two rather different relations that capabilities
have to the human rights traditionally recognized by international
human rights instruments. In what follows, I shall understand a
human right to involve an especially urgent and morally justified claim
that a person has, simply by virtue of being a human adult, and
independently of membership in a particular nation, or class, or sex, or
ethnic or religious or sexual group.
First, there are some areas in which the best way of thinking about
rights is to see them as, what I have called, combined capabilities to
function in various ways. The right to political participation, the right
to religious free exercise, the freedom of speech, the freedom to seek
81. A Theory of Justice, supra note 18, at 83-90, 224-27.
82. Though in one form Aristotle had it too. See Human Nature,supra note 12, at
83. See A Theory of Justice, supranote 18, at 251-57; Freedoms and Needs, supra
note 18, at 38.
84. Cf Freedoms and Needs, supra note 18, at 38 ("The importance of political
rights for the understanding of economic needs turns ultimately on seeing human
beings as people with rights to exercise, not as parts of a 'stock' or a 'population' that
passively exists and must be looked after. What matters, finally, is how we see each
employment outside the home, and the freedom from unwarranted
search and seizure are all best thought of as human capacities to
function in ways that we then go on to specify. The further specification
will usually involve both an internal component and an external
component: a citizen who is systematically deprived of information about
religion does not really have religious liberty, even if the state imposes
no barrier to religious choice. On the other hand, internal conditions
are not enough: women who can think about work outside the home,
but who are going to be systematically denied employment on account
of sex, or beaten if they try to go outside, do not have the right to seek
employment. In short, to secure a right to a citizen in these areas is to
put them in a position of capability to go ahead with choosing that
function if they should so desire.
Of course, there is another way in which we use the term "right" in
which it could not be identified with a capability. We say that A has
"a right to" seek employment outside the home, even when her
circumstances obviously do not secure such a right to her. When we use
the term "human right" this way, we are saying that just by virtue of
being human, a person has a justified claim to have the capability
secured to her: so a right in that sense would be prior to capability, and
a ground for the securing of a capability. "Human rights" used in this
sense lie very close to what I have called "basic capabilities," since
typically human rights are thought to derive from some actual feature
of human persons, some untrained power in them that demands or
calls for support from the world. Rights theories differ about which
basic capabilities of the person are relevant to rights, but the ones
most commonly chosen are the power of reasoning, generally
understood to be moral reasoning, and the power of moral choice.'
On the other hand, when we say, as we frequently do, that citizens
in country C "have the right of free religious exercise," what we
typically mean is that this urgent and justified claim is being answered,
that the state responds to the claim that they have just by virtue of
being human. It is in this sense that capabilities and rights should be
seen to be equivalent: For I have said, combined capabilities are the
goals of public planning.
Why is it a good idea to understand rights, so understood, in terms
of capabilities? I think this approach is a good idea because we then
understand that what is involved in securing a right to people is
usually a lot more than simply putting it down on paper. We see this very
85. This way of thinking derives from the ancient Stoic tradition, continued
through Cicero and on into Grotius and Kant. See Martha C. Nussbaum, Kant and
Stoic Cosmopolitanism,5 J. Pol. Phil. 1 (1997) [hereinafter Kant and Stoic
Cosmopolitanism];Martha C. Nussbaum, The Incomplete Feminism of Musonius Rufus:
Platonis4 Stoic, and Roman, Paper presented at the Conference on Gender and Sexual
Experience in Ancient Greece and Rome, Finnish Academy in Rome (June 22-25,
1997) (on file with the FordliamLaw Review).
clearly in India, for example, where the Constitution is full of
guarantees of Fundamental Rights that are not backed up by effective state
action. Thus, since ratification women have had fights of sex
equality-but in real life they are unequal not only defacto, but also de ]ure.
This inequality results from the fact that most of the religious legal
systems that constitute the entire Indian system of civil law have
unequal provisions for the sexes, very few of which have been declared
unconstitutional.86 So we should not say that women have equal
fights, since they do not have the capabilities to function as equals.
Again, women in many nations have a nominal fight of political
participation without really having this fight in the sense of capability:
for they are secluded and threatened with violence should they leave
the home. This is not what it is to have a fight. In short, thinking in
terms of capability gives us a benchmark in thinking about what it is
really to secure a right to someone.
