Why the Capability Approach?
Sabina Alkire is a Research Associate at the Global Equity Initiative,
Abstract In addressing operational challenges such as poverty or
economic development, many researchers and practitioners wish to
build upon insights raised by Sen’s capability approach and related
writings. This paper argues that the comprehensive reach and foundation
of the human development and capability approach has a value
independent from and additional to their practical outworkings, and yet
also that operational specifications are both possible and vital to the
further development of the approach. The paper begins with a thumbnail
sketch of the core concepts of the capability approach, and supplements
these with additional informational and principle requirements that Sen
argues to be necessary for a more complete assessment of a state of affairs.
It traces some important avenues along which the Human Development
Reports and other empirical studies have operationalized certain aspects
of Sen’s capability approach. The paper then articulates further
developments that might be expected, arguing that such developments
must also build upon cutting edge research in other fields. It also identifies
certain ‘value judgments’ that are inherent to the capability approach and
should not be permanently dismissed by some methodological innovation.
Key words: Amartya Sen, Capabilities, Capability approach, Development
economics, Human development, Poverty, Well-being
Many ask how or whether Sen’s capability approach can be ‘operationa-
lized’ or put into practice.
However cumbersome the wording may seem,
this is a vital and pertinent question upon which many have written.
may not be the only grounds on which to consider the contribution of the
capability approach and, relatedly, of human development.
In an article on the revolutions that occur within economics, John
Hicks acknowledges economists’ need for focus: ‘‘In order that we should
be able to say useful things about what is happening, before it is too late,
we must select, even select quite violently. We must concentrate our
attention, and hope that we have concentrated it in the right place. We
Journal of Human Development
Vol. 6, No. 1, March 2005
ISSN 1464-9888 print/ISSN 1469-9516 online/05/010115-19 #2005 United Nations Development Programme
must work, if we are to work effectively, in some sort of blinkers’’ (1983,
p. 4). Clearly economists need to identify shorthands — theorems,
indicators, and other formal or ‘operational’ tools — to specify broad
ideas. Yet economic revolutions, Hicks argues, emerge when the focal area
of concentration itself shifts. Hicks observes that one way such shifts occur
within economics is ‘‘by generalization, by constructing ‘more general’
theories, theories which put more things into their places, even if we can
do less with them when we have put them there …’’ (1983, p. 6).
This article argues that Sen’s capability and the human development
approach together represent a ‘more general’ approach to the problems
that economics and development (together with other disciplines)
address, and that this has a distinctive value apart from the practical
contributions and empirical outworkings of the approaches — as Hicks
At the same time, even if we acknowledge the considerable value of
the ‘more general’ framework, the pragmatic and insistent questions about
how to use the approach in different contexts are still well worth asking for
a number of reasons. The most evident, of course, is that such tools can be
of direct value to the objectives at hand. A further reason relates to the
political economy of ideas: theories that are not user-friendly do not
spread. Consider the historical trajectory of the basic needs school in the
1970s, when a possibly similar confluence of international vision and
energy for poverty reduction existed to the one at present.
needs approach defended human development very much along the lines
of the then-nascent capability approach (Streeten et al., 1981, pp. 33–34;
Stewart, 1985, chapters 1–2). But its vision was complicated.
Paul Streeten (1984) published a short article that identified
‘unanswered questions’ of the basic needs approach: who defines needs;
whether the goal was ‘human flourishing’ or ‘meeting basic needs’; where
participation fit in; which needs institutions could legitimately plan to
meet; and how to coordinate international funding to meet basic needs.
But in the meantime, before those questions had been adequately
addressed, while the research and discussion was underway, operational
programs run by the World Bank and International Labour Organization
(ILO) among others hastily implemented ‘answers’. They focused on
commodity inputs to health, education, clothing, shelter, sanitation and
hygiene — because it was relatively cheap and easy to measure these. The
problem was that the overemphasis on commodities misinterpreted the
basic needs approach, and in so doing redefined and subverted it.
Unless user-friendly operational procedures arise and spread, the interest
in the capability approach is likewise vulnerable to subversion by misinter-
pretation. To give just one example, the initial World Bank documents
outlining poverty reduction strategies used the term ‘capabilities’ to mean
‘health and education’. Freedom had vanished. A vector of functionings that
included the ability to walk around without shame had vanished, all in the
haste to imbue a ‘popular’ term with easily operational content.
In an effort to give both the ‘general’ and the more ‘operational’
issues their due, this paper proceeds as follows. The first part addresses
the definitional issues and range of the capability approach that undergird
its general value. The second part addresses pragmatic issues. Rather than
surveying the secondary literature to date, this second part discusses in
quite ‘a rough and ready way’ what Keynes might have termed the ‘art’ of
economics: the methods by which appropriate applications of a theory,
informed by empirical evidence, can be assembled and revised (Keynes,
1891; see also Colander, 2004).
Ends and means
A fundamental strength, the capability approach is clarity about the
objective. This insight can be stated briefly: according to the capability
approach, the objective of both justice and poverty reduction (for
example) should be to expand the freedom that deprived people have
to enjoy ‘valuable beings and doings’.
They should have access to the
necessary positive resources, and they should be able to make choices that
matter to them. The key excitement about the capability approach is that it
goes beyond the relentless criticism of income to propose an alternative
space in which to conceptualize both poverty reduction and justice. This
space includes multiple functionings, and freedoms. The hope is that
further elaboration of this objective will build into an alternative paradigm,
an alternative way of identifying and evaluating intermediary actions
(including for example growth, social investment, and participation) that
might contribute to the objective (expanding valuable capabilities).
To set this insight into context, recall the fundamental shift of both
the capability approach and human development with respect to standard
economic approaches. Both argue that human beings and their flourish-
ing, rather than an increase in economic growth, should be the ‘end’ or
objective of development. Sen’s paper ‘The Concept of Development’
stated what seems to be his enduring position: that welfare economics
and development should not vary at all in so far as their ‘objective’ is
concerned. ‘‘The enhancement of living conditions must clearly be an
essential — if not the essential — object of the entire economic exercise
and that enhancement is an integral part of the concept of development’’
(Sen, 1988, p. 11). The status of human beings as ‘ends’ of development
must be reiterated, Sen argues, because human beings ‘‘also happen to be
— directly or indirectly — the primary means of all production’’ (1990a,
There is broad agreement in the pencil sketch of the capability
approach earlier, whether it is applied in welfare economics or
development or other disciplinary frameworks such as health economics.
