The Idea of Justice

by Amartya Sen

  Some of these motivations, like ‘humanity, justice, generosity and public spirit’, may even be very productive for society, as Smith noted.*

  There tends to be, however, more resistance to accepting the possibility that people may have good reasons even to go beyond the pursuit of their own goals (whether or not the goals themselves are based only on self-interest). The argument runs: if you are consciously not pursuing what you think are your goals, then clearly those cannot actually be your goals. Indeed, many authors have taken the view that the claim that one can have reason not to be confined to the pursuit of one’s goals is ‘nonsensical since even strongly heterogeneous or altruistic agents cannot pursue other people’s goals without making them their own’.†

  The point to note here is that in denying that rationality demands (RCT) in its broadest form. I have discussed the reach and the limits of the critique presented by Jolls, Sunstein and Thaler in the introductory essay to Rationality and Freedom: ‘Introduction: Rationality and Freedom’ (Cambridge, MA: Harvard University Press, 2002), pp. 26–37.

  * The Theory of Moral Sentiments, p. 189. Smith considers various reasons for giving room to a variety of such motivations, including moral appeal and behavioural grace as well as their social usefulness.

  † This is how Fabienne Peter and Hans Bernhard Schmid summarize a line of critique of departing from ‘self-goal choice’ in their introductory essay to a very interesting collection of papers on this and related themes: ‘Symposium on Rationality and Commitment: Introduction’, Economics and Philosophy, 21 (2005), p. 1. My treatment of this objection draws on my response to a larger collection of essays put together by Peter and Schmid (with their own important contributions on this subject) included in that volume: ‘Rational Choice: Discipline, Brand Name and Substance’, in Fabienne Peter and Hans Bernhard Schmid (eds), Rationality and Commitment (Oxford: Clarendon Press, 2007).

  191

  t h e i d e a o f j u s t i c e that you must act single-mindedly according to your own goals (subject only to constraints that are not self-imposed), you do not necessarily dedicate yourself to the promotion of the goals of others. We can reason our way towards following decent rules of behaviour that we see as being fair to others as well, which can restrain the unique dominance of single-minded pursuit of our own goals. There is nothing particularly mysterious about our respect for sensible rules of conduct, which can qualify the pursuit of what we rightly – and reasonably – see as goals that we would in general like to advance.

  Consider an example of such restraint that does not force us to take on other people’s goals as our ‘real goals’. You happen to be occupying a window seat in a plane journey, with the window shade up on a sunny day. That is when you hear the occupant of the aisle seat next to you requesting you to pull down the shade (‘if you would, please’) so that he can see his computer screen better, to be able to devote himself fully to playing some computer game. You know that game and it is in your view a ‘plainly silly’ game (‘a great waste of time’).

  You are, in general, frustrated that there is so much ignorance around, with so many people playing inane games rather than reading the news

  – boning up on what is actually happening in Iraq, or Afghanistan, or for that matter in your home town. You decide, nevertheless, to behave well and comply with the game-enthusiast’s request, and you oblige him by pulling the shutter down.

  What can we say about your choice? There is no difficulty in understanding that you are not averse to helping your neighbour – or anyone else – pursue his or her well-being, but it so happens that you do not think that your neighbour’s well-being is, in fact, best advanced by his wasting his time – and by your helping him to waste his time –

  on a very silly game. You remain, in fact, entirely willing to lend him your copy of the New York Times, reading which would be, you are convinced, much better for your neighbour’s edification and well-being. Your action is not a corollary of any general pursuit of well-being.

  The main issue here may rather be whether you should impose – or refuse to dismantle – barriers to the pursuit of other people’s goals, when these goals are not in any sense evil, even if – as in this case –

  you do think that they are not conducive to promoting their own 192

  r a t i o n a l i t y a n d o t h e r p e o p l e well-being. Perhaps you are reluctant, as a general rule, to be unhelpful to your neighbour (no matter what you think of their goals). Or maybe you judge that, while having a window seat gives you control over the proximate shutter, this incidental advantage should not be used by you without taking into account what others want to do and how that would be affected by your choice regarding the window shade (even though you yourself were rather enjoying the sun which would now be shut out and even though you think very little of the goal that the other guy wants to pursue).

  These arguments may be explicitly invoked or implicitly considered in your decision, but is it right to think that your socially influenced behaviour shows that your objective is to help others to pursue their own goals, no matter what you think of their goals? Thanks to your acceptance of social norms of behaviour, you have certainly ended up helping the guy next to you to pursue his own goal. But it is surely too much to say either that your objective is to maximally help others to pursue their respective goals, or that their goals have somehow become yours as well (‘Thank God, no,’ you heave a sigh of relief).

  Rather, you are just following a norm of good behaviour you happen to approve of (let others do what they really want), which is a self-imposed behavioural restraint you accept in your choice of what to do.

  There is nothing very peculiar, or silly, or irrational about your decision to ‘let others be’. We live in a world in which there are a lot of other people, and we can give them room for their own way of living even without adopting their way as something that we must see as a good thing to promote. Commitment may take the form not only of wanting to pursue goals that are not entirely parasitic on self-interest; it can also take the form of following rules of passable, even perhaps generous, behaviour, that restrain our inclination to be guided exclusively by the promotion of our own goals, irrespective of its impact on others. Being considerate of the desires and pursuits of others need not be seen as a violation of rationality.