There is another set of fights, largely those in the area of property
and economic advantage, which seem to me analytically different in
their relationship to capabilities. Take, for example, the fight to a
certain level of income, or the fight to shelter and housing. These are
rights that can be analyzed in a number of distinct ways, in terms of
resources, or utility, or capabilities. We could think of the fight to a
decent level of living as a fight to a certain level of resources; or, less
plausibly, as a fight to a certain level of satisfaction; or as a right to
attain a certain level of capability to function.
Once again, we must distinguish the use of the term "fight" in the
sentence "A has a fight to X," from its use in the sentence "Country C
gives citizens the fight to X." All human beings may arguably have a
right to something in the first sense, without being in countries that
secure these fights. If a decent living standard is a human fight, then
American citizens have that fight although their state does not give
them, or secure to them, such a fight. So far, then, we have the same
distinctions on our hands that we did in the case of the political
liberties. But the point I am making is that at the second level, the analysis
of "Country C secures to its citizens the right to a decent living
standard" may plausibly take a wider range of forms than it does for the
political and religious liberties, where it seems evident that the best
way to think of the secured fight is as a capability. The material fights
may, by contrast, plausibly be analyzed in terms of resources, or
possibly in terms of utility.
86. See Religion and Women's Human Rights, supra note 69, at 121-26 (reviewing
this situation). Typically, only small and unpopular religions get their laws thrown
out. Thus, the Christian inheritance law-or one of them, since Christians in India
are governed by a bewildering variety of different systems of Christian law-was
declared unconstitutional on grounds of sex equality, but the attempt to set aside a part
of the Hindu marriage act on these grounds was reversed at the Supreme Court level.
Id. at 108.
Here again, however, I think it is valuable to understand these
rights, insofar as we decide we want to recognize them, in terms of
capabilities. That is, if we think of a right to a decent level of living as
a right to a certain quantity of resources, then we get into the very
problems I have pointed to: that is, giving the resources to people
does not always bring differently situated people up to the same level
of functioning. If you have a group of people who are traditionally
marginalized, you are probably going to have to expend more
resources on them to get them up to the same living standard-in
capability terms-than you would for a group of people who are in a
favorable social situation.
Analyzing economic and material rights in terms of capabilities
would thus enable us to understand, as we might not otherwise, a
rationale we might have for spending unequal amounts of money on the
disadvantaged, or creating special programs to assist their transition to
full capability. The Indian government has long done this. Indeed,
affirmative action in this sense for formerly despised caste and tribal
groups was written into the Constitution itself, and it has played a
crucial role in creating the situation we have today, in which
lowercaste parties form part of the ruling government coalition. Indeed,
one could also argue that even to secure political rights effectively to
the lower castes required this type of affirmative action. If we think of
these economic rights asking the question-"What are people actually
able to do and to be?"-then I think we have a better way of
understanding what it is really to put people securely in possession of those
rights, to make them able really to function in those ways, not just to
have the right on paper.
If we have the language of capabilities, do we still need, as well, the
language of rights? The language of rights still plays, I believe, four
important roles in public discourse, despite its unsatisfactory features.
When used in the first way, as in the sentence "A has a right to have
the basic political liberties secured to her by her government," rights
language reminds us that people have justified and urgent claims to
certain types of urgent treatment, no matter what the world around
them has done about that. I have suggested that this role of rights
language lies very close to what I have called "basic capabilities," in
the sense that the justification for saying that people have such natural
rights usually proceeds by pointing to some capability-like feature of
persons that they actually have, on at least a rudimentary level, no
matter what the world around them has done about that. And I
actually think that without such a justification the appeal to rights is quite
mysterious. On the other hand, there is no doubt that one might
recognize the basic capabilities of people and yet still deny that this
entails that they have rights, in the sense of justified claims, to certain
types of treatment. We know that this inference has not been made
through a great deal of the world's history, though it is false to
suppose that it only was made in the West, or that it only began in the
Enlightenment.87 So, appealing to rights communicates more than
appealing to basic capabilities: it says what normative conclusions we
draw from the fact of the basic capabilities.