In defining the capability approach further in this article I will rely on
more theoretical or philosophical explications of it — in Inequality
Re-examined (Sen, 1992), and ‘Wellbeing, Agency, and Freedom’ (Sen,
Why the Capability Approach?
1985a), for example (see also Sen, 1985b, 1987b, 1988, 1900a, 1993a,
1995, 1997c, 1999a). Others consider applied writings — including the
background papers for the United Nations’ Human Development Reports
(Dre`ze and Sen, 1989, 1990, 1995, 2002; Ahmad et al., 1991; Anand and
Sen, 1994, 1997, 2000a, 2000b). These writings serve to illustrate in many
ways the distinctiveness of the capability approach when it is applied to
poverty analysis, and have sparked a great number of refinements and
developments. But as Sen as well as Jean Dre`ze signal, ‘operational’
analyses are also appropriately influenced by other matters, such as the
availability of data, and the policy appeal of results to a particular audience
or at a particular point in time. It would seem unnecessarily limiting and
error-prone to extrapolate the capability approach from particular,
context-specific applications of it, especially when well-developed theore-
tical sources are available. So on definitional matters of the capability
approach I will use the more philosophical texts. To complement
Robeyns’ and Deneulin’s papers in this issue of the journal, I quickly
rehearse the approach’s central concepts without belaboring points well
covered in their papers.
Formulations of capability have two parts: valuable beings and
doings (functionings), and freedom. Sen’s significant contribution has
been to unite the two concepts, and any account of capabilities that does
not include both misrepresents this approach. The two component parts
are described in the following.
Functionings represent multiple diverse aspects of life that people value.
Sen argues that functionings — that is ‘‘the various things a person may
value doing or being’’ (1999a, p. 75) — taken together create a better
conceptual space in which to assess social welfare than utility or opulence.
Functionings are ‘beings and doings’, such as being nourished, being
confident, being able to travel, or taking part in political decisions. The
word is of Aristotelian origin and, like Aristotle, this approach claims,
significantly, that ‘‘functionings are constitutive of a person’s being’’.
For many readers, the capability approach seems to be curtained by its
vocabulary — and in particular by non-intuitive phrases such as
‘functionings’ or ‘beings and doings’. This need not be the case. When a
graduate of a Friere-style phonetic literacy course in Lahore, Pakistan
considered what the literacy class had taught her, she eloquently described
a functioning — in this case ‘‘trusting one’s own judgment’’: ‘‘Women
think they are like a flower bud — that they do not understand with their
own eyes. But we are not buds, we are mountains. We can do anything
with our lives. So I tried to open my eyes, and my eyes were opened’’.
Another adult graduate and mother valued several functionings: ‘know-
ledge’ and the better ‘health’ and ‘confidence’ she anticipated the
knowledge would bring. She said, ‘‘We studied the word ‘food’ in class.
We knew that we are poor, we can not drink milk, eat many foods; that we
eat little meat. We learned that it was not necessary … Chick peas are 4
rupees a pound, and they have many vitamins. Apples are expensive;
carrots are not. But carrots are good for health — as good as expensive
things ….’’ Graduates also valued their ‘friendships’ with each other —
and these are another functioning. So functionings are actually very
familiar and intuitive parts of life. The term may seem a bit odd, but it
covers the very different activities and situations people recognize to be
important (Alkire, 2002, chapter 7).
So when a poverty reduction activity undertakes to evaluate a group
of persons’ well-being (in the course, perhaps, of assessing their quality of
life, standard of living, social welfare, or level of poverty), the capability
approach would argue that it must have in view their functionings. Not all
functionings will be relevant to every evaluation. The identification of what
people value, the selection of which priority functionings a particular
poverty reduction initiative should aim to expand, and the actual
expansions that are to be evaluated (which may be wider than the priority
functionings), are each separate questions. Sen does not identify one set
of basic functionings (or basic capabilities
) precisely because no one set
will do for every evaluation.
This intentional breadth allows the approach to be relevant to a wide
variety of circumstances. Temporarily, in an article called ‘The Living
Standard’, Sen (1984) had suggested that one should ‘‘separate ‘material’
functionings and capabilities (e.g., to be well-nourished) from others (e.g.,
being wise and contented)’’ and evaluate standards of living with reference
to material capabilities. But later Sen reversed this, and suggested instead
that considerations of living standard encompass all valued functionings.
‘‘It is possible that this way of drawing the line is a little too permissive, but
the alternatives that have been proposed seem clearly too narrow’’(Sen,
1987b, p. 27).
So the capability approach, fully developed, could
appreciate all changes in a person’s quality of life: from knowledge to
relationships to employment opportunities and inner peace, to self-
confidence and the various valued activities made possible by the literacy
classes. None of these changes are ruled out as irrelevant at all times and
places. One can thus analyze the capabilities of a rich as well as a poor
person or country, and analyze basic as well as complex capabilities.
Herewecanbegintoseeoneofthepractical benefits of analytical clarity
about ends and means: its ability to raise issues requiring value judgments.
For example, if an ‘evaluation’ only takes into account the reading level and
income or employment rates generated by a literacy program, the participants
may point out that such an ‘evaluation’ does not adequately capture their
experience in which cultural or language-based or relational functionings that
they deeply valued expanded (or contracted). If the activity were to be
evaluated with respect to capability expansion, such information from
participants should provoke reconsideration of the evaluative framework to
take into account the expansion or contraction of other valuable and high
Why the Capability Approach?
priority capabilities. Of course, richer information on people’s capability sets
would still not be sufficient to make a decision. Pragmatic considerations of
decision-making as well as procedural issues could also be required, as Sen
points out (see also ‘Principle pluralism’ later).
