themes · focus   
themes literature agenda archive anthology calendar links profile

Amartya Sen

Global Justice

Beyond International Equity


Creating a new perspective for justice, this article investigates global dimensions seriously, in regard to the formation of international solidarity and the constructions of identity patterns that go beyond national borders. It seeks a way between the established approaches of "grand universalism" and "national particularism", while avoiding over-arching generalizations on the one hand and simplistic assumptions about the subjection of individuals to a national framework on the other. Taking off from a critical discussion of Rawls' idea of justice as fairness (where a hypothetical "original position" of human equity is the starting point), the own conception of "plural affiliation" becomes central for the formulation of a third alternative. This idea expresses the range of multiple identities accessible to individuals and makes "justice" applicable to a corresponding diversity of socio-political realities, independent of the idea of national frameworks.



Rawlsian Justice as Fairness
Three Concepts of Global Justice
Institutions and Multiplicity of Agencies


»Global equity is sometimes identified with international equity. The two, however, are very different notion.«


  Global equity is sometimes identified with international equity. The two, however, are very different notion – both in terms of their constitutive contents and with respect to their policy implications. In this article I examine the nature of the distinction, which I believe is quite central to political philosophy as well as policy scrutiny. Its implications for the understanding of global public goods are also quite extensive. The contrast between global and international equity relates to quite deep differences in:


  The domain of social justice: whether relations of justice apply primarily within nations, with relations across borders being seen as relations between nations.


  The concept of a person: whether our identities and responsibilities are parasitic on nationality and citizenship, which must lexicographically dominate over solidarity based on other classifications such as group identities and viewpoints of class (including relations between workers or between businesspeople with particular ethics), gender (including feminist concerns beyond local borders), professional obligations (including the commitments of doctors, educators and social workers without frontiers) and political and social beliefs (with loyalties that compete with other identities).


  Something quite important is involved in these distinctions, which have far-reaching implications on the nature of practical reason at the global level and the choice of actions of potential agents. Ideas of justice – and corresponding actions – that cut across borders must not be confused with international relations in general, or with demands of international equity in particular.

 Rawlsian Justice as Fairness

Ben Rogers:
"Portrait: John Rawls".
In: Prospect, June 1999.
external linkArticle

Thomas Bridges:
"Post-Metaphysical Liberalism: The Case of Rawls".
From: Philosophy and Civil Society Site, 1997.
external linkArticle

Terry Hoy:
Rawls' Concept Of Justice As Political: A Defense Against Critics.
Paideia World Philosophy Conference Paper.
external linkArticle

Michael Anderheiden:
Justification by Reflective Equilibrium in Rawls's More Recent Work.
Paideia World Philosophy Conference Paper.
external linkArticle


  As in many other discussions of social justice, it is quite helpful to begin with the Rawlsian notion of "justice as fairness" (Rawls 1971, 1993). The framework of political and social analysis initiated by John Rawls's classic contributions has had a profound impact on contemporary understanding of the nature of justice. Even though, as I argue below, a serious departure from the ramifications of Rawlsian analysis will ultimately be needed, the basic idea of "justice as fairness" is an appropriate starting point.


  In the Rawlsian framework, fairness for a group of people involves arriving at rules and guiding principles of social organization that pay similar attention to everyone's interests, concerns and liberties. In working out how this may be understood, the Rawlsian device of the "original position" has proved useful. In the hypothetical original position, which is an imagined state of primordial equality, individuals are seen as arriving at rules and guiding principles through a cooperative exercise in which they do not yet know exactly who they are going to be (so that they are not influenced, in selecting social rules, by their own vested interests related to their actual situations, such as their respective incomes and wealth).


  Rawlsian analysis proceeds from the original position to the identification of particular principles of justice. These principles include the priority of liberty (the "first principle") giving precedence to maximal liberty for each person subject to similar liberty for all. The "second principle" deals with other matters, including equity and efficiency in the distribution of opportunities, and includes the Difference Principle, which involves the allocational criterion of "lexicographic maximin" in the "space" of holdings of primary goods (or general-purpose resources) of the different individuals, giving priority to the worst-off people in each conglomeration.


