DRAFT. NOT TO BE CITED.

Development and the limits of Amartya Sen’s The Idea of Justice

Séverine Deneulin1

Abstract
The paper analyzes the contribution of Amartya Sen’s The Idea of Justice for development studies. The
influence of Sen’s book for thinking about development is likely to parallel that of John Rawls’s Theory of
Justice in political theory. This paper argues that The Idea of Justice has a limited reach in relation to addressing
concrete cases of injustice because it is built only on the dual foundations of freedom and reasoning. On the basis
of real world examples of unjust situations derived from Sen’s body of writings itself, the paper discusses the
limits of The Idea of Justice. It contends that remedying injustice requires an understanding of how justice is
structural and which recognizes that discussion of justice is inseparable from reasoning about the nature of the
good society. The paper concludes by pointing out The Idea of Justice’s ambiguous relationship with liberalism.

Keywords: Justice, freedom, reasoning, structural injustice, liberalism, Amartya Sen

Introduction
In the 1960s, a group of Latin American social scientists named the development model
adopted by Latin American countries unjust. Supplying cheap raw materials to Western
economies and buying off them finished products at inflated prices created a situation of
dependence which severely limited the scope for Latin American economies to provide
employment opportunities and reduce poverty. Justice required that Latin American
economies broke their dependence ties to Western economies. However, with the collapse of
import-substitution policies in the early 1980s after the turmoil of the oil and debt crisis, the
intellectual revolution of dependency theory within development studies was short-lived, and
‘justice’ disappeared from the development vocabulary to make room for the ‘pro-poor
growth’, ‘participation’, ‘community-driven development’, ‘empowerment’, ‘social capital’
and all the many other buzzwords that have inhabited development discourses since then.
In the 1990s, with the integration of human rights into development discourses and
practices, justice became again a major concern for development studies, but the language of
justice shifted away from the structural analysis of dependency theory to a liberal discourse of
individual rights and freedoms. Justice is no longer the product of just structural relations
between economies but the product of just outcomes between individuals. Every individual is
1
Centre for Development Studies, University of Bath, UK. E-mail: [email protected]. This article expands
the argument of ‘Michal Sandel’s Justice and Amartya Sen’s Idea of Justice: A Review Essay’ published in
Oxford Development Studies, October 2010, vol 38, issue 3, pp. 383-8. I am grateful to Augusto Zampini for
discussions on the topic, and Nick Townsend and Dana Bates for critical comments on an earlier draft.

1
 DRAFT. NOT TO BE CITED.

born equal and failing to recognize these equal rights is unjust. It is unjust for an Afghan girl
to be denied access to education while her brother is not. It is unjust that a woman in the
Congo is more than 100 times more likely to die in childbirth than a woman who lives in the
UK.2 It is unjust for a Chinese student to be in prison because s/he has opposed a government
policy. While not linked with human rights as such, the Millennium Development Goals and
their targets of achieving gender equality in education, reducing child and maternal mortality,
exemplify a partial and imperfect attempt to bring concerns for justice for individuals to the
heart of development processes.
Amartya Sen’s Idea of Justice situates itself within that liberal tradition of integrating
justice and development. At first glance, The Idea of Justice does not appear to add any new
insight to what is already in the Amartya Sen corpus. Like the central argument of
Development as Freedom, The Idea of Justice holds that the development process should be
about providing the opportunities for people to live the kind of lives they have reason to
value. It is about expanding valuable freedoms, such as freedoms to read and write, to be
healthy, to live in peaceful and secure environments, to participate in the life of the
community, to appear in public without shame, etc. At a second glance however, The Idea of
Justice goes much further than Development as Freedom. It presents the expansion of
valuable freedoms as a matter of justice. That 4,000 children die each day in the world as a
result of diarrhoea, while the means to easily prevent it through oral re-hydration therapy
exist, is unjust. That child malnutrition persists in India despite a decade of high levels of
economic growth is unjust. These situations of injustice require urgent remedial action.
In this sense, Sen’s Idea of Justice constitutes a significant intellectual revolution for
development studies that is similar to what Rawls’s Theory of Justice did for political theory.
In policy discourses dominated by a language which uses development as synonymous to
poverty reduction, The Idea of Justice advances the bold argument that development should
be synonymous to making the world less unjust, for poverty reduction and reduction of
injustices do not necessarily go together. If a country has reduced its income poverty rate
from 40% to 30%, fewer people live below the poverty line, but this poverty reduction has not
necessarily made the country more ‘just’ as a large percentage of people might still be denied
many valuable freedoms. They might still be unable to access education, to avoid easily
preventable diseases or be denied freedom of expression. The Idea of Justice might therefore

2
The most recent WHO data estimate maternal mortality rates in the Democratic Republic of Congo at 990 per
100,000 live births and 7 in the UK. Most recent UNICEF data estimate net primary enrolment rate in
Afghanistan at 74% for boys and 46 % for girls.

2
 DRAFT. NOT TO BE CITED.

change development studies drastically, taking it away from its concern for poverty reduction
towards justice. But how far does The Idea of Justice pass the test of doing what it set out to
do: to diagnose concrete cases of injustice and offer insights to make the world less unjust?
This is the question that the paper will seek to answer.
It starts by examining how The Idea of Justice links development with justice through
two core ideas: freedom and reasoning. It then tests how these two ideas can help us analyze
concrete unjust situations. By confronting the argument of The Idea of Justice with concrete
examples of unjust situations discussed in Sen’s writings itself, the paper underlines some of
the limits of a freedom and reasoning-based idea of justice. It concludes that, for Sen’s idea of
justice to be translated into remedial action, it needs to be structural and not individual, and be
based more explicitly on reasoning about the good life and the good society.