A capability framework points towards a space, but even when
the functionings have been selected and some ranges of weights
specified, there will be the possibility that a particular policy A
would be better for equality, while policy B is better for aggregate
achievement. These trade-offs are also part of public judgment,
and what this exercise requires is … the identification of relevant
considerations, suggesting particular proposals and encouraging
public discussion on those considerations and proposals. (1996a,
The discussion of the appropriate objectives for development interven-
tions could not occur unless the ‘objective’ of the activity was supposed to
be related to what people value and have reason to value (rather than
some mechanical intermediary variable such as growth in income or
number of graduates). Thus, such discussions are a practical ‘benefit’ of
the general capability framework.
The focus on functionings sets the capability approach off from other
approaches to the evaluation of well-being. Sen acknowledges that these
alternative approaches to well-being are relevant. In particular, he
acknowledges that the work of others who have tried to correct the
shortcomings of utilitarian or commodity-focused approaches has ongoing
relevance even if these approaches themselves are not fully adopted (Sen,
1985b, p. 24 and references cited therein). Also, both utility and
commodities can be used as proxies of individual advantage in some
situations. And certainly the capability approach further develops the
trajectory that Rawls (1971, see also 1993) so ably initiated. But Sen’s claim
is that alternative approaches fail to provide an adequate conceptual basis
for comparisons of well-being, and do not provide a sufficient basis for
social evaluation. Functionings, taken together with freedom (to be
discussed next), he argues, provide such a basis.
A person’s achieved functionings at any given time are the particular
functionings he/she has successfully pursued and realized. But in assessing
human development, a focus on achieved functionings alone, like a focus
on utility, is incomplete. It does not necessarily incorporate the freedom to
decide which path to take, or the freedom to bring about achievements
one considers to be valuable, whether or not these achievements are
connected to one’s own well-being or not (reducing national carbon
emissions, for example).
Sen argues that such freedoms have intrinsic as
well as instrumental value. ‘‘The ‘good life’ is partly a life of genuine
choice, and not one in which the person is forced into a particular life —
however rich it might be in other respects’’ (Sen, 1996a, p. 59). The
intrinsic value of freedom does, Sen argues, pertain across classes and
cultures. ‘‘The popular appeal of many social movements in India confirms
that this basic capability is highly valued even among people who lead very
deprived lives in material terms’’ (Dre`ze and Sen, 1995, p. 106; see also
In order to attend to the intrinsic and foundational importance of
freedom, Sen introduces the concept of capability (as well as agency, as we
shall see). Capability refers to a person or group’s freedom to promote or
achieve valuable functionings. ‘‘It represents the various combinations of
functionings (beings and doings) that the person can achieve. Capability
is, thus, a set of vectors of functionings, reflecting the person’s freedom to
lead one type of life or another … to choose from possible livings’’ (Sen,
1992, p. 40). Capability is a budget set; it is a set of real opportunities that
you could use in one way or another, the paths that lie open before you.
A number of clarifications are in order that might circumvent common
misunderstandings of capability. First, in the capability approach, freedom
concerns ‘‘the real opportunity that we have to accomplish what we
value’’ (Sen, 1992, p. 31; see 1999a, p. 74). It does not, therefore, include
freedoms or opportunities that a person might hold theoretically or legally
but that, in reality, lie well beyond their reach.
Second, capabilities are, by definition, limited to functionings of
value; they exclude evil or harmful functionings.
It would be possible to
overlook a vital phrase in Sen’s capability approach: ‘to accomplish what
we value’. Without qualification, the apparent prominence of freedom in
the capability account (and the image of a capability set as a budget set
from which one chooses) would be open to the criticism that freedom of
choice is of more importance in some societies than others; to some
people more than others. But the prominence is qualified: Sen argues that
increases in choices per se do not necessarily lead to an increase in
freedom. There are two reasons for this. First, the options added may not
be ones we value anyway. And second, however valuable or not options
may be, an increase in ‘freedom of choice’ may crowd out our ability to live
‘‘a peaceful and unbothered life’’ (Sen, 1992, p. 63). Sen writes: ‘‘Indeed
sometimes more freedom of choice can bemuse and befuddle, and make
one’s life more wretched’’ (1992, p. 59).
A third often overlooked distinction is freedom versus ‘control’ (Sen,
1982). Freedom includes ‘‘a person’s ability to get systematically what he
would choose no matter who actually controls the levers of operation’’
(Sen, 1992, p. 65). For example, if, given the choice, we would choose to
work in a smoke-free environment, then ceteris paribus a public program
to prohibit smoking in shared work areas does indeed enhance our
freedom, even if we were not asked directly about this matter, because in
the absence of this public program we would not have the effective
freedom to work in a smoke-free environment. This is the case even
Why the Capability Approach?
though the ‘number of alternatives’ we have to choose between does not
increase (in fact we lose the freedom to smoke). Clearly, often what is
important actually is who has the levers of control. But Sen points out that
direct control is not the only expression of freedom, although it has often
been mistaken as such.
A final point relates to the internal plurality of capability space. The
capability approach notes that individual advantage can be assessed in at
least four different spaces: well-being achievement,well-being freedom,
agency achievement,oragency freedom. Individual advantage can be
assessed in relation to one’s well-being whether defined in an elementary
fashion (nutritional status) or in a more complex manner (self-esteem). Or
it can relate to agency — one’s ability to pursue goals that one values
(getting funding for a new school, promoting the protection of rare
seabirds). In either case, advantage can refer to the well-being or agency
achievements, or to well-being and agency freedom. Sen argues that we
cannot simply choose to focus on one or another of these four possible
spaces and ignore the rest; there are good arguments for keeping all in
mind. He argues this while accepting that these objectives may conflict.
For instance, if your riverside picnic is interrupted by the chance to
rescue someone from drowning, then your agency freedom (and hopefully
achievement) increases, because you can save someone’s life; but
your achieved well-being diminishes, as you emerge cold wet and hungry.
Sen’s Arrow lectures further develop quite extensive accounts of the
‘‘opportunity’’ and ‘‘process’’ aspects of freedom (Sen, 2002, chapters
This means that when Sen advocates that social arrangements should
be evaluated with respect to ‘‘freedom’’ (1992, p. 129; see 1993a, p. 49;
2002, chapters 19–21), he is advocating equality in a ‘space’ that has quite
a substantial degree of internal plurality. It includes a medley of things like
the literacy instructor’s freedom to be an agent of social change in Lahore,
and the students’ capability to be nourished, and women’s capability to
read and act on their own behalf.