  Questions can be raised about the plausibility of the specific principles of justice that Rawls derives from his general principles of fairness, and it can, in particular, be asked whether the device of the original position must point inescapably to these principles of justice (my own scepticism on this point is presented in Sen 1970 and 1990). The adequacy of Rawls's focus on primary goods, which makes his Difference Principle resource-oriented rather than freedom-oriented, can be particularly questioned.  1  With those specific debates I am not primarily concerned in this article (though, when the more basic groundwork regarding the idea of the original position is completed, the bearing of these differences on the application of the groundwork will have to be taken up in the analysis to follow).

 Three Concepts of Global Justice

»Who are the individuals who are seen, hypothetically, as having gathered together in the original position to hammer out deals on rules and guiding principles?«


  My concentration in this article is on the more elementary issue of the composition of the "original position" and its implications for the understanding of fairness as well as its manifest practical consequences. In particular, who are the individuals who are seen, hypothetically, as having gathered together in the original position to hammer out deals on rules and guiding principles? Are they all the people in the world – irrespective of their nationality and citizenship – who are seen as arriving at rules that are going to govern the affairs of the whole world? Or are they instead the citizens of each nation, each country separately, gathered together in their own original positions?


  These two different conceptions can be identified, respectively, as "universalist" in a most comprehensive sense and "particularist" in its nation-based orientation.


  Grand universalism. The domain of the exercise of fairness is all people everywhere taken together, and the device of the original position is applied to a hypothetical exercise in the selection of rules and principles of justice for all, seen without distinction of nationality and other classifications.


  National particularism. The domain of the exercise of fairness involves each nation taken separately, to which the device of the original position is correspondingly applied, and the relations between nations are governed by a supplementary exercise involving international equity.

Rodney G. Peffer:
What is to be Distributed?
Paideia World Philosophy Conference Paper.
external linkArticle

R.A. Hill:
Government, Justice, and Human Rights.
Paideia World Philosophy Conference Paper.
external linkArticle

Antonio Perez-Estevez:
Intercultural Dialoque and Human Rights: A Latinamerican reading of Rawls The Law of Peoples.
Paideia World Philosophy Conference Paper.
external linkArticle


  Even though the original position is no more than a figment of our constructive imagination, the contrast between these rival conceptions can have very far-reaching implications on the way we see global justice. The formulation of the demands of global justice as well as the identification of the agencies charged with meeting those demands are both influenced by the choice of the appropriate conception and the characterization of the domain of fairness. Even the understanding of the nature of the twin concepts of "global public goods" and "global housekeeping" cannot but be influenced in the choice of domain and the concept of justice. Questions such as whose house is to be kept in shape and which joint and indivisible results are to be seen as the relevant public goods invoke the underlying issues regarding the domain of reciprocal concern and the identification of appropriate agencies.


  I shall presently argue that neither of these two conceptions – grand universalism and national particularism – can give us an adequate understanding of the demands of global justice, and that there is a need for a third conception with an adequate recognition of the plurality of relations involved across the globe. But let me first elaborate a little on the claims of each of these two classic conceptions.


  Grand universalism has an ethical stature that is hard to match in terms of comprehensive coverage and non-sectarian openness. It rivals the universalism of classical utilitarianism and that of a generalized interpretation of the Kantian conception of reasoned ethics (see Kant 1785; Bentham 1789; Mill 1861; Sidgwick 1874; Edgeworth 1881; and Pigou 1920). It can speak in the name of the whole of humanity in a way that the separatism of national particularist conceptions would not easily allow.


  And yet grand universalism is hard to adopt in working out the institutional implications of Rawlsian justice as fairness. The exercise of fairness through a device like the original position is used, in Rawlsian analysis, to yield the choice of the basic political and social structure for each society, which operates as a political unit and in which the principles of justice find their application. There are great difficulties in trying to apply this mode of reasoning to the whole of humanity without an adequately comprehensive institutional base that can implement the rules hypothetically arrived at in the original position for the entire world. It would not be, I hope, taken to be disrespectful of the United Nations to suggest that it is in no way able to play this role. Indeed, even the very conception of the United Nations – as its name indicates – is thoroughly dependent on drawing on the basic political and social organizations prevalent in the respective national states.