Justice: Freedom and reasoning

Sen’s Idea of Justice is mainly directed at two audiences: political philosophers and social
scientists. So far it is in the area of political theory that The Idea of Justice has generated most
reaction.3 One of the reasons for this lies in the recent history of political philosophy. Since
the publication of John Rawls’s Theory of Justice in 1971, academic discussions in political
theory conducted in Anglo-Saxon circles on the issue of justice could hardly avoid engaging
with this work.4 To date, no account of justice coming from the English-speaking world has
rivalled the primacy of Rawls’s Theory of Justice. Even Rawls’s communitarian critiques5
have not succeeded in offering an alternative theory of justice with the same influence as
Rawls had on Anglo-Saxon political philosophy. Sen’s Idea of Justice could do that as it
provides a robust alternative to Rawls’s Theory of Justice for thinking about social
arrangements and distributive matters outside the utilitarian tradition.
The thrust of the argument of The Idea of Justice, already presented in his article,
‘What do we want from a theory of justice? (Sen, 2006), is that the question ‘What is a just
society?’, is not a good starting point for thinking about justice. What is needed is a
comparative, not transcendental, approach to justice. One does not need to know what a
perfectly just society is, and what constitutes just institutional arrangements, as is the case in

3
Book reviews in the press and academic journals have mainly been written by political philosophers who
discussed The Idea of Justice in relation to theories of justice in philosophy.
4
Rare exceptions to this are given by English-speaking academics working from within the German
philosophical tradition of Hegel, Marx and critical theory.
5
See Mulhall and Swift (2002), for a summary of the ‘liberal-communitarian’ debate.

3
 DRAFT. NOT TO BE CITED.

Rawls’s Theory of Justice, in order to identify injustices and seek remedial action. Justice and
what constitutes ‘just institutions’ need not be defined in order to be able to say that a state of
affairs where fifty per cent of the population under five are malnourished is more unjust than
one where five percent are. A comparative framework, which enables people to evaluate
states of affairs and judge whether one is better or worse than another, is sufficient, according
to The Idea of Justice, to address injustice.
By making the capability space the comparative framework for justice, Sen’s idea of
justice is freedom-based.6 One state of affairs is more just if people enjoy more freedoms to
live a life they have reason to value. If in situation A, more people are free to be healthy, go to
school, express themselves and participate in running the affairs of their community, than in
situation B, where some minorities are excluded from health and educational services and
political life, situation A is more just because more people enjoy more valuable freedoms.
According to Sen’s idea of justice, comparing situation A with situation B within the
capability space is sufficient to start remedying injustice. One does not need to identify ‘just’
institutional arrangements, e.g. whether collective ownership of capital by the workers is
more just or unjust than a handful of shareholders owning a company. It suffices to compare
these institutional arrangements according to their consequences for people’s freedoms. All
we need for justice is to have a comparative framework which evaluates whether one
institutional arrangement leads to a situation where more people are living the kind of life
‘they have reason to choose and value’.
Already back in 1990, in an article entitled ‘Justice: Means vs. Freedoms’, Sen
articulated his freedom-based idea of justice. In his 1979 Tanner Lectures, and more
expansively in his 1984 Dewey Lectures, Sen (1980, 1985) made the case for ‘capabilities’ as
a more appropriate space for assessing wellbeing than the utility space, and as a more
appropriate informational basis for justice than Rawls’s primary goods. His argument was that
personal heterogeneity meant that people would require different levels of primary goods in
order to do the same things. For example, a paraplegic person requires a larger amount of
some primary goods, such as income, in order to travel about in their neighbourhood than a
person who is able to walk.
Rawls’s rejoinder to Sen’s critique was that making capabilities instead of primary
goods the informational basis of justice would lead to a comprehensive view of the good
which goes against a political conception of justice. It would require making a judgement
6
For his capability approach, see among others Sen (1980, 1985, 1992, 1993). For a general introduction to the
capability approach, see Alkire and Deneulin (2009).

4
 DRAFT. NOT TO BE CITED.

about the nature of what people do or are. The capability to move, to eat well, to participate in
the life of the community are the informational basis for justice, in Sen’s account, because
they are worthwhile as such, because they constitute what good living is about. Rawls’s
objection is that if people wish, for example, to live a reclusive life, without going out, eating
a decent diet or participating in the political affairs of their countries by voting, they should be
left free to do so. Rawls’s concern is that, on Sen’s account, a hermit who chooses to live in a
secluded location without transport and eat a diet of bread and water should be counted as
suffering from injustice.
Sen responds to Rawls’s concerns and reaffirms the deep liberal foundations of his
idea of justice. Central to (political) liberalism is the idea of freedom. People have different
understandings of what it means to live well and the government may not advance a specific
conception of the good. It has to be neutral and provide the conditions for the freedom of
every individual to live a life of his or her own choosing. Justice, in the liberal philosophical
tradition, requires foremost respect for individual freedom. A state of affairs is just if it has
enabled each individual freely to live his or her conception of the good life.
Sen reassures Rawls that the move from primary goods to capabilities remains
consistent with a liberal conception of justice: ‘A theory of justice based on fairness must be
deeply and directly concerned with the actual freedoms enjoyed by different persons –
persons with possibly divergent objectives – to live different lives that they can have reason to
value’ (Sen, 1990: 112). Because a ‘capability-based assessment of justice’ rests on ‘the
freedoms they actually enjoy to choose between different ways of living that they can have
reason to value’ (Sen, 1990: 115) and not in what they achieve, it does not assume a
comprehensive view of the good life: ‘Capability reflects a person’s freedom to choose
between alternative lives (functioning combinations), and its value need not be derived from
one particular “comprehensive doctrine” demanding one specific way of living’ (Sen, 1990:
118). 7
This freedom-based conception of justice does not only have an opportunity aspect in
the capability sense, it has also a process aspect in the agency sense, expressed through public
reasoning.8 The opportunities that people have to live the kinds of lives they have reason to
7
For a discussion of the capability approach as a liberal school of thought, see Robeyns (2009).
8
On the distinction between the opportunity and process aspect of freedom, Sen (2002a: 10) writes: ‘Freedom
can be valued for the substantive opportunity it gives to the pursuit of our objective and goals. In assessing
opportunities, attention has to be paid to the actual ability of a person to achieve those things that she has reason
to value. The focus is not directly on what the processes involved happened to be, but on what the real
opportunities of achievement are for the persons involved. The process aspect of freedom focuses on the