The definition of capability, then, combines functioning and freedom.
But the capability approach is a proposition, and the proposition is this:
that social arrangements should be evaluated according to the extent of
freedom people have to promote or achieve functionings they value. If
equality in social arrangements is to be demanded in any space — and
most theories of justice advocate equality in some space — it is to be
demanded in the space of capabilities (Sen, 1992, 1996b).
After exposing the undesirably narrow effect of utilitarianism because of
the information it excludes, Sen argues for an approach to welfare
economics that takes into account additional information of two kinds.
‘‘One is in terms of plurality of principles (I shall call this principle
pluralism), and the other in terms of plurality of informational variables
(to be called information pluralism)’’ (1985a, p. 176).
As is by now evident, the capability approach has considerable reach
and depth, and its informational pluralism is one feature that gives it a
kind of generality that Hicks might have commended. Capabilities may
relate to things near to survival (the capability to drink clean water) or
those that are rather less central (the capability to travel for pleasure, the
capability to read ancient history). The definition of capability does not
delimit a certain subset of capabilities as of peculiar importance; rather,
the selection of capabilities on which to focus is a value judgment (that
also depends partly on the purpose of the evaluation), as is the weighting
of capabilities relative to each other (Sen, 1992, pp. 42–46; 1999a, pp. 76–
85). Clearly, in order to construct pertinent individual capability sets for a
particular evaluative exercise, much less to aggregate and compare
capabilities operationally, we need a great deal of information that will
not be straightforward to obtain.
Sen has written not only on capabilities and functionings, but also,
significantly, on agency,
social choice (Sen, 1998), rationality (Sen,
2002), justice (Sen, 1990b, 1992, 1995), the market (Sen, 1993b),
public action (Dre`ze and Sen, 1989, 2002), economic methodology
(Sen, 1989), consequentialism (Sen, 2000a), human rights (Sen, 2004),
public debate (see Sen, 2005; see also Sen 1999b; Alkire, forthcoming),
cost–benefit analysis (Sen, 2000b), women’s agency (Sen, 1990c),
population (Sen, 1997b), and the institutions by which capabilities are
realized in human communities (Sen, 1999). I would suggest that further
developments of the capability approach should consider these and
related conceptual writings, and should not restrict attention to the bare
definitions of capability and functioning and the proposition already
Many others have chosen to conceive of the capability approach as
being (only) the proposition that justice (for example) is to be evaluated in
the space of capabilities rather than in the space of utility or primary goods
or functionings or the satisfaction of some deontological principle(s).
There are good reasons for focusing on this proposition, probing and
evaluating it as distinctive of the capability approach. But it would seem
rather arbitrary to disregard Sen’s accounts of the processes by which the
proposition can be put into action — especially if one is fundamentally
interested in operationalizing the capability approach. Furthermore, a
number of authors ‘complain’ that the capability approach does not
address questions they put to it — when Sen has actually developed very
clear responses to their very questions in other writings.
Consider for example the other information that may be relevant to the
assessment of affairs.
Taken as a whole, Sen’s writings articulate a number
of kinds of information to be pertinent to the evaluation of social
arrangements. These go well beyond information on people’s capabilities
and agency freedoms. For example, they may include information on
Why the Capability Approach?
people’s current levels of achieved functionings — but also about the ‘menus’
from which they chose (Sen, 1997a). Further information on the exact
circumstances of a choice, which includes a description of individual
responsibility, is also key for what Sen terms ‘situated evaluation’. This might
include, for example, situations of active harm (murder) and situations in
which people omit to do what they should or violate their ‘imperfect
obligations’ to do what they can to help one another (Sen, 2000a).
Not only must the direct and intended consequences of an action be
considered, but also the unintended but foreseeable consequences —
whether these be expansions of capabilities, contractions of capabilities, or
trade-offs (Sen, 1999a; see Alkire, 2004). Sen has also identified ‘human
rights’ as being information that could and should be considered in
evaluating states of affairs. In the Arrow lectures, Sen (2002, p. 624) also
identifies how persons might have concerns both about the processes that
occur in their own lives and also those that occur within their society.
Such a list of possibly relevant information may seem onerous and
unfeasible. However, while such an approach has many degrees of
freedom, concrete situations have far fewer — and the categories of
information that will be relevant to a particular circumstance are almost
sure to be considerably fewer.
Agency and process
In addition to expanding the informational basis of economic choices,
there might be diverse principles by which to evaluate states of affairs —
distributional concerns for the least well-off; efficiency concerns; concerns
for the fairness of the decision-making procedure (was the ‘voice’ of the
marginalized adequately considered?); concerns for the impact of the
decision on future generations; concerns of universalizability; and so on.
Sen argues that no one principle — for example, efficiency maximization
— suffices for normative economic problems. Rather, a plurality, not only
of informational ingredients, but also of combining principles, should be
Welfare economics is a major branch of ‘practical reason’. There
are no good grounds for expecting that the diverse considera-
tions that are characteristic of practical reason, discussed, among
others, by Aristotle, Kant, Smith, Hume, Marx, or Mill, can, in any
real sense, be avoided by taking refuge in some simple formula
like the utilitarian maximization of utility sums, or a general
reliance on optimality, or going by some mechanical criterion of
technical efficiency or maximization of the gross national
product. (Sen, 1996a, p. 61)
And of course broadening the informational basis of welfare considerations,
as well as the combining principles, implies a rather different and more
sophisticated set of assumptions about human beings. For example, given the
capability approach’s view of persons as agents who have diverse valued goals
and commitments on behalf both of themselves and of their society, and who
contribute to public discussion about social goals, the approach cannot
coherently employ an entirely self-interested model of human motivation. A
complex of other motivations, perhaps including identity, cooperation,
altruism,habit,andsympathy,mustalso enter (see Sen, 1997e; Alkire and
Deneulin, 2002). Thus the capability approach has implications also for the
model of homo economicus.