  All this may forcefully suggest that we seek the tractability and coherence of the nationalist particularist conception of Rawlsian justice. That is, in fact, the direction in which Rawls himself has proceeded, considering separately the application of justice as fairness in each political society, but then supplementing this exercise through linkages between societies and nations through the use of intersocietal norms. These interactions take the form of what Rawls calls "the law of peoples" (see Rawls 1996). The "peoples" – as collectivities – in distinct political formations consider their concern for each other (and the imperatives that follow from such linkages). The principle of justice as fairness can be used to illuminate the relation between these political communities (and not just between individuals, as in the original Rawlsian conception).



  It must be noted, however, that in this particularist conception the global demands of justice primarily operate through intersocietal relations rather than through person-to-person relations, which some may see as central to an adequate understanding of the demands of global justice. The nation-based characterization identifies, in fact, the domain of international justice, broadly defined. The imperatives that follow, despite the limits of the formulation, have far-reaching moral content, which has been analyzed with characteristic lucidity by Rawls. However, the restrictions (identified in the introduction of this paper) of an "international" – as opposed to a more directly "global" – approach apply forcefully to this approach, which limits the reach of the Rawlsian "law of peoples".


  How should we take note of the role of direct relations across borders between different people whose identities include, inter alia, solidarities based on classifications other than partitioning according to nations and political units such as class, gender or political and social beliefs? How do we account for professional identities (such as being a doctor or an educator) and the imperatives they generate, without frontiers? These concerns, responsibilities and obligations may not only not be parasitic on national identities and international relations, they may also occasionally run in contrary directions to international relations. Even the identity of being a "human being" – perhaps our most basic identity – may have the effect, when fully seized, of broadening our viewpoint, and the imperatives that we may associate with our shared humanity may not be mediated by our membership in collectivities such as "nations" or "peoples". As I write this article sitting in Calcutta, with the Indian subcontinent still shaking with the aftershocks of nuclear explosions, the perspective of direct interpersonal sympathies and solidarities across borders has a cogency that can substantially transcend the national particularism of the estranged polities.

»The starting point of this approach – I shall call it 'plural affiliation' – can be the recognition of the fact that we all have multiple identities, and that each of these identities can yield concerns and demands that can significantly supplement, or seriously compete with, other concerns and demands arising from other identities.«


  We do need, I believe, a different conception of global justice – one that is neither as unreal as the grand universalism of one comprehensive "original position" across the world, nor as separatist and unifocal as national particularism (supplemented by international relations). The starting point of this approach – I shall call it "plural affiliation" – can be the recognition of the fact that we all have multiple identities, and that each of these identities can yield concerns and demands that can significantly supplement, or seriously compete with, other concerns and demands arising from other identities.


  With plural affiliation the exercise of fairness can be applied to different groups (including – but not uniquely – nations), and the respective demands related to our multiple identities can all be taken seriously (irrespective of the way any conflicting claims are ultimately resolved). The exercise of "fairness", which can be illustrated with the device of the original position, need not look for a unique application. The original position is a rich way of characterizing the discipline of reciprocity and within-group universalization, and it can be used to provide insights and inspirations for different group identities and affiliations. Nor is it entirely necessary, in order to benefit from Rawls's foundational characterization of fairness, to work out an elaborate system – as in Rawls's own theory – of detailed specification of a stage-by-stage emergence of basic structures, legislation and administration. The device of the original position can be employed in less grand, less unique and less fully structured forms without giving complete priority to one canonical formulation involving national particularism.


  For example, a doctor could well ask what kind of commitments she may have in a community of doctors and patients, but the parties involved need not necessarily belong to the same nation. (It is well to remember that the Hippocratic oath was not mediated – explicitly or by implication – by any national contract.) Similarly, a feminist activist could well consider what her commitments should be to address the special deprivation of women in general – not necessarily only in her own country. The obligations that are recognized cannot, of course, each be dominant over all competing concerns, because there may well be conflicting demands arising from different identities and affiliations. The exercise of assessing the relative strength of divergent demands arising from competing affiliations is not trivial, but to deny our multiple identities and affiliations just to avoid having to face this problem is neither intellectually satisfactory nor adequate for practical policy. The alternative of subjugating all affiliations to one overarching identity – that of membership in a national polity – misses the force and far-reaching relevance of the diverse relations that operate between persons. The political conception of a person as a citizen of a nation – important as it is – cannot override all other conceptions and the behavioural consequences of other forms of group association.

