The Idea of Justice

by Amartya Sen

  Even though the first part may suggest that this is a straightforward demand for non-discriminatory institutions, which need not be conditional on behavioural norms, it is plausible to think that the requirements of ‘fair equality of opportunity’ could give a much greater role to behavioural features (for example, what kind of selection criteria would be effective given behavioural characteristics, and so on) in determining the appropriate choice of institutions.

  When we turn to the second part of this principle for institutional choice (the important requirement that goes by its own name of ‘the Difference Principle’), we have to examine how the different potential institutional arrangements would mesh with, and interact with, behavioural norms standard in the society. Indeed, even the language of the difference principle reflects the involvement of this criterion with what would actually happen in the society (that is, whether inequalities will work out to be of ‘the greatest benefit of the least-advantaged members of society’). Again, this gives Rawls much more room to build in sensitivity to behavioural differences.

  78

  i n s t i t u t i o n s a n d p e r s o n s b e h a v i o u r a l r e s t r i c t i o n

  t h r o u g h c o n t r a c t a r i a n

  r e a s o n i n g

  There is, however, also a second issue that is relevant in discussing the relation between actual behaviour and the choice of institutions.

  This issue, which was introduced in the last chapter, concerns Rawls’s presumption that once the social contract has been arrived at, people would abandon any narrow pursuit of self-interest and follow instead the rules of behaviour that would be needed to make the social contract work. Rawls’s idea of ‘reasonable’ behaviour extends to the actual conduct that can be presumed once those chosen institutions –

  unanimously chosen in the original position – have been put in place.5

  Quite demanding assumptions are made by Rawls on the nature of post-contract behaviour. He puts the issue thus in Political Liberalism: Reasonable persons . . . desire for its own sake a social world in which they, as free and equal, can cooperate with others on terms all can accept. They insist that reciprocity should hold within that world so that each benefits along with others. By contrast, people are unreasonable in the same basic aspect when they plan to engage in cooperative schemes but are unwilling to honour, or even to propose, except as a necessary public pretense, any general principles or standards for specifying fair terms of cooperation. They are ready to violate such terms as suits their interests when circumstances allow.6

  By assuming that actual behaviour in the post-social contract world would incorporate the demands of reasonable behaviour in line with the contract, Rawls makes the choice of institutions that much simpler, since we are told what to expect in the behaviour of individuals once the institutions are set in place.

  Rawls cannot, then, be accused in any way of either inconsistency or incompleteness in presenting his theories. The question that remains, however, is how this consistent and coherent political model will translate into guidance about judgements of justice in the world in which we live, rather than in the imagined world with which Rawls is here primarily concerned. Rawls’s focus does indeed make sense, if the intention is to outline how to achieve the perfectly just social 79

  t h e i d e a o f j u s t i c e arrangements and, with the additional help of reasonable behaviour, a totally just society.* But this makes the distance between transcendental thinking and comparative judgements of social justice, on which I commented in the Introduction, that much larger and more problematic.

  There is a real similarity here between Rawlsian presumptions about reasonable behaviour following the presumed agreements in the original position, and Ashoka’s vision of a society led by right behaviour (or dharma), except that in Rawls’s critical hands we get a much fuller picture of how things are supposed to work out in a world that we can try to get to, taking note of the dual role of institutions and behaviour.

  This can be seen as an important contribution to thinking about transcendental justice seen on its own. Rawls outlines his idealized transcendental vision for institutions and behaviours with force and clarity: Thus very briefly: i) besides a capacity for a conception of the good, citizens have a capacity to acquire conceptions of justice and fairness and a desire to act as these conceptions require; ii) when they believe that institutions and social practices are just, or fair (as these conceptions specify), they are ready and willing to do their part in those arrangements provided they have reasonable assurance that others will also do their part; iii) if other persons with evident intention strive to do their part in just or fair arrangements, citizens tend to develop trust and confidence in them; iv) this trust and confidence becomes stronger and more complete as the success of cooperative arrangements is sustained over a longer time; and v) the same is true as the basic institutions framed to secure our fundamental interests (the basic rights and liberties) are more firmly and willingly recognized.7

  * There is, however, an important issue here with regard to the adequacy of Rawlsian theory for the characterization of transcendental justice because of the concession that Rawls makes to inequalities needed to cater for the demands of incentives. If we accept G. A. Cohen’s argument, presented in Rescuing Justice and Equality (Cambridge, MA: Harvard University Press, 2008), that this makes Rawls’s theory quite unsatisfactory as a theory for perfect justice since there should be no concession to inequality to cajole people to behave right (what they should do even without personal incentives, in a just world), then surely the substantive content of the Rawlsian theory of perfect justice would be undermined. As was discussed in the last chapter, there is an important issue of theory here, since Rawls does make strong behavioural demands on individual conduct in the post-contract world, but exempts the need for ideal behaviour without incentives by accommodating incentives in the social contract itself.