5
 DRAFT. NOT TO BE CITED.

value are not to be provided by a benevolent dictator who knows what is good but by the
people themselves, through processes of collective reasoning and decision-making. Public
reasoning is the exercise of democracy par excellence. Democracy is not only about free and
fair elections but also about freedom of expression, of information, freedom to protest and
dissent (Sen, 2009: 327; Sen and Drèze, 2002). If in both situations A and B, people benefit
from a good health care system and are given the opportunities to avoid easily preventable
diseases, but if in situation A, the health care system has been designed by a bureaucracy in
which people have no voice, and if in situation B, it has been designed through collective
processes where people themselves had a say, then according to Sen’s idea of justice,
situation B is more just than situation A, even if both situations are equally just in the
opportunity sense of freedom.
The Idea of Justice is replete with references to the importance of reasoning for
thinking about justice. To cite a few: ‘Reasoning is central to the understanding of justice’ (p.
xviii); ‘The role of unrestricted public reasoning is quite central to democratic politics in
general and to the pursuit of social justice in particular’ (p. 44); ‘Public reasoning is so
critically important for the practice of justice’ (p. 328); ‘Open-minded engagement in public
reasoning is quite central to the pursuit of justice’ (p. 390); ‘When we try to determine how
justice can be advanced, there is a basic need for public reasoning, involving arguments
coming from different quarters and divergent perspectives’ (p. 392).
The Idea of Justice does not sideline the prevalence of disagreement and ‘unreason’ in
public reasoning processes. When people come together to discuss matters of collective
concerns and try to reach a decision about these, it is sensible to expect diversity of opinions
and considerable opposition to one’s views. When feminist pioneer Mary Wollstonecraft
advanced her views that women were capable of reason and should enjoy the same political
and civil rights as men, she encountered a lot of opposition. Many men (and women) did not
readily accept her views. The suffragettes also had to encounter a lot of ‘unreason’ from men,
who had their own ‘reasons’ to keep women outside the economic, social and political sphere.
Nonetheless, by persistent reasoning, Sen argues, men’s ‘unreason’ was finally overcome by
reason and the case for women’s rights eventually won.
The reality of clashing reasons does not rule out the possibility of people changing
their views on the basis of accepting others’ reasons – e.g. men coming to realize that their
reasons to keep women outside economic and political influence are flawed. This can be
because the reason for holding certain views are often based on prejudices that do not

freedom involved in the process itself (e.g. whether the person was free to choose herself).’

6
 DRAFT. NOT TO BE CITED.

withstand critical scrutiny: ‘Actual disagreements that exist may be removed through
reasoning, helped by questioning established prejudices, vested interests and unexamined
preconceptions’ (Sen, 2009:396). The case of the abolition of slavery constitutes another
example of reasoning overcoming ‘unreason’ with people finding the reasons to abolish
slavery more convincing than maintaining it. Disagreement and opposition does thus not
discredit the overarching importance of reasoning for justice. Through reason, ‘good’
reasoning can overcome ‘bad’ reasoning:

The pervasiveness of unreason presents good grounds for scepticism about the practical effectiveness of
reasoned discussion of confused social subjects. […] This particular scepticism of the reach of
reasoning does not yield any ground for not using reason to the extent one can, in pursuing the idea of
justice […] Unreason is mostly not the practice of doing without reasoning altogether, but of relying on
a very primitive and very defective reasoning. There is hope in this since bad reasoning can be
confronted by better reasoning. (p. xvii-xviii)

The cases of women’s rights and abolition of slavery represent complete agreements
over a specific issue of justice, but justice does not always involve complete agreement, as
Sen insists repeatedly: ‘An engagement with contrary arguments does not, however, imply
that we must expect to be able to settle the conflicting reasons in all cases and arrive at agreed
positions on every issue’ (p. 392); ‘If the importance of public reasoning has been one of the
major concerns of this book, so has been the need to accept the plurality of reasons that may
be sensibly accommodated in an exercise of evaluation’ (p. 394); ‘Judgements about justice
have to take on board the task of accommodating different kinds of reasons and evaluative
concerns’ (p. 395). In many cases, Sen argues, it is sufficient to stop at a partial ranking
without having to look for complete agreement over all rankings: ‘Reasoned conclusions can
easily take the form of partial rankings’ (p. 396). For example, there may not be complete
agreement on the extent of the role of the state in providing welfare for vulnerable people, but
there can be a minimum agreement that a situation where vulnerable children are provided
free state-funded school meals is more just than a situation where they are unable to
concentrate because of lack of food. Justice, according to The Idea of Justice, does not require
complete agreement on what just states of affairs are, even less on what constitute just
institutions, but reasoned partial agreement that one state of affairs is more just than another.
With its rejection of a transcendental approach to justice, The Idea of Justice has been
heralded as a theory of justice ‘for an imperfect world’ (Osmani, 2010), as ‘dedicated to the
reduction of injustice on Earth rather than to the creation of ideally just castles in the air’. 9 But