 Institutions and Multiplicity of Agencies

Amartya Sen:
"Humane Development. An interview by Akash Kapur".
In: The Atlantic Unbound, 15 December 1999.
external linkInterview

Amartya Sen:
"Dix vérités sur la mondialisation".
In: Le Monde, 19.7.2001.
external linkArticle

Amartya Sen:
"Sobre Conceptos y medidas de Pobreza".
In: Comercio Exterior 42.4 (1992) (México).
external linkArticle


  There are a great many agencies that can influence global arrangements and consequences. Some of them are clearly "national" in form. These include domestic policies of particular states as well as international relations (contracts, agreements, exchanges) between states, operating through national governments. However, other cross-border relations and actions often involve units of economic operation quite different from national states – such as firms and businesses, social groups and political organizations, non-governmental organizations and so on – that may operate locally as well as beyond the frontiers. Transnational firms constitute a special case of this. There are also international organizations, which may have been set up directly by individual states acting together (such as the League of Nations or United Nations) or indirectly by an already constituted international organization (such as the International Labour Organization, United Nations Children's Fund, United Nations University, or World Institute for Development Economics Research). Once formed, these institutions acquire a certain measure of independence from the day-to-day control of individual national governments.


  Still other institutions involve non-governmental, non-profit entities that operate across borders, organizing relief, providing immunization, arranging education and training, supporting local associations, fostering public discussion and engaging in a host of other activities. Actions can also come from individuals in direct relation to each other in the form of communication, argumentation and advocacy that can influence local social, political and economic actions (even when the contacts are not as high profile as, say, Bertrand Russell's writing to Nikita Kruschev on the nuclear confrontations of the Cold War). For an adequate understanding of global justice (and a fortiori for seeing the role of "global public goods" not to mention "global housekeeping"), it is extremely important to take adequate note of the multiplicity of agencies and of the rationale of their respective operations.

»Sin embargo, quizá por conocer el medio intelectual en que se mueve, Sen se cuida mucho de indicar claramente las 'causas' de la pobreza en el mundo. Siempre analiza parcialmente el problema, porque va al 'hecho' y estudia los criterios de su medida, pero jamás habla de que la pobreza (tanto absoluta como relativa) pueda presuponer una relación de dominación.«

Enrique Dussel
(In this issue, 34)


  In operating across boundaries, cross-national institutions (and more generally, cross-national contacts) inevitably have to face issues of purpose, relevance and propriety, and these issues cannot really be dissociated from concerns of justice. In dealing with this requirement, one approach would be to repudiate the direct linkages across borders and to embed every cross-boundary relation within the limited structure of "international relations", including the "law of peoples". This can be achieved, but only at the cost, I would argue, of great impoverishment of content and reach, and certainly of massive circumlocution.


  A more appropriate alternative is to pose the issue of justice – and that of fairness – in several distinct though interrelated domains involving various groups that cut across national boundaries. These groups need not be as universally grand as the collectivity of "all" the people in the world, nor as specific and constrained as national states. There are many policy issues that cannot be reasonably addressed in either of these two extremist formats.


  How should a transnational conglomerate treat the local labour force, other businesses, regional customers or – for that matter – national governments or local administration? If there are issues of fairness involved, how should these issues be formulated – over what domain? If the spread of business ethics (generating rules of conduct, fostering mutual trust or keeping corruption in check) is a "global public good", then we have to ask how the cogency and merits of particular business ethics are to be evaluated. Similarly, if the solidarity of feminist groups helps to generate social change across borders (perhaps by providing support for local groups, by generating critiques of policies of governments or businesses or simply by helping to place the addressing of neglected inequalities on the agenda for public discussion), then the claim of such organizations – and indeed of such modes of thinking – may well be integrated with the class of global public goods. But we need to address the question as to how the affiliations and interactions, and their consequences, are to be normatively assessed, invoking such ideas as justice and fairness. All this calls for extensive use of the perspectives of plural affiliations and the application of the discipline of justice and fairness within these respective groups.


