Amartya Sen, The Idea of Justice, Belknap Press, 2009, pp. 496.
It is a mark of the greatness of John Rawls’ A Theory of Justice that, after nearly 40 years, it still retains its position as the one work of post-war political theory which must be contended with. One might say of Amartya Sen’s new book what Jürgen Habermas, in his own debate with Rawls, once said: namely, that any disagreement between the two is basically a family feud. The dispute, to be sure, is a significant one. It concerns the viability of an idealized theory of society such as Rawls has presented it in A Theory of Justice. I say viability since it is not really Sen’s concern to argue about the philosophical foundations of Rawls’ approach. Rather, Sen wishes to substitute a more practically oriented theory of justice, the capabilities approach, for Rawls’ foundationalism about institutional structures and their relation to justice. Sen’s main criticism is thus that Rawls’ idealized theory excludes too much of what we should care about: the person’s
real position in the world, as well as parts of the world which are not included in Rawls’ closed system of the state. In short, Sen is driven by the practical need to provide a theory which is truly universal in scope, but which is also able to deal with vexing questions of real world politics like persistent inequalities among people and relative lack of basic freedoms. This situates Sen’s work at the intersection of political theory and social science.
Sen’s book is in many ways an introduction to and summary of work done over the past 50 years. Readers who are already familiar with Sen will find here a compelling elaboration and systematization of the many topics that Sen has worked on: economics, philosophy, and, it seems, all of the social sciences which lie in between. New readers will be dazzled by the copious references not only to Sen’s own work, but by the near encyclopedic knowledge Sen evinces of work done in the past 50 years in the social sciences and political philosophy. Sen peppers his pages with references to friends and colleagues, especially at Harvard, Oxford and Cambridge, who have influenced him and who, inevitably, have been influenced by him. His style in these references is chatty and warm, and further contributes to the impression that what we have in these pages is a disagreement only within liberalism, which, in different forms, Sen believes has carried the day. (The only prominent references to a non-liberal thinker that I could detect are to the late G. A. Cohen, whose own recent book takes issues with the central tenets of Rawls’ theory.)
Sen has many philosophical heroes but chief among these is Adam Smith who, for Sen, himself an avowed child of the enlightenment, represents a sort of counter current to the Kantian enlightenment, which Sen sees Rawls embodying in his emphasis on abstraction. This counter-current includes thinkers like Mary Wollstonecraft, the Marquise de Condorcet, Jeremy Bentham, John Stuart Mill and even Karl Marx. What all of these thinkers have in common, according to Sen, is their comparative rather than foundationalist approach. Thus, Smith’s concept of the impartial spectator is championed as a comparative version of Kant’s foundational categorical imperative. The difference between the two is that the impartial spectator, while impartial, proceeds by comparison of different perspectives rather than by moral judgment concerning the correct position to take, as the Kantian categorical imperative does. Smith’s view is meant to permit us to see other positions without judging them right or wrong, but rather giving the spectator an appreciation of the relative advantages and disadvantages of a particular comprehensive position.
Sen’s thought, as he makes clear on several occasions, is also influenced by Indian philosophical thought. Here the distinction between niti and nyaya conceptions of justice is particularly important to him. The niti conception of justice is characterized by “organizational propriety and behavioral correctness”, while the nyaya conception “stands for a comprehensive concept of realizing justice” (20). And with this, the centralconcern of theory has been named: Sen is concerned with realization of justice rather than the determination of its pure definition. Sen writes: “an approach to justice can be both entirely acceptable in theory and eminently useable in practice, even without its being able to identify the demands of perfectly just societies.” (401) What Sen wants, in other words, is a theory that is both responsive to the needs of justice and responsive to real world problems.
For someone who has worked as extensively on real world problems, from the Bengal famine of 1943, which Sen witnessed first hand, to the problems of the ‘missing’ generation of women in Asia (due to the impact of inequality on life expectancy of children and newborns), Rawls’ idealized theory of justice surely leaves something to be desired. The problem with a procedure like Rawls’ original position is that, as Senuts it, “there is still a large question about how the chosen institutions would work in a world in which everyone’s actual behavior may or may not come fully into line with the identified reasonable behavior [stipulated by Rawls].” (68) The problem, however, is not only that people are not sufficiently rational, but also, that formulating the question of justice in the way Rawls does means that certain important features of justice will be neglected. These are, as I have noted, the agent’s real position and her real possibilities.