9
Book commendation by the late political philosopher G.A. Cohen on Amazon’s website.

7
 DRAFT. NOT TO BE CITED.

despite its dual ambition of putting political philosophy in touch with the reality of people’s
lives blighted by injustices and of bringing concerns for justice to the heart of development
thinking, The Idea of Justice does not do much to show that it has the reach to enable
remedying injustice ‘in the real world’, beyond generic references to famines, gender injustice
or malnutrition. The next section considers specific cases of poverty and injustice and
examines how Sen’s freedom and reasoning-based idea of justice offers insights to improve
people’s lives and further justice in those situations.

The empirical reality test of The Idea of Justice

The Idea of Justice gives very few concrete examples of how its theory translates into
practice. The hypothetical situation of three children quarrelling over the use of a flute is the
closest the book comes to when discussing the details of a real life dilemma of justice – a
constructed tale reported in most book reviews. The issue is about the allocation of a flute to
one of three children who have distinctive attributes: one who plays the flute, one who made
it, and one who has no toy. How to allocate the flute justly? The Idea of Justice does not say
whom should be given the flute or what a just allocation would be. It concludes instead that
there are competing moral frameworks and that there are disagreements about what
constitutes a just distribution of resources. It is the nature of justice to engage in collective
reasoning processes and to seek partial agreements on ranking of social arrangements.
The Idea of Justice is better at addressing justice from a general perspective. As noted
earlier, the case of women’s rights is discussed as an example of how reasoning can overcome
unreason and advance justice for women. There are many references to Sen’s earlier works on
famines. Because of the possibility of collective reasoning, no famine ever occurred in a
democracy. People who suffer from a lack of entitlement to food have channels to voice their
claims and denounce the injustice done to them so that remedial action can be taken. The
cause for justice is thus closely linked to democracy.
In his works with Jean Drèze, Sen is more detailed about the political economy of how
a capability-view of justice can help reduce malnutrition and advance the cause of justice.
Drèze and Sen (2002: 336-40) describe the food policy of the Indian government of
supporting a minimum price for food producers, which has led to grain stocks being left to rot
because the government had to buy surplus food to maintain prices. Assessing the situation
from a capability perspective, they give evidence that this food policy takes place in a context
of widespread malnutrition, with a large proportion of children being born below average
weight and women suffering from anaemia. The cause for such policy lies in the

8
 DRAFT. NOT TO BE CITED.

disproportionate power of large-scale farmers over subsistence farmers and rural labourers,
the former being better organized politically. Justice requires the political empowerment of
the latter group so that they can participate in the public reasoning process and overcome the
unreason of the large-scale farmers.
To sum up how The Idea of Justice works in practice: first, justice demands an
evaluation of the state of affairs in the capability space, i.e., an assessment of the extent to
which people are able to enjoy valuable freedoms;10 second, justice demands inclusive
reasoning processes which allow all parties to be heard and to reach a collective decision
about what should be done to enable more people to enjoy more valuable freedoms. In the
above case of ‘hunger amidst plenty’, the most valuable freedom selected is the freedom to be
adequately nourished (measured by calorie intake, percentages of children being born with
low birth weight and of women suffering from anaemia). The evaluation makes the judgement
that a situation where more people are adequately nourished is more just. Justice then
demands that this unjust state of affairs is addressed through reasoning, by making a
convincing argument that it is absurd to have food stocks which are the ‘equivalent of about
one tonne of food for each household below the poverty line’ (Drèze and Sen, 2002: 336) in a
context of widespread child malnutrition. Drèze and Sen highlight especially the role of
intellectuals in speaking on behalf of the marginalized, advancing their cause through
analysis, and the role of the political empowerment of the marginalized through political
organizations so they can counteract the ‘bad’ reasoning of the most powerful who make
policy decisions that harm them.
In contrast to its political economy works with Jean Drèze, The Idea of Justice is not
much illustrative of how its argument works in practice. The major injustice in the world
today, climate change and environmental degradation receives little detailed attention. In the
three pages on the topic, Sen (2009: 248-250) presents the case for sustainable development
to be conceived in terms of freedoms and not needs. The environment is to be seen in terms of
the opportunities it offers to current and future generations to live a life they have reason to
value. Sen underlines the close connection between capability expansion and environmental
protection – greater female education leads to lower fertility rate, greater education among the
general population leads to greater environmental awareness. Consistent with itself, The Idea
of Justice emphasises the importance of reasoning about values for ensuring sustainability:

10
Sen has notoriously refrained from specifying the valuable freedoms that constitute the capability evaluation
space, and left it to reasoning processes to determine what ‘valuable’ freedoms are, see Robeyns (2003), Sen
(2004) and Alkire (2002, 2007).