  80

  i n s t i t u t i o n s a n d p e r s o n s This vision is both illuminating and in many ways hugely inspiring. And yet if we are trying to wrestle with injustices in the world in which we live, with a combination of institutional lacunae and behavioural inadequacies, we also have to think about how institutions should be set up here and now, to advance justice through enhancing the liberties and freedoms and well-being of people who live today and will be gone tomorrow. And this is exactly where a realistic reading of behavioural norms and regularities becomes important for the choice of institutions and the pursuit of justice.

  Demanding more from behaviour today than could be expected to be fulfilled would not be a good way of advancing the cause of justice.

  This basic realization must play a part in the way we think about justice and injustice today, and it will figure in the constructive work that follows in the rest of the book.

  p o w e r a n d t h e n e e d

  f o r

  c o u n t e r va i l i n g

  This is perhaps also the place where we must take note of a basic insight of John Kenneth Galbraith on the nature of appropriate social institutions that society may need. Galbraith was very aware of the negative influence of unchecked power, both because institutional balance is very important for society, but also because power corrupts.

  He argued for the importance of distinct social institutions that could exercise ‘countervailing power’ over each other. This requirement and its relevance are spelt out in Galbraith’s 1952 book, American Capitalism, which also provides an unusual and illuminating account of how the success of American society is deeply dependent on the operation of the power of a multiplicity of institutions that check and balance the force and possible domination that might otherwise be exercised by one institution.8

  Galbraith’s analysis has much to offer on what has tended to go rather badly wrong in recent years in the USA as the executive branch has tried to exercise more unrestrained power than the American Constitution would seem to have intended. But even more strikingly, it also
tells us a lot about what goes wrong in one-party states with 81

  t h e i d e a o f j u s t i c e commanding central control, like the former Soviet Union. Despite the early political enthusiasm and justice-related expectations that the October Revolution generated, huge political and economic failures soon came to characterize the USSR (including the purges, the show trials, the Gulags as well as its bureaucracy-dominated non-functioning economic and social institutions). The origin of these failures can, at least partly, be traced, I would argue, to the complete absence of countervailing powers in the Soviet institutional structure.

  The issue relates obviously to the absence of democracy, a subject to which I will return later on (in Chapter 15 ‘Democracy as Public Reason’). The issue of democratic practice can be closely linked with the existence and use of countervailing power in a society with a plurality of sources of voice and strength.

  i n s t i t u t i o n s a s f o u n d a t i o n s Any theory of justice has to give an important place to the role of institutions, so that the choice of institutions cannot but be a central element in any plausible account of justice. However, for reasons already discussed, we have to seek institutions that promote justice, rather than treating the institutions as themselves manifestations of justice, which would reflect a kind of institutionally fundamentalist view. Even though the arrangement-centred perspective of niti is often interpreted in ways that make the presence of appropriate institutions themselves adequate to satisfy the demands of justice, the broader perspective of nyaya would indicate the necessity of examining what social realizations are actually generated through that institutional base. Of course, the institutions themselves can sensibly count as part of the realizations that come through them, but they can hardly be the entirety of what we need to concentrate on, since people’s lives are also involved.*

  * Justice Stephen Breyer has brought out with much force and clarity the importance of paying ‘attention to purpose and consequence’ in interpreting a democratic constitution, emphasizing the role of ‘consequences as an important yardstick to measure a given interpretation’s faithfulness to these democratic purposes’ ( Active Liberty: Interpreting Our Democratic Constitution (New York: Knopf, 2005), p. 115).

  82

  i n s t i t u t i o n s a n d p e r s o n s There is a long tradition in economic and social analysis of identifying the realization of justice with what is taken to be the right institutional structure. There are a great many examples of such a concentration on institutions, with powerful advocacy for alternative institutional visions of a just society, varying from the panacea of wonderfully performing free markets and free trade to the Shangri-La of socially owned means of production and magically efficient central planning. There are, however, good evidential reasons to think that none of these grand institutional formulae typically deliver what their visionary advocates hope, and that their actual success in generating good social realizations is thoroughly contingent on varying social, economic, political and cultural circumstances.9 Institutional fundamentalism may not only ride roughshod over the complexity of societies, but quite often the self-satisfaction that goes with alleged institutional wisdom even prevents critical examination of the actual consequences of having the recommended institutions. Indeed, in the purely institutional view, there is, at least formally, no story of justice beyond establishing the ‘just institutions’. Yet, whatever good institutions may be associated with, it is hard to think of them as being basically good in themselves, rather than possibly being effective ways of realizing acceptable or excellent social achievements.

  All this would seem to be easy enough to appreciate. And yet institutionalist fundamentalism is very often implied by the nature of the chosen institution-focused advocacy, even in political philosophy.