  In this article I have argued for the need to distinguish between global and international equity. The distinction has, I believe, far-reaching implications for public policy as well as for conceptual clarity. I have tried to examine some of these implications.


  Individuals live and operate in a world of institutions, many of which operate across borders. Our opportunities and prospects depend crucially on what institutions exist and how they function.


  Not only do institutions contribute to our freedoms, their roles can be sensibly evaluated in the light of their contributions to our freedoms. To see development as freedom provides a perspective in which institutional assessment can systematically occur (see Sen 1999).

Amartya Sen
is Master of Trinity College, University of Cambridge, and Professor of Economics and Philosophy at Harvard University.


  Even though different commentators have chosen to focus on particular institutions (such as the market, the democratic system, the media or the public distribution system), we have to consider them all to be able to see what they can do, individually or jointly. Many of these institutions – not just the market mechanism – cut vigorously across national boundaries and do not operate through national polities. They make contributions that have strong elements of indivisibility and non-exclusiveness that are characteristic of public goods, and their claim to be seen as "global public goods" is quite strong. The literature has to take note of this important issue.


Sudhir Anand / Martin Ravallion (1993): "Human Development in Poor Countries: On the Role of Private Incomes and Public Services". In: Journal of Economic Perspectives 7.1, 133-50.

Jeremy Bentham (1789): An Introduction to the Principles of Morals and Legislation. London: Payne. Republished Oxford: Clarendon Press, 1907.

Meghnad Desai (1995): Poverty, Famine and Economic Development. Aldershot: Elgar.

Jean Drèze / Amartya K. Sen (1989): Hunger and Public Action. Oxford: Clarendon Press.

Francis Y. Edgeworth (1881): Mathematical Psychics. London: Kegan Paul.

Keith Griffin / John Knight (eds.) (1989): "Human Development in the 1980s and Beyond". In: Journal of Development Planning 19 (special issue).

Geoffrey Hawthorn (ed.) (1987): The Standard of Living. Cambridge: Cambridge University Press.

Immanuel Kant (1785): Fundamental Principles of Metaphysics of Ethics. English translation by T.K. Abbott. Republished London: Longman, 1907.

John Stuart Mill (1861): Utilitarianism. London: Longman.

A.C. Pigou (1920): The Economics of Welfare. London: Macmillan.

John Rawls (1971): A Theory of Justice. Cambridge/Mass.: Harvard University Press.

John Rawls (1993): Political Liberalism. New York: Columbia University Press.

John Rawls (1996): A Theory of Justice. Oxford: Oxford University Press.

Amartya K. Sen (1970): Collective Choice and Social Welfare. San Francisco: Holden-Day. Republished Amsterdam: North Holland 1979.

Amartya K. Sen (1984): Resources, Values and Development. Oxford: Blackwell – Cambridge/Mass.: Harvard University Press.

Amartya K. Sen (1985a): Commodities and Capabilities. Amsterdam: North-Holland.

Amartya K. Sen (1985b): "Well-being, Agency and Freedom: The Dewey Lectures 1984". In: Journal of Philosophy 82.

Amartya K. Sen (1990): "Justice: Means versus Freedoms". In: Philosophy and Public Affairs 19.

Amartya K. Sen (1999): Development As Freedom. New York: Alfred A. Knopf.

Henry Sidgwick (1874): The Method of Ethics. London: Macmillan.

United Nations Development Programme – UNDP (1990): Human Development Report 1990. New York: Oxford University Press.



The contrast in the informational perspective in the conceptual framework can have many practical implications, which I discuss in Sen 1985b. On their relevance for economic policy and related issues, see also Sen 1984, 1985a, Hawthorn 1987, Drèze / Sen 1989, Griffin / Knight 1989, UNDP 1990, Anand / Ravallion 1993 and Desai 1995. 

themes literature agenda archive anthology calendar links profile

home  |  search  |  sitemap  |  newsletter  |  interphil  |  imprint  |  donations