9
 DRAFT. NOT TO BE CITED.

Consider another subject, which is beginning, at long last, to receive the attention it deserves, that is, the
neglect and deterioration of the natural environment. It is, as is increasingly clear, a hugely serious
problem and one that is closely linked with the negative effects of human behaviour, but the problem
does not arise from any desire of people today to hurt those yet to be born, or even to be deliberately
callous about the future generations’ interests. And yet, through lack of reasoned engagement and
action, we do still fail to take adequate care of the environment around us and the sustainability of the
requirements of good life. To prevent catastrophes caused by human negligence or callous obduracy,
we need critical scrutiny, not just goodwill towards others (Sen, 2009: 48).

Thus, according to Sen, more in-depth collective reasoning is the best route to secure
greater inter- and intra-generational justice. In that context, the Copenhagen Summit in
December 2009 exemplifies how The Idea of Justice works in practice. Governments,
international organizations, non-governmental and civil society organizations reasoned
together about how to judge different states of affairs, often using competing moral
frameworks in a way that is reminiscent of Sen’s flute example. There is the discussion on
resource allocation according to utilitarianism, best exemplified by Nick Stern’s report with
its call to reduce carbon emissions to ensure maximum utility levels in the future. There is
also discussion of resource allocation according to social contract theory where justice
requires an equal distribution of rights across generations. There are moral frameworks which
are anthropocentric (like the capability approach which judges the value of the environment
according to its impact on human wellbeing), and others which are bio-centric. Despite these
fundamental disagreements about a ‘just’ resource allocation in the context of climate change,
there is partial agreement about the ranking that a world with less carbon emissions is better
than the current one, even if there is no binding agreement on how much exactly carbon
emissions should be reduced.
Despite being consistent with Sen’s idea of justice – the Copenhagen agreement is
based on reasoning and reached a partial agreement that leads to a better situation than the
current one – the comparative judgement reached does not provide the conditions for future
generations (or indeed this generation) to live well or, to use Sen’s jargon, to live a life they
have reason to value. The next section goes on to examine some limits of The Idea of Justice
which become apparent when it is confronted with the nitty-gritty details of injustice.

The limits of The Idea of Justice

In his earlier writings, Sen emphasised the strong ethical individualism of his capability
approach. Individuals are the only units for assessing development processes – what matters is
whether they have expanded individual freedoms. Institutions play a crucial role in people’s

10
 DRAFT. NOT TO BE CITED.

wellbeing (both enabling and constraining) but they are to be ‘evaluated in virtue of the causal
importance that they have for individual well-being’ (Robeyns, 2008: 90), for it is individuals
that ‘think, choose and act’ (Sen, 2002b: 81). A focus on groups or institutions may hide
forms of oppression and inequalities within the group (Alkire 2008). Along the same line, The
Idea of Justice contends that individuals are the concerns of justice because reasoning is only
carried out by individuals and not by groups:

There is indeed no particular analytical reason why group capabilities must be excluded a priori from
the discourse on justice and injustice. The case for not going that way lies in the nature of the reasoning
that would be involved. […] Ultimately, it is individual evaluation on which we would have to draw,
while recognizing the profound interdependence of the valuation of individuals who interact with each
other. […] In valuing a person’s ability to take part in the life of society, there is an implicit valuation of
the life of the society itself, and that is an important enough aspect of the capability perspective. (Sen,
2009: 246)

Humans may be the only living beings capable of reason but is their capacity for reasoning a
property of only individuals? Is an unjust situation usually the result of bad reasoning between
individuals? Let us consider the case of two individuals, a Maasai pastoralist whose livelihood
is increasingly endangered by the reality of climate change, and a financial trader in Wall
Street. Both seek to live lives they value, the Maasai pastoralist a life of livestock grazing in
the community in which he was born, and the trader a life of high stimulation in a big city.
The injustice between these two individuals is that climate change, with its changes of rain
patterns, increasingly prevents the Maasai pastoralist from living the life he has reason to
choose and value. However, injustice in this case is not a matter of different individual
attributes – one person enjoying fewer individual freedoms than another. The trader, through
his profession, contributes to the existence of an economic system which prioritizes profits
over environmental protection and encourages reckless risk-taking and wasteful consumption.
Joined by multiple other similar actions and decisions, the behaviour of the trader creates a
structure which is self-reinforcing and which destroys the environment that supports the life
the Maasai pastoralist values.
By structure, I mean, following Ricoeur (1992), something which emerges from
interpersonal relations but which, over time, becomes irreducible to these relations and yet
remains bound up with these.11 The economic system is such structure. It emerges from inter-
personal relations – behaviours which trade goods for the sake of maximum profit – but such
economic system acquires an existence which is irreducible to these relations and which
11
For a discussion of the concept of ‘structures of living together’ in Paul Ricoeur in relation to Sen’s capability
approach, see Deneulin (2008).