Even at the second level, when we are talking about rights
guaranteed by the state, the language of rights places great emphasis on the
importance and the basic role of these things. To say, "Here's a list of
things that people ought to be able to do and to be" has only a vague
normative resonance. To say, "Here is a list of fundamental rights,"
means considerably more. It tells people right away that we are
dealing with an especially urgent set of functions, backed up by a sense of
the justified claim that all humans have to such things, by virtue of
Third, rights language has value because of the emphasis it places
on people's choice and autonomy. The language of capabilities, as I
have said, was designed to leave room for choice, and to communicate
the idea that there is a big difference between pushing people into
functioning in ways you consider valuable and leaving the choice up to
them. At the same time, if we have the language of rights in play as
well, I think it helps us to lay extra emphasis on this very important
fact: that what one ought to think of as the benchmark are people's
autonomous choices to avail themselves of certain opportunities, and
not simply their actual functionings.
Finally, in the areas where there is disagreement about the proper
analysis of right talk-where the claims of utility, resources, and
capabilities are still being worked out-the language of rights preserves a
sense of the terrain of agreement, while we continue to deliberate
about the proper type of analysis at the more specific level.
One further point should be made. I have discussed one particular
view about human capabilities and functioning, my own, and I have
indicated its relationship to Sen's very similar view. But of course
there are many other ways in which one might construct a view based
on the idea of human functioning and capability without bringing
capabilities nearly so close to rights. As I have suggested, the view Sen
and I share is a liberal view of human capabilities, which gives a strong
priority to traditional political and religious liberties, and which
fo87. On Indian discussions of religious pluralism and liberty, see Amartya Sen,
Human Rights and Asian Values, New Republic, July 14 & 21, 1997, at 33-40. For
related discussion of Indian conceptions of pluralism, see Amartya Sen, Tagore and
His India, N.Y. Rev. Books, June 26, 1997, at 55-56. On the Greek and Roman
origins of ideas of human rights, see Fred D. Miller, Jr., Nature, Justice, and Rights, in
Aristotle's Politics (1995), arguing that Aristotle's political theory contains the basic
ingredients of a theory of rights; Nature, Function, and Capability,supra note 11,
arguing that Aristotle's political theory contains the view that the job of politics is to
distribute to citizens the things that they need for a flourishing life; Kant and Stoic
Cosmopolitanism,supra note 86, arguing that Kant's view of basic human rights is in
many ways indebted to the views of the Greek and Roman Stoics.
cuses on capability as the goal precisely in order to leave room for
choice. In addition, as I have more recently stressed, the items on my
list of basic capablities are to be regarded as the objects of a
specifically political consensus, rather like a Rawlsian list of primary goods,
and not as a comprehensive conception of the good.
A capabilities theorist might construct a view that departed from
our view in all of these ways. First, the content of the list might be
different: it might not give the same importance to the traditional
liberal freedoms. Second, government might be given much more
latitude to shoot directly for functioning as a goal, and to penalize people
who do not exhibit the desired mode of functioning. Such, indeed, is
the strategy of some natural-law thinkers in the Catholic tradition, and
in this regard they are closer to Aristotle himself than I am.' In that
sense, as I have written, they construe the account of the human good
as a source of public discipline on the choices of citizens, whereas we
construe the good as an account of freedoms citizens have to pursue a
variety of different plans of life. Finally, one might think of the
account of human functioning as a comprehensive conception of human
flourishing for both public and private purposes, rather than as the
object of a specifically political consensus. Again, natural law
theorists sometimes understand the view this way, as does Aristotle
himself-although some Catholic thinkers have themselves adopted a
political-liberal interpretation of their tradition.' Insofar as any of
these alternatives are pursued, the relationship between capabilities
and rights will shift accordingly.
RIGHTS AS GOALS AND SIDE-CONsTRAINrs
One final question remains to be discussed. Sen has argued that
thinking of rights in terms of capabilities should lead us to opt for a
particular way of thinking about rights and to reject another way.
Specifically, it should encourage us to think of rights as goals, and thus
as part of a more general account of social goals that it is reasonable
to promote, rather than to think of them as "side-constraints," or as
justified claims of individuals that should be respected no matter what,
and that thus constrain the ways in which we may promote our social
goals.90 Since Sen's target here is the libertarian theory of Robert
Nozick, and since I believe his critique has force primarily ad
hominem against Nozick, and not against all versions of a
side-constraints view, I must describe Nozick's position.
88. See Finnis,supra note 25; George, supra note 25. For a detailed discussion of
differences between the Sen/Nussbaum view and those views in a range of areas of
public policy, see The Good as Discipline, The Good as Freedom, supra note 12.
89. For an eloquent example, see Jacques Maritain, Truth and Human Fellowship,
in On the Use of Philosophy: Three Essays 16, 24-29 (1961).