We began with an observation that the fundamental insight of the
capability approach concerns the objective of human development:
namely, that it should not be economic growth as an end-in-itself, but
rather be the expansion of people’s real freedoms to do and be what they
value. However, as is overwhelmingly evident by now, such a change in the
objective has direct implications for the information that is considered and
the conception of rationality that is invoked. Because people’s values are
involved both in the identification of the objective and in the processes by
which it is realized, the capability approach requires ethical rationality as
well as (not instead of!) a more narrow technical or engineering rationality
Capability and welfare economics
Poverty reduction initiatives, development economics, and welfare
economics all address the problem of how to generate and allocate
productive resources to achieve the best social state. Reflections on this
problem can be broken into subcomponents: (i) what kinds of information
are necessary in order to define social states?, (ii) how are more valuable
social states to be distinguished from less valuable?, and (iii) what rules or
principles guide (or constrain) the procedures of attaining/sustaining
The Bergson–Samuelson Welfare Theorem, on which the greater part
of welfare and development economics depends, suggested that social
states are to be measured as sum-rankings of individual ordinal utility, with
greater aggregate sums defining better social states. It also identified one
necessary and sufficient principle: maximization (to maximize aggregate
utility) (Sen, 1996a).
Sen’s capability approach sketches a distinctive ‘general’ overview of
the same landscape. For example, with respect to the aforementioned
subcomponents, the capability approach would argue that: (i) social states
should be defined primarily in the space of human capabilities although
other kinds of information will pertain, (iiA) more valuable states are those
that have ‘expanded’ valuable human capabilities, (iiB) the determination
of which and whose capabilities are valuable and their relative weights
should be subject to explicit scrutiny and public discussion over time, (iiiA)
the single rule of social utility maximization is insufficient, and (iiiB) plural
rules, based on practical reason, apply.
As Sen has well demonstrated,
Why the Capability Approach?
this approach challenges the fundamental basis of welfare economics as
well as its schematic model of rational economic man.
The overview of the capability approach prepared us to hear the questions
that have surged through the literature: How are capabilities to be
measured, compared, and aggregated? How are value conflicts to be
resolved? Which capabilities should be selected for study? As Sugden
(1993, p. 1953) already noted, ‘‘Given the rich array of functionings that
Sen takes to be relevant, given the extent of disagreement among
reasonable people about the nature of the good life, and given the
unresolved problem of how to value sets, it is natural to ask how far Sen’s
framework is operational’’.
Yet it may be equally evident that research over the past 20 years
has already been informed by the capability approach and has taken it
forward, giving rise to a substantial literature.
For example, the Human
Development Reports (United Nations Development Programme, 1990–
2004) have, annually, drawn upon both the core ‘general’ ideas of human
development that overlap with Sen’s capability approach, and also
developed practical policy positions on key issues such as participation,
gender, globalization, and human rights. The reports influenced and
continues to influence policy priorities. Additional streams of academic
research have explored a number of questions raised by the capability
approach in considerable depth, as Robeyns’ article in this issue surveys.
This section addresses the second stream of questions raised in the
Introduction; namely, how to generate additional theoretical and practical
specifications of the capability approach.
Human Development Reports: a first port of call
The Human Development Reports, published annually since 1990,
represent an extensive and sustained effort to translate some core ideas
of the capability approach among other work into accessible language and
operational policy prescriptions. In particular, they have continued to
attend to human beings as the ‘end’ of development and have articulated
some implications of that perspective through their analysis of a
considerable range of topics (see Table 1). They have also given rise to
one particularly prominent operational tool; namely, the Human
Development Index (and related indices of gender and poverty).
Regional and national human development reports have, similarly, tried
to identify and articulate the distinctive analyses that arise from this
perspective. Given the diverse and unconsolidated nature of secondary
literature on the capability approach, greater consideration of, and critical
interaction with, these texts could be of considerable benefit. These
reports could provide a vehicle for communicating key research results to
a wider audience of both researchers and practitioners.
But, to reiterate the cumbersome question that is often asked, how does
one ‘operationalize’ the capability approach? The first observation to make
about the capability approach is that operationalizing it is not a one-time
thing. Some critics seem to be nostalgic for an approach that would
cleanse the capability approach from all of the value choices and provide
an intellectual breakthrough — like finding a cure for AIDS. If that is
the case then researchers are competing teams who are trying to find
the magic missing insight. But many of the residual value judgments in the
capability approach will need to be made on the ground over and over
again. They are not of the sort that are susceptible to a magic missing
insight. That was what Sen means by fundamental or assertive incomplete-
ness. For example, no one ‘list’ of basic capabilities will be relevant to
every evaluation or assessment or measurement exercise or index: the
selection of functionings or capabilities upon which to focus will need to
be done repeatedly. The same might apply to principles.
A second observation is that operationalization needs to occur not
only in many countries, but also at many different levels, and in respect
to different problems. For example, how does the capability approach to
poverty relate to the Millennium Development Goal indicators? How do
capability poverty measures compare with income poverty measures? What
implications does the capability approach have for those who are
constructing databases and designing household questionnaires in the
international institutions so we can track capability deprivation over time?
How does the capability approach change the way that a non-
governmental organization decides to allocate its resources between
alternative poverty reducing activities? How can the capability approach
address foundational issues of practical reason? How can the capability
Table 1. Themes of Human Development Reports to date
2004 Cultural liberty in today’s diverse world
2003 Millennium Development Goals: a compact among nations to end human poverty
2002 Deepening democracy in a fragmented world
2001 Making new technologies work for human development
2000 Human rights and human development
1999 Globalization with a human face
1998 Consumption for human development
1997 Human development to eradicate poverty
1996 Economic growth and human development
1995 Gender and human development
1994 New dimensions of human security
1993 People’s participation
1992 Global dimensions of human development
1991 Financing human development
1990 Concept and measurement of human development
Why the Capability Approach?
approach be brought into dialogue with the theoretical work on modeling
multi-dimensionality? How can a national government facilitate ‘public
debates’ about the value judgments that are required so their poverty
reduction strategy addresses key capability poverties of women? What
empirical regularities are there worldwide in terms of the capabilities
that poor people understand to be central to their condition? What are
the inter-correlations between indications of capability poverty generat-
ed by subjective data from questionnaires, by participatory data from
Participatory Reflection and Action and similar exercises, and by
quantitative data on life expectancy and literacy?
A third observation is that, in the abstract, the capability approach may
seem unwieldy. But our problems are not abstract. The capability
approach has many degrees of freedom; concrete situations have far
fewer. The feasibility considerations can usually be jotted in, and the actual
scope for both analysis and action narrows considerably.