11
 DRAFT. NOT TO BE CITED.

structures them – in such economic system, it becomes difficult for individual economic
behaviour not to be aimed at maximum profit.
Arendt identified three fundamental kinds of social structure (Arendt, 1958; Deneulin
et al. 2006): those belonging to the cultural sphere which enable a human being to become a
human person, with a specific language, set of behavioural norms and practices; those
belonging to the economic sphere which enable people’s needs to be met, through production,
distribution and consumption; and those belonging to the political sphere which enable
humans to act and shape their destiny through their own free action.
What The Idea of Justice fails to recognize is that injustice is more than a comparative
reasoning exercise between individuals; it is quintessentially structural at two levels. First,
structures are the very support of individual reasoning. When people reason about what they
should value or should do, they must rely on a collective framework of meanings that give
their actions and decisions significance (Deneulin 2009; Deneulin and McGregor 2010); they
must rely on structures which belong to the cultural sphere in order to make these choices.
The person who chooses to be a financial trader does so because this is what makes sense and
seems worthwhile to him given the collective framework of meanings on which he draws and
which is supported by the relationships he engages with. Similarly the Maasai pastoralist
relies on his own collective framework of meanings when he decides that a semi-nomadic life
of livestock grazing with strong community bonds is more meaningful than farming with
cattle enclosures in a way that prioritizes economic returns over community bonding. This
does not subtract from the importance of individual choices. The trader could decide to resign
and become a teacher because on account of another collective framework of meaning he
finds that activity more meaningful. The Maasai pastoralist could become an agricultural
entrepreneur, breaking the bonds with his community because he finds that activity more
meaningful than semi-nomadic communal lifestyle.
The second level in which justice is structural is that the cause of injustice lies not as
much in individual actions as in the structures in which these actions take place. Taking up
again the example of hunger amidst plenty in India, the unjust situation of high levels of
malnutrition together with surplus food supply is the result of a failure of the democratic
structure, of subsistence farmers not having equal voice to that of large-scale farmers who are
more organized to have their own interests represented when policy decisions are made, such
as deciding a minimum food price for producers.
By emphasising the importance of public reasoning for reducing injustice, The Idea of
Justice implicitly situates the subject of justice in the quality of the democratic structure.

12
 DRAFT. NOT TO BE CITED.

However, like earlier writings of Sen,12 it views democracy as instrumental to individual

wellbeing and not a good as such in itself, which can be perverted and become ‘bad’, as was
the case of the democratic political structure of Germany in the 1930s. This omission has far-
reaching consequences, for it ignores the reality that structures can be perverted in such way
that individuals who act within that unjust structure may even have a sense of acting justly, as
Hannah Arendt has powerfully documented in her account of the trial of Eichmann in
Jerusalem (Arendt, 1963). The SS soldiers were only doing their work and believed that they
acted justly.
Central to Arendt’s political thought is recognition that each of the kinds of structures
she identified, cultural, economic and political, can be corrupted. Nazism, slavery, apartheid
regime in South Africa, the Israeli occupation of Palestinian territories are all cases where the
structures of life in common have become perverted to such a high degree that the conditions
for human living for a large group of people have become severely undermined. The unjust
situation of a group being denied access to resources (e.g. untouchables in India being denied
access to a well) is a clear manifestation of structural injustice, of the perversion of the
cultural and political structures which are not oriented to providing the conditions for human
flourishing. Injustice is not about an individual having more or less of a good than another
person (whether resources, freedoms or rights), but about structures being corrupted and
deviated from the good they serve. Injustice is structural. One could even add that it goes
beyond interpersonal relations for it is the very structure which defines what these relations
are and how they are to be conducted, to the extent to which it is difficult, if not impossible,
for an individual not to engage in personal relations in other ways than those defined by the
structure – it was nearly impossible for a white person under South African apartheid to have
relations with black people in a way that was not defined by the apartheid system.
Structural injustice generates a sense of powerlessness, with a risk of alienation
(Deneulin et al., 2006). Even if people disapprove of an unjust structure, there is nothing each
individual alone can do about it. They have to submit to the logic of a structure they
disapprove of but from which they cannot escape. Structural injustice may therefore also
generate a sense of alienation. People might become blind to the injustice (e.g. it was very
difficult for people within the slavery system to see slavery as unjust). This is why structural
injustice can only be overcome through collective action. While, in the short run, there may
seem to be no other possibility than maintaining structural injustice, in the long run,
12
Democratic freedom is ‘a significant ingredient – a critically important component—of individual capabilities’
(Sen 2002b: 79).

13
 DRAFT. NOT TO BE CITED.

individual victims have the power to unite and overcome structural injustice. This collective
action may be initiated within the structure (e.g. black people under apartheid in South Africa)
or/and from outside (e.g. Westerners and white south Africans denouncing the injustice of the
apartheid system).
It is therefore not a coincidence that in his writings, Sen emphasizes the importance of
collective action to overcome injustice (Alkire, 2006) but it is odd that, at the anthropological
level, The Idea of Justice continues to treat structures insofar as they promote justice for
individuals, i.e. the expansion of individual freedoms, and not as themselves manifestation of
injustice.13 Limiting the idea of justice to comparative judgements about individual lives does
not do justice to the structural nature of human life. To be effective, the idea of justice must
include a judgement of the nature of structures, whether they are ‘just’ or ‘good’, whether
they provide the conditions for people to live flourishing human lives. The question of ‘just
institutions’ so central to Rawls’s Theory of Justice is thus not redundant, as Sen contends.
The difference with Rawls in this case is that the justice of these institutions does not lie in
their respect of principles but in their orientation to the human good.
As noted earlier, Sen is reluctant to give substance to the ‘freedoms that people have
reason to choose and value’ for this would commit him to what is perceived as a non-liberal
position which does not leave scope for individuals to choose the conception of the good they
wish to pursue. But a reasoning approach to justice cannot do away with addressing explicitly
questions of the good life and the extent to which structures are consistent with the aim of the
good life. The ability of each person to live a life s/he has reason to value is constituted by
structures which may, or may not, be conducive to the good. The Israeli settler in Hebron has
good reasons to value living there. Zionism provides him the structure through which he
values being able to live on the land of his ancestors. But most voices in the international
community hold that, in this case, Zionism is a perverted cultural structure which affirms the
superiority of one race over another. 14 This is why an idea of justice for the ‘real world’ has to
incorporate an analysis of the just or unjust nature of economic, cultural and political
13
‘We have to seek institutions that promote justice, rather than treating the institutions as themselves
manifestations of injustice’ (Sen, 2009: 82).
14
The initial report of the 2000 UN conference on ‘Racism, Racial Discrimination, Xenophobia and Related
Intolerance’ stated that it recognized ‘with deep concern the increase of racist practices of Zionism and anti-
Semitism in various parts of the world as well as the emergence of racial and violent movements based on
racism and discriminatory ideas, in particular the Zionist movement, which is based on racial superiority’. The
American and Israeli delegations walked out of conference and these references had to be removed from the
final draft (Gelvin, 2007).