90. Rights and Capabilities,supra note 16; Rights as Goals, supra note 6; Rights
and Agency, supra note 16.
Nozick's basic argument, in Anarchy, State, and Utopia9,1 is that
people have rights, in the sense-apparently, since no account of
rights is presented-that these rights should not be overridden for the
sake of the greater good. The rights people have are a function of
their initial entitlements, together with a theory of just transfer. One of
the notoriously frustrating aspects of Nozick's theory is that he refuses
to present his own account of initial entitlements, although he alludes
to a controversial interpretation of Locke, in order to illustrate the
type of thing he has in mind. Through this process, he derives the
view-which must be advanced tentatively, since the account of initial
entitlement has not been given-that people have a right to the
property they hold, just in case they acquired it by a series of just transfers
from the original owners. It is wrong of the state to take any of this
property away from them for redistributive purposes. Nozick focuses
on property throughout the book, and says little about political,
religious, and artistic liberty.
Nozick's theory has been criticized in a number of ways. First of all,
in the absence of a theory of initial entitlement, it is very difficult to
see what the upshot will be, and thus impossible to know whether a
procedural conception of justice like Nozick's will produce results that
are acceptable or quite bizarre and unacceptable. And of course one
might answer questions about entitlement very differently from the
way in which Nozick seems inclined to answer them, saying, for
example, that individuals are never entitled to any property they do not
need for their own use, or that they are never entitled to accumulate a
surplus. Such, for example, was Aristotle's view of entitlement, and
this meant that for Aristotle the very existence of private ownership of
land was a highly dubious business.' In Aristotle's ideal city, fully
half of the land is publicly owned, and the rest is "common in use,"
meaning its produce can be taken by anyone who is in need.93 So
Aristotle's view of entitlement, combined with his strong moral
distaste for hoarding and accumulation, would certainly not yield the
Nozickian conclusion that: "Capitalist acts between consenting adults
are no crime."
Second, it has been pointed out that even if individuals do have
entitlements to what they have acquired in a just transfer, it does not
follow that they are entitled to the surplus value of these goods, when
for contingent reasons they rise in value during the time they hold
them. In fact, even the Lockean tradition is much divided on this
91. Supra note 6.
92. See AristotelianSocial Democracy, supra note 12, at 203-06, 231-32.
93. See id. at 205.
94. Barbara Fried, Wilt ChamberlainRevisited: Nozick's "Justice in Transfer" and
the Problem of Market-Based Distribution,24 Phil. & Pub. Aff. 226 (1995).
Third, one might point out that the economic inequalities
apparently tolerated in Nozick's minimal state would erode the meaningful
possession of other rights that Nozick apparently thinks people have,
such as the right to political participation. Nozick nowhere
confronted possible tensions between two parts of his libertarian view, so
we do not even know whether he would be willing to tax people in
order to get the money to support the institutions that make
meaningful political and religious liberties for all a social reality. In these
ways, his attitude toward rights remained obscure.
Fourth, the view of self-ownership on which much of Nozick's
argument rested was both rather obscure and somewhat questionable.' 5
What does it mean to say of people that they own themselves, and
how, precisely, does and should this affect arguments on a variety of
topics, from the morality of slavery to the legality of prostitution?
These are only some of the ways in which one might criticize
Nozick's view. Let me now describe Sen's critique. Sen argues that if
we allow rights to function the way Nozick says they should, as
"sideconstraints" that can almost never be overridden for the sake of the
general good, then we will be led to tolerate an unacceptable level of
The question I am asking is this: if results such as starvation and
famines were to occur, would the distribution of holdings still be
morally acceptable despite their disastrous consequences? There is
something deeply implausible in the affirmative answer. Why
should it be the case that rules of ownership, etc., should have such
absolute priority over life-and-death questions?
.*. But once it is admitted that consequences can be important in
judging what rights we do or do not morally have, surely the door is
quite open for taking a less narrow view of rights, rejecting
assessment by procedures only.96
Sen seems to be saying two things not easily made compatible. First,
that Nozick has given the wrong account of what rights people have:
they do not have the right to keep their surplus when others are dying.