  For example, in his deservedly famous exploration of ‘morals by agreement’, David Gauthier relies on agreements between different parties that take the form of accord on institutional arrangements, and this is supposed to take us all the way to social justice. The institutions are given an overwhelming priority – a priority that may seem immune to the nature of the actual consequences generated by the agreed institutions. As it happens, Gauthier relies heavily on the market economy doing its job in producing efficient arrangements, on which the parties seeking agreement are imagined to focus, and once the ‘right’ institutions have been set up, we are supposed to be in the secure hands of these institutions. Gauthier argues lucidly that the setting up of the right institutions liberates the parties from having to be constantly constrained by morality as well. The chapter of 83

  t h e i d e a o f j u s t i c e Gauthier’s book where all this is explained is aptly named ‘The Market: Freedom from Morality’.10

  Giving such a foundational role to institutions for the assessment of social justice, in the way that David Gauthier does, may be somewhat exceptional, but there are many other philosophers who have been clearly tempted in that direction. There is evidently considerable attraction in assuming institutions to be inviolable once they are imagined to be rationally chosen by some hypothetical just agreement, irrespective of what the institutions actually achieve. The general point at issue here is whether we can leave matters to the choice of institutions (obviously chosen with an eye to results to the extent that they enter the negotiations and agreements) but without questioning the status of the agreements and of the institutions once the arrangements have been chosen, no matter what the actual consequences prove to be.*

  There are some theories, which do not take the form of being institutionally fundamentalist in the way that Gauthier’s theory is, but which share the priority of chosen institutions over the nature of the outcomes and realizations. For example, when Robert Nozick argues for the necessity, for reasons of justice, of guaranteeing individual liberties, including the rights of property ownership, free exchange, free transfer and free inheritance, he makes the institutions needed for these rights (the legal as well as the economic framework) essential requirements of his vision of a just society.11 And he is ready to leave matters in the hands of these institutions rather than calling for any revision based on an assessment of outcomes (no ‘patterning’ of outcome is allowed in his theory, at least in its pure form). Formally, there is still a difference between valuing the institutions themselves and seeing an institution as being essential to justice because of its being necessary for the realization of something else, such as the

  ‘rights’ of people, as in the Nozickian system. The distinction is, however, rather formal, and it would not be entirely misleading to see Nozick’s theory to be derivatively fundamentalist about institutions.

  * The advocacy for a market economy need not, however, ignore the conditional nature of the support; see, for example, John Gray’s strong defence of the market as an institution which takes a consequence-contingent form ( The Moral Foundations of Market Institutions (London: IEA Health and Welfare Unit, 1992)).

  84

  i n s t i t u t i o n s a n d p e r s o n s But what if the collectivity of what are taken to be ‘just institutions’

  generates terrible results for the people in that society (without actually violating their immediate concerns, such as the guarantee of libertarian rights, as in Nozick’s case)?* Nozick did recognize that there could be a problem here. Indeed, he proceeded to make a possible exception to the case in which the system advocated by him, with complete priority of libertarian rights, would lead to what he called

  ‘catastrophic moral horror’.† The institutional requirements might well be dropped in those extreme cases. But once such an exception is made, it is not clear what remains of the basic priorities in his theory of justice, and the fundamental place that is given to the necessary institutions and rules within that theory. If catastrophic moral horrors are adequate for abandoning the reliance on the allegedly right institutions altogether, could it be the case that bad social consequences that are not absolutely catastrophic but still quite nasty might be adequ
ate grounds for second-guessing the priority of institutions in less drastic ways?

  The more general issue, of course, is the basic unreliability of not being constantly sensitive to what actually happens in the world, no matter how excellent the institutions are taken to be. Even though John Rawls is quite clear in motivating the discussion on institutions in terms of the social structure they promote, nevertheless, through defining his ‘principles of justice’ entirely in institutional terms, Rawls too goes some distance towards a purely institutional view of justice.‡

  * It can be shown that economic and political forces that generate even gigantic famines can work to yield that result without violating anyone’s libertarian rights. On this, see my Poverty and Famines: An Essay and Entitlement and Deprivation (Oxford; Oxford University Press, 1981). See also Chapter 1, ‘Reason and Objectivity’. See Cormac O

  ´ Gra´da, Ireland’s Great Famine: Interdisciplinary Perspectives (Dublin: University College Dublin Press, 2006).

  † Nozick does, however, leave the question open: ‘the question of whether these side constraints reflecting rights are absolute, or whether they may be violated in order to avoid catastrophic moral horror, and if the latter, what the resulting structure might look like, is one I hope largely to avoid’ (Robert Nozick, Anarchy, State and Utopia (Oxford: Blackwell, 1974), p. 30).

  ‡ It is, of course, true that in the Rawlsian system of ‘justice as fairness’ institutions are chosen with an eye to results. But once they are chosen through ‘the principles of justice’, there is no procedure within the system to check whether the institutions are, in fact, generating the anticipated results.

  85

  t h e i d e a o f j u s t i c e So do a number of other leading theorists of justice through their ultimate reliance on the soundness of the institutions they recommend on the basis of how they are expected to operate.