14
 DRAFT. NOT TO BE CITED.

structures, whether they constitute the structuring conditions of a good life in common or
whether they are perverted from that aim.
The Idea of Justice emphasizes the importance of ‘reason overcoming unreason’ but it
falls short of acknowledging that the activity of reasoning is done for the sake of an end,
namely the good that we seek to pursue. Public reasoning is about reflecting on the nature of
the good life and the kind of society one wants to create and live in. When the suffragettes
confronted the ‘bad’ reasoning of men and argued that women had the same rights as men,
their reasoning was about the nature of the ‘good’ society, whether the cultural, economic and
political structures of the time were consistent with the good life in common for all, including
women, or whether they were perverted from that aim. Their argument was that a society
which excluded women from economic and political life was ‘bad’. In contrast, men argued
that a society where women are kept at home was a ‘good’ society. The feminist struggles
were not a case of ‘reason overcoming unreason’ but a case of one conception of the good
society overcoming another.
The Idea of Justice requires explicit acknowledgment of the nature of the aim that
reasoning pursues. Justice does not consist of freedom and reasoning alone, but of ‘reasoning
together about the good life’ (Sandel, 2009: 261). Questions about justice cannot be separated
from questions about the good society and the nature of the kind of lives that people live.
Confronting The Idea of Justice with concrete situations of lack of freedoms leads to the
conclusion, following Sandel, that ‘it may not be possible to say what’s just without arguing
about the nature of the good life’ (Sandel, 2009: 207), for ‘Thinking about justice seems
inescapably to engage us in thinking about the best way to live’ (Sandel, 2009: 10). Even if
unanimous agreement cannot be reached, as Sen rightly pointed out, judgments about the
good cannot be escaped, for they are central to justice. The idea of justice is about the idea of
the good society. But this good society requires ‘good’ people who act in such a way as to
create a good society, and it requires ‘good’ structures for people to be ‘good’. If people live
in a context of structural racism, their actions are not likely to be inclusive of other races. If
people live in an environmentally destructive economic system, their actions are not likely to
be environmentally friendly.

Conclusion
The Idea of Justice is a skilful mastery of embracing different, often opposing, positions into a
seemingly consistent body of thought that pulls together major alternative ethical approaches.
It embraces the liberal political thought tradition with its focus on individual freedom and its

15
 DRAFT. NOT TO BE CITED.

non-commitment to a conception of the good. It embraces consequentalism (a key feature of

utilitarianism), with its comparative approach to justice and its assessment of states of affairs
in terms of their consequences for people’s wellbeing. It also embraces implicitly virtue ethics
as the application of The Idea of Justice unavoidably leads to questions about the good life
and the good society.
This generous philosophical embrace is one of The Idea of Justice’s greatest
strengths.15 It can criticize the Rawlsian position while remaining firmly rooted in liberalism.
It can criticize utilitarianism while remaining allied to its consequentialism. It can criticize
neo-classical economics while appealing to its claimed founder Adam Smith (especially in
relation to Smith’s impartial spectator which it proposes as an alternative to Rawls’s original
position). Responding to socialist critics, it can appeal to its Marxist roots and Marx’s idea of
human flourishing, while being careful not to engage in a critique of a capitalist mode of
production. To the Aristotelian virtue ethicists, Sen affirms he is greatly indebted to Aristotle
in his central concept of ‘capability’ while remaining careful not to advocate the ‘good
society’ as the aim of public reasoning.
I have tried to show in this paper that such generous embrace becomes problematic
when The Idea of Justice is to give insights for development studies in its task of seeking to
remedy situations where people are denied the basic conditions to live well. Assessing states
of affairs in terms of individual freedoms and expecting that reasoning will lead to a better
state is not enough. What is required is an analysis of the justice of the economic, social and
political structures constitutive of a human life lived in common with fellow human beings
and the environment. These structures are to be ‘good’, enabling people to live ‘good’ lives.
Freedom and reasoning are certainly excellent starting points for thinking about justice but the
journey needs to continue. The reality of environmental degradation and human suffering
calls The Idea of Justice towards a more structural and comprehensive destination, which
indeed would sit at odds with the liberalism of (Anglo-Saxon) Western political thought that
currently dominates development thinking. An idea of justice for the imperfect real world, as
Sen’s The Idea of Justice claims to be, has to be based on an anthropology of the human being
that inhabits that world. This entails a recognition that structures are intrinsic to human
wellbeing, for there can be no good human life without a good human life in common.16

15
See Sánchez (2009) for a discussion of the Aristotelian, Kantian, Marxist and Smithian roots of Sen’s thought.
16
For the idea of the common good and its relation to the capability approach, see Deneulin and Townsend
(2007).

