The Idea of Justice

by Amartya Sen

  A different take on the same issue is that it is hard to imagine that the need for incentives can be made to go away on the basis of the expectation that the conception of justice in the original position will make everyone spontaneously play their full productive role with no

  * Cf. ‘Everyone is presumed to act justly and to do his part in upholding just institutions’ (Rawls, A Theory of Justice, p. 8).

  61

  t h e i d e a o f j u s t i c e incentive arrangements. Cohen may well be right that a society that can be seen as perfectly just should not have the impediment of incentive-based inequality, but that is one more reason for not concentrating so much on transcendental justice in developing a theory of justice. Rawls’s halfway house may not be quite transcendental enough for Cohen, but there are other problems of transcendental concentration (for reasons already discussed) that Rawls has to face, even in the absence of taking a leaf from Cohen’s book. In a world of comparative justice, Cohen’s just world may stand above the one that Rawls outlines in justice as fairness, but the main use of the theory of comparative justice would be in making comparisons between feasible possibilities less exalted – in terms of justice – than both Cohen’s and Rawls’s ‘just’ worlds.

  s o m e p o s i t i v e l e s s o n s f r o m t h e r aw l s i a n a p p r o a c h

  It is not hard to see that there are some contributions of great importance in the Rawlsian approach to justice as fairness and in the way Rawls has presented and explicated its implications. First, the idea that fairness is central to justice, which is illuminatingly defended by Rawls, is a major avowal that takes us well beyond the understanding generated by the previous literature on the subject of justice (for example, the justificatory basis of the Benthamite utilitarian theory).

  Even though I do not believe that the impartiality captured in the reflective device of the ‘original position’ (on which Rawls greatly relies) is adequate for the purpose, this is in no way a rebellion against the basic Rawlsian idea of the foundational priority of fairness in developing a theory of justice.

  Second, I must reiterate a point I have already made concerning the far-reaching importance of Rawls’s thesis about the nature of objectivity in practical reason, in particular his argument that ‘The first essential is that a conception of objectivity must establish a public framework of thought sufficient for the concept of judgement to apply and for conclusions to be reached on the basis of reasons and evidence after discussion and due reflection.’10 The issue was discussed fairly 62

  r aw l s a n d b e y o n d

  extensively in Chapter 1 (‘Reason and Objectivity’) and I will not belabour it further here.

  Third, aside from clarifying the need for the idea of fairness as preceding justice, Rawls makes another basic contribution in pointing to ‘the moral powers’ that people have, related to their ‘capacity for a sense of justice’ and ‘for a conception of the good’. This is a far cry from the imagined world, on which exclusive attention is showered on some versions of ‘rational choice theory’ (to be more fully discussed in Chapter 8, ‘Rationality and Other People’), in which human beings only have a sense of self-interest and prudence but evidently not any capacity or inclination to consider ideas of fairness and justice.11 Aside from enriching the concept of rationality, Rawls also pursues the distinction between being ‘rational’ and being ‘reasonable’ in a very useful way,12

  and that is a distinction that will be used fairly extensively in this work.

  Fourth, Rawls’s prioritization of liberty, admittedly in the rather extreme form of its total priority, does draw attention to the strong case for seeing liberty as a separate and, in many ways, overriding concern in the assessment of the justice of social arrangements. Liberty also, of course, works alongside other concerns in determining a person’s overall advantage: it is included in the list of ‘primary goods’

  specified by Rawls as a part of the picture of individual advantage for use in his difference principle. But going much beyond that shared role with other concerns as a primary good, liberty also has, more selectively, an additional status, which has an importance of its own.

  Giving a special place – a general pre-eminence – to liberty goes well beyond taking note of the importance of liberty as one of many influences on a person’s overall advantage. While personal liberty is indeed useful, like income and other primary goods, that is not all that is involved in its importance. It is a central concern both in a person’s freedom, touching the most private aspects of personal life, and it is also a basic necessity (for example, in the form of freedom of speech) for the practice of public reasoning, which is so crucial to social evaluation.* The reasoned perception of the importance of

  * On the various ways in which liberty, including freedom of expression, is crucially important for justice, see also Thomas Scanlon, The Difficulty of Tolerance (Cambridge: Cambridge University Press, 2003).

  63

  t h e i d e a o f j u s t i c e personal liberty has, not surprisingly, moved people to defend and fight for it over the centuries. By separating out the importance of liberty shared by all, Rawls draws attention to a distinction – between liberty and other helpful facilities – that is really important to note and pursue.13

  Fifth, by insisting on the need for procedural fairness under the first part of the second principle, Rawls provided a significant enrichment of the literature on inequality in the social sciences, which has often tended to concentrate too exclusively on disparities in social status or economic outcomes, while ignoring disparities in the processes of operation, for example, those associated with excluding people from offices on grounds of their race or colour or gender.*

  Sixth, after giving liberty its due and after recognizing the need to have openness in allowing people to compete equitably for offices and positions, the difference principle indicates the importance of equity in social arrangements so that attention is drawn particularly to the predicament of the worst-off people.14 The removal of poverty measured in terms of the deprivation of primary goods is given a big place in Rawls’s theory of justice, and this Rawlsian focus has indeed been powerfully influential on the analysis of public policy for poverty removal.