A fourth observation is that it is not entirely up to any researchers,
however august, to operationalize the capability approach — to hastily fill
in all of the boxes with information and value judgments. Sen has
provided an analytical map of important variables that can be useful to
practitioners who are deeply sensitive to the context, and that can be
adapted, shaped, and fitted to many different institutional levels, time
periods, groups, and so on. Sen’s refusal to ‘fill in all of the blanks’, his
decision to leave the prioritization of basic capabilities to others who are
engaged directly with a problem, demonstrates respect for the agency of
those who will use this approach. If researchers apply the capability
approach in a way that is consistent with its own tenets, then its
operationalization depends upon the thoughtful participation of many
users and much public debate. For that reason the capability approach is
very conducive to participatory undertakings but alas not to crystalize into
So what is the way ahead? Researchers and practitioners seem quite
determined to put the capability approach to work. The difference
between an insight (or in our case a proposition) and an operational
approach is that the latter has been driven around the entire discipline; it
has stopped off in every laboratory or field office to pay a visit, and it has
been helpful, thought-provoking, and challenging in all of the places
where it can be. In short, the insight has traced its implications all the way
through and shown where it made a difference and what that difference is.
You could think of the application fields of the capability approach, then,
as a set of boxes, each consisting of the related technical disciplinary tools,
whether of gender analysis or nutritional science or econometrics or
decision theory or policy-making.
The ‘things’ inside the boxes are relatively well worked out. We know,
for example, how to execute a large survey, how to input the data into the
best software, how to clean it, how to make the data publicly available on
the web, how to link it to past and future surveys. We know how to train
facilitators to develop rapport and trust with local communities, what to
wear and how to encourage diagrams so that participatory monitoring and
evaluation generates the best possible insights both for the community
and for the facilitator. We know how to measure child malnutrition and
stunting and literacy and assets. We are learning how to measure different
kinds of freedom and empowerment. We know how to do controlled
experiments, and analyze quasi-experiments that isolate the effects of a
particular policy or activity on a population. We know alternative
approaches to addressing incommensurability. Some know what kinds
of questions generate the best subjective answers about one’s feeling of
being empowered; others know how to use fuzzy set theory; yet others
know the dynamics of non-governmental organizations and international
donor agencies, how they resist new ideas and how to enable them to
Sen has been very acute in observing the ‘blind spots’ of traditional
approaches to utility, or revealed preference, or maximization, and so on.
At the same time he has continued to stress that these have something to
contribute. Their analysis conveys some fascinating insights that are part of
the picture even if they are not sufficient for a complete analysis. The
question we need to consider is what each set of tools or axioms can do.
One problem is that the tools inside the boxes are not connected to
the capability approach — the information flow between the emphases of
the capability approach and the various literatures does not exist; the
implications are not driven through. The other problem is that the tools
inside the boxes are not easy for those outside to use. They require skills
and techniques that take years and years to learn. Experts in their field
know them yet other researchers may not have mastered them. And —
going back to the first issue — the experts are not necessarily motivated by
the same objectives as the capability approach.
How do we address this? Part of the further research must be to
carefully study the best existing work in different areas and try to trace
through how the capability approach could be applied. This will require
rather a great deal of creativity. We began by talking of the need for the
capability approach to become user friendly so it is not left behind by
frustrated practitioners as the visionary basic human needs work was. But
we left off having explored the bewildering breadth and dexterity of the
capability approach. Operationalization may well be more art than science.
Another part of the research is either to master new techniques (the
very worst thing researchers could do is second-class technical work) or to
develop collaborative relationships with colleagues who are experts in a
particular tool or field that is of interest Take for example the work by
Diener and Biswas-Diener, psychologists who study subjective well-being.
After years of work in more affluent settings both turned their attention
to the subjective well-being of slum-dwellers, pavement dwellers and
sex-workers in Calcutta. Interestingly, they found from 83 interviews
and survey responses that these persons who are objectively very poor,
Why the Capability Approach?
subjectively were more than satisfied with their life (Biswas-Diener and
Diener, 2001). It is this kind of empirical work where we need to drill
down beyond Sen’s example about being grateful for small mercies, and
begin empirical studies on how to facilitate value judgments by poor
persons whose subjective preferences have adapted to abject surrounds.
To indicate another track, a number of people are working in the literature
of public deliberation and deliberative democracy, and need to link Sen’s
advocacy of public scrutiny and value judgments by public debate with
these explorations (Bohman, 2000).
So what will operationalizing the capability approach look like? It may
be a collaborative enterprise, with many researchers working on different
aspects at the same time. Researchers and practitioners will have to keep
communicating with one another in order to build up a consistent set of
simplifications and also create more momentum. The task is considerable.
As Sen wrote in Farina et al.’sFestschrift, ‘‘Welfare economics deals with
the basis of normative judgements, the foundations of evaluative
measurement, and the conceptual underpinnings of policy-making
in economics. It is not a modest subject’’ (Sen, 1996a, p. 50). To
operationalize an alternative approach to welfare economics — which is
what the capability approach is — is not a modest task, nor is it nearly
accomplished. But it is a task well worth continued attention.
1 Martha Nussbaum’s rich account of the capability approach merits separate treatment
that lies beyond the scope of this article.
2 For overviews of issues see, for example, Fukuda-Parr and Kumar (2003), Robeyns
(2000), Sugden (1993), Nussbaum and Sen (1993), Crocker (1991, 1992) and Alkire
3 For brief historical accounts see, for example, Gasper (1996), Streeten (1995) and
Doyal and Gough (1991).
4 See Sen (1992, p. 39; 1999a, p. 75); see also Sen (1990a). For the context of this paper I
often use examples of poverty reduction, but of course the capability approach is not
limited to these; other focal issues might be chosen with equal relevance.
5 Sen traces the roots of this approach to human flourishing to Aristotle’s writings in
both The Nicomachean Ethics and Politics (1992, p. 39; 1999a, p. 73).
6InQuality of Life Sen pointed out that he had not used the term ‘basic capabilities’ in
his Dewey Lectures, nor in Commodities and Capabilities, in order to avoid confining
the capability approach ‘‘only to the analysis of basic capabilities’’ (Sen 1993a, p. 41, n.