16
 DRAFT. NOT TO BE CITED.

Liberalism and the cause of reducing injustice in the real world might not be as reconcilable
as The Idea of Justice would hope they would. Something will have to go.

References
Alkire, Sabina (2002), Valuing Freedoms, Oxford University Press
_______ (2006), ‘Structural Injustice and Democratic Practice: The Trajectory in Sen’s Writings’, in Deneulin et
al. (eds.), Transforming Unjust Structures, Dordrecht: Springer
______ (2007), ‘Choosing Dimensions: the Capability Approach and Multidimensional Poverty’, in N. Kakwani
and J. Silber (eds), The Many Dimensions of Poverty, New York: Palgrave
_____ (2008) ‘Using the Capability Approach: Prospective and Evaluative Analyses’, in S. Alkire, M. Qizilbash
and F. Comim (eds) The Capability Approach: Concepts, Measures and Applications. Cambridge:
Cambridge University Press.
Alkire, S. and S. Deneulin (2009), ‘The Human Development and Capability Approach’, in S. Deneulin (ed.),
An Introduction to the Human Development and Capability Approach, London: Earthscan/IDRC
Arendt, Hannah (1958), The Human Condition, Chicago: University of Chicago Press
_____ (1963), Eichmann in Jerusalem: A Report on the Banality of Evil, London: Penguin
Deneulin, Séverine (2008), ‘Beyond Individual Freedom and Agency: Structures of Living Together in Sen’s
Capability Approach to Development’, in S. Alkire, F. Comim and M. Qizilbash (eds.), The Capability
Approach: Concepts, Measures and Application, Cambridge University Press
______ (2009), ‘Advancing Human Development: Values, Groups, Power and Conflict’, WeD Working Paper,
forthcoming in J. Heyer, R. Thorp and V. Fitzgerald (eds), Overcoming Persistent Inequality and Poverty,
Basingstoke: Palgrave, forthcoming.
______ Nebel, M. and N. Sagovsky (eds) (2006), Transforming Unjust Structures, Dordrecht: Springer
______ and J. A. McGregor (2010), ‘The Capability Approach and the Politics of a Social Conception of
Wellbeing’, European Journal of Social Theory, 2010, 13(4): 501-519.
______ and N. Townsend (2007), ‘Public Goods, Global Public Goods and the Common Good’, International
Journal of Social Economics, 34(1/2): 19-36.
Drèze, Jean and Amartya Sen (2002), India: Development and Participation, Delhi: Oxford University Press
Gelvin, James (2007), The Israel-Palestine Conflict: One hundred years of war, Cambridge University Press
Mulhall, S. and J. Swift (1996), Liberals and Communitarians, 2nd ed., Oxford: Blackwell
Osmani, Siddiq (2010), ‘Theory of Justice for an Imperfect World: Exploring Amartya Sen’s Idea of Justice’,
Journal of Human Development and Capabilities, 11(4): 629-40.
Rawls, John (1971), A Theory of Justice, Cambridge, Mass.: Harvard University Press
_________ (1993), Political Liberalism, New York: Columbia University Press
Ricoeur, Paul (1992), One Self As Another, Chicago University Press
Robeyns, Ingrid (2003) ‘Sen’s Capability Approach and Gender Inequality: Selecting Relevant Capabilities’,
Feminist Economics 9(2/3): 61–92.
_______ (2008) ‘Sen’s Capability Approach and Feminist Concerns’, in S. Alkire, M. Qizilbash and F. Comim
(eds) The Capability Approach: Concepts, Measures and Applications. Cambridge: Cambridge University
Press.

17
 DRAFT. NOT TO BE CITED.

______ (2009), ‘Equality and Justice’, in S. Deneulin (ed), An Introduction to the Human Development and
Capability Approach: Freedom and Agency, London: Earthscan/IDRC
Sandel, Michael (2009), Justice: What’s the Right Thing to Do?, London: Allen Lane
Sen, Amartya (1980), ‘Equality of What?’, in S. McMurrin (ed.) Tanner Lectures on Human Values, Cambridge:
Cambridge University Press
________ (1985), ‘Well-Being Agency and Freedom: The Dewey Lectures 1984’, Journal of Philosophy 82(4):
169–221.
________ (1990), ‘Justice: Means versus Freedoms’, Philosophy and Public Affairs, 19(2); 111-121.
________ (1992), Inequality Re-examined. Oxford: Clarendon Press
________ (1993), ‘Capability and Well-Being’, in M. Nussbaum and A. Sen (eds), The Quality of Life, Oxford:
Clarendon Press
________ (1999), Development as Freedom, Oxford University Press
________ (2002a), Rationality and Freedom, Harvard: Harvard University Press
——— (2002b) ‘Symposium on Development as Freedom: Response to Commentaries’, Studies in Comparative
International Development, 37(2), pp. 78-86.
_________ (2004), ‘Capabilities, Lists and Public Reason: Continuing the Conversation’, Feminist Economics
10(3): 77–80.
_______ (2006), ‘What Do We Want from a Theory of Justice’, Journal of Philosophy, 103 (5): 215-238.
_______ (2009), The Idea of Justice, London: Allen Lane
Sánchez Garrido, Pablo (2008), Raíces intelectuales de Amartya Sen: Aristóteles, Adam Smith y Karl Marx,
Madrid: Centro de Estudios Políticos y Constitucionales. Book review in English in Journal of Human
Development and Capabilities, 2009, vol. 10 (2): 305-6.

18