Sen develops this theory out of the social choice model developed by Kenneth Arrow. While much social contract theory depends on an idealized conception of rational agents, the social choice model, as developed by Sen, relies on our actual and only somewhat rational ability to rank different outcomes based on our particular situation. Such ranking, for Sen, means that we can weigh outcomes that are not ideal againsteach other, regardless of how close to the ideal outcome they get. Social choice theory thus makes room for incompleteness.
Sen identifies several features of the social choice framework. (1) Focus on the comparative, not just the transcendental. For Sen this means that “a theory of justice must have something to say about the choices that are actually on offer”. (2) Recognition of the inescapable plurality of competing principles. This means that there will be different conceptions of freedom that can be employed in social choice theory and this, in turn, means that there will be different and potentially incompatible outcomes of social choice theory. (3) Permissibility of partial resolution. This means that the result of social choice may be tentatively incomplete (a work in progress) or that it may arrive at an impasse which is conceptually accounted for, though also open to further revision. (4) Social choice theory reflects diversity of input which means that social choice can give us functional connection between individual rankings and priorities on the one hand and results relevant to social policy on the other. These too will be helpful in determining a closer approximation to justice. (5) Finally, social choice, even if it yields conclusions which conflict, tells us a great deal about current positions and thus should help us come up with better solutions than the ones we already have (106-111).
Sen argues that this framework can help us evaluate relative freedom and justice in terms of outcome as well as in terms of agency. Avowing that he shares Rawls’ commitment to the priority of liberty (62), Sen argues that justice in actual human affairs is not simply a matter of cumulative outcome (what results) but also of comprehensive outcome (what results and how it is brought about), as in Rawls’ proceduralism. Sen thus sides with Rawls against economists and utilitarians who tend to see human freedom merely as welfare. Sen’s approach of freedom as capability holds that freedom is valuable for at least two reasons: freedom must provide opportunity to pursue our ends, and freedom must give us a choice about which opportunities to pursue (228). Sen calls this capability. The approach is pluralist in the sense that it “points to an informational focus in judging and comparing overall individual advantages” without specifying how
information may be used (232). It does not specify any particular ‘primary goods’ as Rawls does. Its second decisive feature is that it “is inescapably concerned with a plurality of different features of our lives and concerns.” (233) These are what Sen calls functionings. Sen writes: “The capability that we are concerned with is our ability to achieve various combinations of functionings that we can compare and judge againsteach other in terms of what we have reason to value.” (233) Freedom as capability thus seeks to do justice to the diversity of human positions as well as to the diversity of human interests.
But how are these different interests to be coordinated? Sen’s answer is: through public debate. For Sen the model of debate is essentially given by the comparative model of Smith’s impartial spectator, as I have noted. The impartial spectator permits open and open-ended discussion, hence the opportunity for people to revise their views based on their changing positions and changing information. This, for Smith, is taken as a fairly unproblematic point.
It is difficult to do justice to such a synthetic work in a few pages, and perhaps more difficult to launch any substantive criticisms of it. However, philosophy being what it is, it is perhaps worth dwelling for a few moments at least on what Sen’s book is and what it is not. As I said at the outset, Sen’s book is not concerned with the grounding of its theory in any meta-ethical sense, that is, Sen does not seek to give a final justification of his pluralist approach to liberty nor to his claim that the impartial spectator is the best
sort of model for deliberation. He does argue, however, that the impartial spectator can help us become more clear about the ways in which human agency could be more fully realized in terms of both functionings and capability. He has, however, also argued that Rawls’ conception of the original position is idealized in a way that makes it unsuitable to tackle real world problems. This too I will not dispute. Sen’s book is a work that seeks to form a bridge between philosophy and social policy in the sense that it argues for the methodology of the capability approach largely from a mixture of philosophical and empirical observation but does not try to systematize these in a philosophical foundational way.
For all of Sen’s criticism of Rawls there is one concept that Sen barely mentions: the concept of the reflective equilibrium. The reflective equilibrium, which in Rawls is meant to ground the whole theory of justice, is the idea that we can abstract from our own perspectives, check these against those of others and adjust our own views accordingly. The reflective equilibrium, in my view, is remarkably close to Sen’s own view of public deliberation. When all is said and done, Sen’s and Rawls’ theories bare remarkable similarity to one another when it comes to their philosophical commitments about the role of rationality and human agency. What they share is a rejection of Kant’s attempt to provide a metaphysical foundation for morality. This means that they have, from a philosophical point of view, more in common than Sen may think.
Resource: Public Reason, Vol. 2, No. 2, December 2010