33). Dre`ze and Sen’s Indian Development takes up again the language of basic human
capabilities, as does Development as Freedom (Sen, 1999a, p. 20). However although
Development as Freedom does identify five instrumental freedoms, Sen does not, ever,
identify and defend a canonical ‘list’ of basic capabilities; indeed, he argues
convincingly against such a list (as presented by Sen in his Keynote Address at the
Human Capability Conference, Pavia, Italy, September 2004).
7 See the exchange between Williams and Sen (1987b, pp. 98–101, 108–109). Sen
(1993a, p. 37) writes that assessments of the standard of living focus on ‘‘those
influences on well-being that come from the nature of [the person’s] life, rather than
from ‘other-regarding’ objectives or impersonal concerns’’.
8 See Sen (1992, pp. 56–57; 1999a, p. 191; 2002, chapters 19–21) and Sen’s third Dewey
lecture (1985a, pp. 203–221).
9 A further research agenda might be developed to consider harmful or negative beings
and doings — and people’s interest in advancing them. See Frances Stewart’s keynote
address given at Pavia in September 2004, forthcoming in the Journal of Human
Development, July 2005.
10 See also Sen (1985b, 1991, 1997a, and references cited therein).
11 See Sen (1985a, lecture 3). For a summary of Sen’s writings on individual and collective
agency see Alkire (forthcoming, and references cited therein). See also Deneulin in this
issue. For each of the remaining topics I give only one citation; for more extensive
referencing, see Alkire (2002).
12 Attention to the informational requirements of moral principles was the substance of
Sen’s first Dewey lecture (1985a) and has been sustained in subsequent writings. See
also Sen (1979).
13 Sen (1997c) offers alternative ways of measuring capabilities.
14 See, for example, the bibliography online (www.hd-ca.org).
15 See, for example, Anand and Sen (1993, 1994, 1995, 1997, 2000a, 2000b) and Sen
(1997d). But also see Fukuda-Parr (2002).
Ahmad, E., Dre`ze, J., Hills, J. and Sen, A. (Eds.) (1991) Social Security in Developing
Countries, Clarendon Press, Oxford.
Alkire, S. (2002) Valuing Freedoms: Sen’s Capability Approach and Poverty Reduction,
Oxford University Press, Oxford.
Alkire, S. (2004) ‘The role of external actors’, in V. Rao and M. Walton (Eds.), Culture and
Public Action: A Cross-Disciplinary Dialogue on Development Policy, Stanford
Alkire, S. (2005) ‘Structural injustice and democratic practice: the trajectory in Sen’s
writings’, in S. Deneulin, M. Nebel and N. Sagovsky (Eds.), Capability and Justice,
Alkire, S. and Deneulin, S. (2002) ‘Individual motivation, its nature, determinants and
consequences for within group behaviour’, in J. Heyer, F. Stewart and R. Thorp (Eds.),
Group Motivation and Development: Is the Market Destroying Cooperation?, Oxford
University Press, Oxford.
Anand, S. and Sen, A. (1993) ‘Human Development Index: methodology & measurement’,
Occasional Paper 12, Human Development Report Office, UNDP, New York.
Anand, S. and Sen, A. (1994) ‘Sustainable human development: concepts and priorities’,
Occasional Paper 8, Human Development Report Office, UNDP, New York.
Anand, S. and Sen, A. (1995) ‘Gender inequality in human development: theories and
measurement’, Background paper for the 1995 Human Development Report, Human
Development Report Office, United Nations Development Programme, New York.
Anand, S. and Sen, A. (1997) ‘Concepts of human development and poverty: a
multidimensional perspective’, Background Paper for Human Development Report
1997, Human Development Report Office, United Nations Development Programme,
Anand, S. and Sen, A. (2000a) ‘Human development and economic sustainability’, World
Development, 28(12), pp. 2029–2049.
Anand, S. and Sen, A. (2000b) ‘The income component of the Human Development Index’,
Journal of Human Development, 1(1), pp. 83–106.
Biswas-Diener, R. and Diener, E. (2001) ‘Making the best of a bad situation: satisfaction in
the slums of Calcutta’, Social Indicator Research, 55, pp. 329–352.
Bohman, J. (2000) Public Deliberation: Pluralism, Complexity, and Democracy, MIT
Why the Capability Approach?
Colander, D. (2004) ‘The making of an Economist II’, Middlebury College Working Paper
Crocker, D.A. (1991) ‘Toward development ethics’, World Development, 19(:),
Crocker, D.A. (1992) ‘Functioning and capability: the foundations of Sen’s and Nussbaum’s
development ethic’, Political Theory, 20(4), pp. 584–612.
Doyal, L. and Gough, I. (1991) A Theory of Human Need, MacMillan, Basingstoke.
Dre`ze, J. and Sen, A. (1989) Hunger and Public Action, Clarendon Press, Oxford.
Dre`ze, J. and Sen, A. (1990) The Political Economy of Hunger, volume 1, Clarendon Press,
Dre`ze, J. and Sen, A. (1995) India: Economic Development and Social Opportunity,
Oxford University Press, Delhi.
Dre`ze, J. and Sen, A. (2002) India: Development and Participation, Oxford University
Fukuda-Parr, S. (2002) ‘Rescuing the human development concept from the HDI:
reflections on a new agenda’, in S. Fukuda-Parr and A.K. Shiva Kumar (Eds.), Readings
in Human Development, Oxford University Press, Delhi.
Fukuda-Parr, S. (2003) ‘The human development paradigm: operationalizing Sen’s ideas
on capabilities’, Feminist Economics, 9(2/3), pp. 301–317.
Fukuda-Parr, S. and Shiva Kumar, A.K. (Eds.) (2003) Readings in Human Development,
Oxford University Press, Delhi.
Gasper, D. (1996) ‘Needs and basic needs: a clarification of meanings, levels and different
streams of work’, Institute of Social Studies Working Paper, No. 210, Institute of Social
Studies, The Hague.
Hicks, J. (1983) ‘‘‘Revolutions’’ in economics,’, in, Classics and Moderns: Collected Essays
on Economic Theory, volume III, Basil Blackwell, Oxford, pp. 3–16, [Reprinted from S. J.