Second, that the consideration of consequences shows that the type of
view of rightsNozick advances must be wrong: a side-constraints view
is implausible, and we should think of rights as parts of a total system
of social goals. But if the first point is correct, as I believe it certainly
is, then we have had as yet no reason to accept the second claim. If
we question the whole way Nozick thinks about what people's rights
and entitlements are, as we most certainly should, then we have no
reason to think that a correct list of rights should not be used as
95. G.A. Cohen, Self-Ownership, Freedom, and Equality 1-115 (1995).
96. Rights and Capabilities,supra note 16, at 312.
This realization is important, since a list of human rights typically
functions as a system of side-constraints in international deliberation
and in internal policy debates. That is, we typically say to and of
governments, let them pursue the social good as they conceive it, so long
as they do not violate the items on this list. I think this is a very good
way of thinking about the way a list of basic human rights should
function in a pluralistic society, and I have already said that I regard my
list of basic capabilities this way, as a list of very urgent items that
should be secured to people no matter what else we pursue. In this
way, we are both conceiving of capabilities as a set of goals-a subset
of total social goals-and saying that they have an urgent claim to be
promoted, whatever else we also promote. Indeed, the point made by
Sen, in endorsing the Rawlsian notion of the priority of liberty, was
precisely this.97 We are doing wrong to people when we do not secure
to them the capabilities on this list. The traditional function of a
notion of rights as side-constraints is to make this sort of anti-utilitarian
point, and I see no reason why rights construed as capabilities-or
analyzed in terms of capabilities-should not continue to play this
Of course there will be circumstances in which we cannot secure to
all the citizens the capabilities on my list. Sen and I have argued that
the political liberties and liberties of conscience should get a high
degree of priority within the general capability set.98 But we also
conceive of the capabilities as a total system of liberty, whose parts
support one another. Thus we also hold that there is something very
bad about not securing any of the items. The precise threshold level
for many of them remains to be hammered out in public debate; but
there are surely levels easy to specify, beneath which people will have
been violated in unacceptable ways if the capabilities are not secured.
Viewing capabilities as rather like side-constraints also helps here: for
it helps us to understand what is tragic and unacceptable in such
situations, and why individuals so treated have an urgent claim to be
treated better, even when governments are in other ways pursuing the
good with great efficiency.
97. See Freedoms and Needs, supranote 18, at 32 (defending the Rawlsian priority
98. See Religion and Women's Human Rights, supra note 69, at 113-14 (religious
liberty); Freedoms and Needs, supra note 18, at 32-38; The Good as Discipline, The
Good as Freedom, supra note 12, at 314-21 (defending the general liberal approach);
id. at 332-33 (political liberty).
18. See John Rawls, Political Liberalism ( 1993 ) [hereinafter Political Liberalism] ; John Raws , A Theory of Justice ( 1971 ) [hereinafter A Theory of Justice]. Sen discusses, and supports, the Rawlsian notion of the priority of liberty in Freedoms and Needs , New Republic, Jan. 10 & 17 , 1994 , at 31-38 [hereinafter Freedoms and Needs]. I discuss the relationship between my own version of the capabilities view and Rawls's theory in Aristotelian Social Democracy, supra note 12, and The Good as Discipline, The Good as Freedom, supra note 12. In The Good as Discipline,The Good as Freedom, I emphasize the liberal roots of my own Aristotelianism, contrasting my view with two non-liberal forms of Aristotelianism.
19. This is my term from Aristotelian Social Democracy , supra note 12 , at 217, contrasting with Rawls's "thin theory of the good." A Theory of Justice , supra note 18, at 395-99.
20. See Human Capabilities, supra note 12 , at 85.
21. Id at 84.
22. Id at 84-85.
23. Women and Cultural Universals, supra note 12 , at 25-26.
24. For the close relationship between the capabilities approach and Enlightenment liberalism, see Freedoms and Needs, supra note 18, and The Good as Discipline, The Good as Freedom , supra note 12.
25. I put things this way because the most prominent anti-liberal natural law theorists do not explicitly reject rights language, see John Finnis, Natural Law and Natural Rights ( 1980 ), and Robert P. George , Making Men Moral: Civil Liberties and Public Morality ( 1993 ), and the most prominent Catholic opponent of rights language does not endorse the capabilities approach , see Mary Ann Glendon, Rights Talk: The Impoverishment of Political Discourse (1991), but the combination is easy enough to imagine .
26. See Rights and Capabilities,supra note 16, at 310-12.
27. A valuable beginning, bringing together all that Sen and I have said on the topic, is in Functioningand Capabilities: Part2 , supra note 14, at 186-91.