  Finally (though this is very much my own reading, which others may or may not find to be a good interpretation of Rawls), by focusing on ‘primary goods’ (that is, on the general-purpose means for the pursuit of one’s comprehensive goals), Rawls gives indirect acknowledgement to the importance of human freedom in giving people real

  – as distinct from only formally recognized – opportunity to do what they would like with their own lives. I shall argue later on, in Chapters 11 and 12, that the fit between a person’s holding of primary goods and the substantive freedoms that the person can in fact enjoy, can be very imperfect, and that this problem can be addressed through focusing instead on the actual capabilities of people.15 And yet by

  * One of the reasons for the extraordinarily positive response that the election of Barack Obama to the presidency of the United States has received across the world is its demonstration of the weakening of the racial barrier in the politics of the country.

  This is a distinct issue from the evident suitability of Obama himself as a visionary leader, irrespective of his racial background.

  64

  r aw l s a n d b e y o n d

  instrumentally highlighting the importance of human freedom, Rawls has, I would argue, given a definitive place to freedom-related thinking within the main corpus of his theory of justice.*

  p r o b l e m s t h a t c a n b e

  a d d r e s s e d

  e f f e c t i v e l y

  There are, however, problems and difficulties as well. Let me begin with a couple of problems that are important, but which can be, I believe, accommodated without going against Rawls’s basic approach, and which have already been receiving considerable attention in the literature.

  First, it has been argued that the total pr
iority of liberty is too extreme. Why should we regard hunger, starvation and medical neglect to be invariably less important than the violation of any kind of personal liberty? That question was first raised powerfully by Herbert Hart shortly after Rawls’s Theory of Justice was published,16 and in his later works (particularly in his Political Liberalism), Rawls himself has gone some distance towards making the priority, in effect, less extreme.17 It is indeed possible to accept that liberty must have some kind of priority, but total unrestrained priority is almost certainly an overkill. There are, for example, many different types of weighting schemes that can give partial priority to one concern over another.†

  Second, in the difference principle, Rawls judges the opportunities

  * Similarly, Philippe Van Parijs’s powerful arguments for a basic income for everyone draws on its role in advancing each person’s freedom; see his Real Freedom for All: What (If Anything) Can Justify Capitalism (Oxford: Clarendon Press, 1995).

  † There is a mathematical issue of ‘weighting’ that might have had a role in influencing Rawls in the direction of the total lexicographic priority he gives to liberty. Rawls clearly found it wrong to put no more emphasis on liberty than on other facilities for human flourishing. This, it would appear, made him give liberty irresistible priority in every case of conflict, which seems much too strong, if my reading of Rawls’s intention is right. In fact, the mathematics of weighting allows many intermediate positions of higher weighting of liberty (with varying degrees of intensity). Some of the methods of using weights much more flexibly are discussed in my Choice, Welfare and Measurement (1982), particularly in essays 9–12. There are many different ways of attaching some priority to one concern over another, without making that priority totally unbeatable under any circumstances (as implied by the ‘lexical’ form chosen by Rawls).

  65

  t h e i d e a o f j u s t i c e that people have through the means they possess, without taking into account the wide variations they have in being able to convert primary goods into good living. For example, a disabled person can do far less with the same level of income and other primary goods than can an able-bodied human being. A pregnant woman needs, among other things, more nutritional support than another person who is not bearing a child. The conversion of primary goods into the capability to do various things that a person may value doing can vary enormously with differing inborn characteristics (for example, propensities to suffer from some inherited diseases), as well as disparate acquired features or the divergent effects of varying environmental surroundings (for example, living in a neighbourhood with endemic presence, or frequent outbreaks, of infectious diseases). There is, thus, a strong case for moving from focusing on primary goods to actual assessment of freedoms and capabilities.* However, if my reading of Rawls’s motivation in using primary goods is right (that is, to focus indirectly on human freedom), then I would argue that a move from primary goods to capabilities would not be a foundational departure from Rawls’s own programme, but mainly an adjustment of the strategy of practical reason.†

  d i f f i c u l t i e s t h a t n e e d

  f r e s h

  i n v e s t i g a t i o n

  The problems discussed in the last section have received considerable attention, and continue to do so. While they have not been fully

  * On this, see my ‘Equality of What?’ in S. McMurrin (ed.), Tanner Lectures on Human Values, vol. I (Cambridge: Cambridge University Press, and Salt Lake City, UT: University of Utah Press, 1980); Commodities and Capabilities (Amsterdam: North-Holland, 1985); Inequality Reexamined (Cambridge, MA: Harvard University Press, and Oxford: Oxford University Press, 1992); jointly with Martha Nussbaum (eds)., The Quality of Life (Oxford: Clarendon Press, 1993). The underlying issues are taken up in Chapters 11, ‘Lives, Freedoms and Capabilities’, and 12, ‘Capabilities and Resources’ of this work.