Latsis (Ed.), Method and Appraisal in Economics, Cambridge University Press,
Cambridge, pp. 207–218].
Keynes, J.N. (1891) The Scope and Method of Political Economy, Macmillan and Co,
Nussbaum, M. and Sen, A. (Eds.) (1993) The Quality of Life, Clarendon Press, Oxford.
Rawls, J. (1971) A Theory of Justice, Harvard University Press, Cambridge, Mass.
Rawls, J. (1993) Political Liberalism, Columbia University Press, New York.
Robeyns, I. (2000) ‘An unworkable idea or a promising alternative? Sen’s capability
approach re-examined’, CES discussion paper 30, Katholleke Universitat, Leuven.
Sen, A.K. (1979) ‘Informational analysis of moral principles’, in R. Harrison (Ed.), Rational
Action, Cambridge University Press, Cambridge, pp. 115–132.
Sen, A.K. (1982) ‘Liberty as control: an appraisal’, Midwest Studies in Philosophy,7,
Sen, A.K. (1984) ‘The living standard’, Oxford Economic Papers, 36, pp. 74–90.
Sen, A.K. (1985a) ‘Well-being agency and freedom: the Dewey Lectures 1984’, Journal of
Philosophy, 82(4), pp. 169–221.
Sen, A.K. (1985b) Commodities and Capabilities, Elsevier, Amsterdam.
Sen, A.K. (1987a) On Ethics and Economics, Basil Blackwell, New York.
Sen, A.K. (1987b) ‘The standard of living’, in A. Sen, Muellbauer, J. Muellbauer, R. Kanbur,
K. Hart and B. Williams (Eds.), The Standard of Living: The Tanner Lectures on Human
Values, Cambridge University Press, Cambridge.
Sen, A.K. (1988) ‘The concept of development’, in H. Chenery and T.N. Srinivasen (Eds.),
The Handbook of Development Economics, volume I, Elsevier Publishers, Amsterdam.
Sen, A.K. (1989) ‘Economic methodology: heterogeneity and relevance’, Social Research,
56(2), pp. 299–329.
Sen, A.K. (1990a) ‘Development as capability expansion’, in K. Griffin and J. Knight (Eds.),
Human Development and the International Development Strategy for the 1990s,
Sen, A.K. (1990b) ‘Justice: means versus freedoms’, Philosophy & Public Affairs, 19,
Sen, A.K. (1990c) ‘Gender and cooperative conflicts’, in I. Tinker (Ed.), Persistent
Inequalities, Oxford University Press, New York.
Sen, A.K. (1991) ‘Welfare, preference and freedom’, Journal of Econometrics, 50,
Sen, A.K. (1992) Inequality Reexamined, Harvard University Press, Cambridge, Mass.
Sen, A.K. (1993a) ‘Capability and well-being’, in A. Sen and M. Nussbaum (Eds.), The
Quality of Life, Clarendon Press, Oxford, pp. 30–53.
Sen, A.K. (1993b) ‘Markets and freedoms: achievements and limitations of the market
mechanism in promoting individual freedoms’, Oxford Economic Papers, 45,
Sen, A.K. (1995) ‘Gender inequality and theories of justice’, in M. Nussbaum and J. Glover
(Eds.), Women, Culture and Development: A Study of Human Capabilities, Clarendon
Press, Oxford, 259–273.
Sen, A.K. (1996a) ‘On the foundations of welfare economics: utility, capability and practical
reason’, in F. Farina, F. Hahn and S. Vannucci (Eds.), Ethics, Rationality and Economic
Behaviour, Clarendon Press, Oxford, 50–65.
Sen, A.K. (1996b) ‘On the status of equality’, Political Theory, 24(3), pp. 394–400.
Sen, A.K. (1997a) ‘Maximization and the act of choice’, Econometrica, 65(4), pp. 745–779.
Sen, A.K. (1997b) ‘Population policy: authoritarianism versus cooperation’, Journal of
Population Economics, 10(1), pp. 3–22, [Text of an address to the MacArthur
Foundation, New Delhi, August 1995; reprinted in Social Change, Journal of the
Council for Social Development, New Delhi, 1996].
Sen, A.K. (1997c) On Economic Inequality: with a substantial annexe ‘after a Quarter
Century’ by J Foster and A Sen, 2nd edition, Clarendon Press, Oxford.
Sen, A.K. (1997d) ‘Poverty in the human development perspective: concept and
measurement’, Chapter 1, in, Human Development Report, 1997, pp. 15–23.
Sen, A.K. (1997e) ‘Economics, business principles and moral sentiments’, Business Ethics
Quarterly, 7(3), pp. 5–15.
Sen, A.K. (1998) ‘The possibility of social choice’, American Economic Review, 89(3),
Sen, A.K. (1999a) Development As Freedom, Knopf Press, New York.
Sen, A.K. (1999b) ‘Democracy as a universal value’, Journal of Democracy, 10(3), pp. 3–17.
Sen, A.K. (2000a) ‘Consequential evaluation and practical reason’, The Journal of
Philosophy, XCVII(9), pp. 477–502.
Sen, A.K. (2000b) ‘The discipline of cost-benefit analysis’, Journal of Legal Studies, XXIX,
Sen, A.K. (2002) Rationality and Freedom, Belknap Press, Cambridge, MA.
Sen, A.K. (2004) ‘Elements of a theory of human rights’, Philosophy &Public Affairs, 32(4),
Sen, A.K. (2005) The Argumentative Indian, Allen Lane.
Stewart, F. (1985) Basic Needs in Developing Countries, Johns Hopkins University Press,
Streeten, P. (1984) ‘Basic needs: some unsettled questions’, World Development, 12(9),
Streeten, P. (1995) ‘Human development means and ends’, UNDP Roundtable on Global
Change Paper 6, United Nations Development Program, New York.
Streeten, P., et al. (1981) First Things First: Meeting Basic Human Needs in Developing
Countries, Oxford University Press for the World Bank, London.
Sugden, R. (1993) ‘Welfare, resources and capabilities: a review of Inequality Reexamined
by Amartya Sen’, Journal of Economic Literature, 31, pp. 1947–1962.
United Nations Development Program (1990–2004) The Human Development Report,
United Nations Development Program, New York.
Why the Capability Approach?