35. For discussion of this approach, see Equality of What? , supra note 7 , at 358-64.
36. See A Theory of Justice , supra note 18; Amartya Sen & Bernard Williams , Introduction to Utilitarianism and Beyond 1 , 4 - 5 ( Amartya Sen & Bernard Williams eds., 1982 ) (arguing that utilitarianism views persons simply as locations of their respective utilities).
37. See A Theory of Justice, supra note 18, at 179-83 (arguing that utilitarianism treats people as means, rather than as ends).
38. Id . at 207 (arguing that it is unacceptable to take chances with basic liberties).
39. Here I am combining arguments from Williams's essay, A Critique of Utilitarianism , in Utilitarianism: For and Against 77 ( 1973 ), arguing that utilitarianism cannot give an adequate account of a person's special connection to his or her own. actions, and therefore of personal integrity, with his Persons, Characterand Morality, in The Identities of Persons 197 (Amdlie 0 . Rorty ed., 1969 ), arguing that the separateness of persons is a central fact of ethical life .
40. See the discussion in Martha C. Nussbaum, Plato on Commensurability and Desire, in Love's Knowledge 106 ( 1990 ) [hereinafter Commensurability and Desire], and Martha C. Nussbaum, The Discernment of Perception: An Aristotelian Conception of Privateand PublicRationality, in Love's Knowledge, supra , at 54 [ hereinafter Discernmentof Perception],arguing that the plurality and distinctness of the valuable things in life make any single metric a damaging distortion .
41. See The Discernmentof Perception , supra note 40; Plato on Commensurability and Desire ,supra note 40; see also Martha C. Nussbaum, The Fragility of Goodness 290-317 ( 1986 ) (arguing that Aristotle was right to recognize a type of deliberation that does not rely on a single metric); Amartya Sen , On Ethics and Economics 62 - 63 ( 1987 ) [hereinafter On Ethics and Economics] (discussing plurality and non-commensurability); Amartya Sen , PluralUtility, 81 Proc. Aristotelian Soc'y 193 ( 1981 ) [hereinafter Plural Utility] (arguing that the right way to think of utility is as a plurality of vectors).
42. See Human Capabilities, supra note 12 , at 85-86; On Ethics and Economics, supra note 41, at 63-64.
43. Rights and Capabilities,supra note 16, at 309.
44. Id .
45. Id .
46. Id .
47. Id .
56. Sen has insisted, however, that happiness is "a momentous functioning," in Well-Being, supranote 15, at 200, and I have insisted that emotional functioning is one of the important types of functioning we should consider . See Martha C. Nussbaum, Emotions and Women's Capabilities ,in Women, Culture, and Development, supra note 12, at 360.
57. See Human Capabilities, supra note 12 , at 85-86; Plural Utility,supra note 41.
58. In this sense, the approach takes its inspiration from Marx's discussion of fully human functioning in several early works in which he was in turn much influenced by Aristotle. For discussion of these links, see Hnman Nature ,supra note 12, at 119-20.
59. See supra notes 9-10 and accompanying text.
60. See , e.g., Jean Drize & Amartya Sen , India: Economic Development and Social Opportunity 13-16 , 109 - 39 ( 1995 ) (discussing the relationship between health and education and other capabilities); id . at 155-178 (discussing the relationship between gender inequality and women's functioning and capability). For an enumeration of all the examples Sen has given in a variety of different works, see Functioningand Capability: Part1, supra note 14, and Functioning and Capability: Part2, supra note 14.
61. This is especially evident in Human Nature , supra note 12 , at 90-95.
68. See Human Nature, supra note 12 , at 119-20 (discussing Marx).
69. See Martha C. Nussbaum , Religion and Women's Hnnan Rights , in Religion and Contemporary Liberalism 93 , 107 - 10 (Paul J. Weithman ed., 1997 ) [hereinafter Religion and Women's Human Rights]; Martha Nussbaum , Double MoralStandards?, Boston Rev., Oct .- Nov . 1996 , at 28 , 30 (replying to Yael Tamir's Hands Off Clitoridectomy , Boston Rev., Summer 1996 , at 21- 22 ).
70. See Human Capabilities, supra note 12 , at 88 (discussing the basic capabilities); Nature , Function, and Capability,supra note 11, at 160-64 ( referring to Aristotle's similar distinctions) . Sen does not use these three levels explicitly, although many things he says assume some such distinctions .