  † See Philippe Van Parijs, Real Freedom for All (1995) on the strategic advantage in making use of the instrument of income even when the basic objective is to advance freedom. See also Norman Daniels, Just Health (2008).

  66

  r aw l s a n d b e y o n d

  resolved, there is reason to think that their central points are, by now, reasonably clear and understood. They will not be neglected in the rest of the book, but I will suggest that more immediate attention is needed to clarify some other problems with the Rawlsian approach that have not figured much in the ongoing literature.

  (1) The Inescapable Relevance of Actual Behaviour First, the exercise of fairness through the approach of social contract is geared, in the Rawlsian case, to identifying only the ‘just institutions’, through arriving at ‘an agreement on the principles that are to regulate the institutions of the basic structure itself from the present into the future’.18 In the Rawlsian system of justice as fairness, direct attention is bestowed almost exclusively on ‘just institutions’, rather than focusing on ‘just societies’ that may try to rely on both effective institutions and on actual behavioural features.

  Samuel Freeman, who with Erin Kelly has done a great job of gathering together and editing Rawls’s extensive writings, summarizes Rawls’s strategy in ‘justice as fairness’ in the following way: Rawls applies the idea of a hypothetical social agreement to argue for principles of justice. These principles apply in the first instance to decide the justice of the institutions that constitute the basic structure of the society.

  Individuals and their actions are just insofar as they conform to the demands of just institutions . . . How [these institutions] are specified and integrated into a social system deeply affects people’s characters, desires and plans, and their future prospects, as well as the kind of persons they aspire to be. Because of the profound effects of these institutions on the kinds of persons we are, Rawls says that the basic structure of society is ‘the primary subject of justice’.19

  We can see how different this niti-centred approach is from any nyaya-based approach to justice, for example that of social choice theory (see the contrast drawn in the Introduction). The latter would tend to ground the assessment of combinations of social institutions and public behaviour patterns on the social consequences and realizations they yield (taking note inter alia of any intrinsic importance that particular institutions and behaviour patterns may have within the social realizations to be assessed).

  67

  t h e i d e a o f j u s t i c e Two issues in particular in this comparison deserve special attention. First, the understanding of justice as nyaya cannot neglect the actual social realizations that may be expected to emerge from any choice of institutions, given other social features (including actual behaviour patterns). What really happens to people cannot but be a central concern of a theory of justice, in the alternative perspective of nyaya (without ignoring any intrinsic valuation that may be reasonably attached to having institutions and behaviour norms that are also seen as important on their own).

  Second, even if we do accept that the choice of basic social institutions through a unanimous agreement would yield some identification of ‘reasonable’ behaviour (or ‘just’ conduct), there is still a large question about how the chosen institutions would work in a world in which everyone’s actual behaviour may or may not come fully into line with the identified reasonable behaviour. The unanimous choice of the principles of justice is ground enough, Rawls argues, for their forming a ‘political conception’ of justice that all accept, but that acceptance may still be a far cry from the actual patterns of behaviour that emerge in any actual society with those institutions. Since no one has argued more powerfully and more elaborately than John Rawls for the need for ‘reasonable’ behaviour by individuals for a society to function well, he is clearly very aware of the difficulty in presuming any kind of spontaneous emergence of universal reasonable behaviour on the part of all members of a society.

  The question to ask, then, is this: if the justice of what happens in a society depends on a combination of institutional
features and actual behavioural characteristics, along with other influences that determine the social realizations, then is it possible to identify ‘just’ institutions for a society without making them contingent on actual behaviour (not necessarily the same as ‘just’ or ‘reasonable’ behaviour)? The mere acceptance of some principles as forming the right ‘political conception of justice’ does not resolve this issue if the theory of justice sought has to have any kind of applicability in guiding the choice of institutions in actual societies.

  Indeed, we have good reasons for recognizing that the pursuit of justice is partly a matter of the gradual formation of behaviour patterns – there is no immediate jump from the acceptance of some 68

  r aw l s a n d b e y o n d

  principles of justice and a total redesign of everyone’s actual behaviour in line with that political conception of justice. In general, the institutions have to be chosen not only in line with the nature of the society in question, but also co-dependently on the actual behaviour patterns that can be expected even if – and even after – a political conception of justice is accepted by all. In the Rawlsian system, the choice of the two principles of justice is meant to ensure both the right choice of institutions as well as the emergence of appropriate actual behaviour on the part of everyone, making individual and social psychology thoroughly dependent on a kind of political ethics. Rawls’s approach, developed with admirable consistency and skill, does involve a formulaic and drastic simplification of a huge and multi-faceted task –

  that of combining the operation of the principles of justice with the actual behaviour of people – which is central to practical reasoning about social justice. This is unfortunate since it can be argued that the relationship between social institutions and actual – as opposed to ideal – individual behaviour cannot but be critically important for any theory of justice that is aimed at guiding social choice towards social justice.*