  193

  9

  Plurality of Impartial Reasons

  It was argued in the last chapter that there is nothing extraordinary or irrational in making choices and decisions that cross the narrow boundaries of exclusive pursuit of personal self-interest. People’s goals can go well beyond the single-minded promotion of self-interest only, and their choices may even go beyond the single-minded pursuit of their personal goals, perhaps moved by some concern for decency of behaviour, allowing others to pursue their goals as well. The insistence of so-called rational choice theory on defining rationality simply as intelligent promotion of personal self-interest sells human reasoning extremely short.

  The connection between rationality of choice and the sustainability of the reasons behind the choice was discussed in the last chapter. In this understanding, rationality is primarily a matter of basing –

  explicitly or by implication – our choices on reasoning that we can reflectively sustain, and it demands that our choices, as well as our actions and objectives, values and priorities, can survive our own seriously undertaken critical scrutiny. It was also discussed why there is no particular ground for imagining that every motivation other than the pursuit of self-interest must somehow be guillotined by such a critical scrutiny.

  However, while rationality of choice can easily allow non-self-interested motivations, rationality does not on its own demand this.

  While there is nothing odd or irrational about someone being moved by concern for others, it would be harder to argue that there is some necessity or obligation to have such concern on grounds of rationality alone. We can have sustainable reasons for action that reflect our
inclinations and our own individual lines of self-scrutiny. Rationality 194

  p l u r a l i t y o f i m p a r t i a l r e a s o n s as a characteristic of choice behaviour rules out neither the dedicated altruist, nor the reasoned seeker of personal gain.

  If Mary decides, in a cogent and intelligent way, to pursue her idea of the social good, even at great sacrifice to herself, it would be hard to see her, for that reason, as being ‘irrational’. And yet the charge of irrationality may be difficult to sustain against Paul even if he were a no-nonsense maximizer of self-interest, provided his values, priorities and choices would survive his own serious scrutiny.* Commitment to the concerns of others might simply be less important for Paul than for Mary.† We may well think that Paul is a less ‘reasonable’ person than Mary, but as John Rawls has discussed, this is a different issue from irrationality as such.1 Rationality is in fact a rather permissive discipline, which demands the test of reasoning, but allows reasoned self-scrutiny to take quite different forms, without necessarily imposing any great uniformity of criteria. If rationality were a church, it would be a rather broad church. Indeed, the demands of reasonableness, as characterized by Rawls, tends to be more exacting than the requirements of mere rationality.‡

  * Paul would have to take note, among other considerations, of the fact that a no-nonsense pursuit of self-love may adversely affect his relations with others, which could be a loss even for self-interested reasons.

  † The term ‘rational’ allows a further distinction that has been illuminatingly pursued by Thomas Scanlon: (1) what a person has most reason to do, and (2) what a person must do to avoid being irrational (see Scanlon, What We Owe to Each Other (Cambridge, MA: Harvard University Press, 1998), pp. 25–30). Mary and Paul could be seen as being rational in both senses. There remains, however, the further issue of sustainability of the reasons invoked, which is central to the idea of rationality as seen in this work (see Chapter 8), and is more fully discussed in my book Rationality and Freedom (2002).

  ‡ However, in illustrating the ‘familiar distinction between reasonableness and rationality’, Thomas Scanlon gives an example that seems to go the other way ( What We Owe to Each Other, pp. 192–3). A person, Scanlon points out, may find a possible objection to a powerful person’s behaviour to be entirely ‘reasonable’, and yet may decide that it would be ‘irrational’ to express that indictment because of the likely anger of that person: so a reasonable statement need not, rationally, be expressed in some circumstances. There are, it seems to me, two distinct issues placed together here. First, the respective demands of rationality and reasonableness are different and need not coincide (and I would tend to argue, in general, that reasonableness would typically demand something more than just rationality). Second, the rationality of an understanding or a decision has to be distinguished from the rationality of publicly 195

  t h e i d e a o f j u s t i c e The demands of scrutiny would have to be sharpened and tightened when we move from the idea of rationality to that of reasonableness, if we broadly follow John Rawls in interpreting that distinction. As was discussed in Chapter 5 (‘Impartiality and Objectivity’), the idea of objectivity in practical reason and behaviour can be systematically linked to demands of impartiality. Drawing on this, we can take the relevant standard of objectivity of ethical principles to be linked to their defensibility in an open and free framework of public reasoning.*

  Other people’s perspectives and assessments as well as interests would have a role here in a way that rationality alone need not demand.†

  We must, however, investigate more closely the idea of defensibility in reasoning with others. What does defensibility demand and why?

  w h a t o t h e r s c a n n o t

  r e a s o n a b l y

  r e j e c t

  In the play King John by William Shakespeare, Philip the Bastard remarks that our general evaluation of the world is often influenced by our own special interests:

  Well, whiles I am a beggar, I will rail

  And say there is no sin but to be rich;

  expressing that understanding or decision. The distinction between ‘a good statement’

  and ‘a good statement to make’ can often be quite momentous in the dual discipline of thought and communication. I have tried to analyse the distinction in my essay

  ‘Description as Choice’, in Choice, Welfare and Measurement (Oxford: Blackwell, 1982, and Cambridge, MA: Harvard University Press, 1997).

  * Rawls’s own wording seems to concentrate on open dialogue, not with all, but only with ‘reasonable people’, and the distinction between this approach with some clearly stated normative elements (reflected in the diagnosis of ‘reasonable persons’ and what they would find to be ‘reasonable’) and the more fully procedural view of Habermas was discussed in Chapter 5. I argued there that the distinction may not be as sharp as it might at first appear.

  † It is possible to define in different ways the reach of ‘an open and free framework of public reasoning’, and the differences in formulation may be quite significant in seeing the precise – and sometimes subtle – distinctions between Rawls’s use of this approach and the uses made by others, including Kant and Habermas. I shall not, however, go further into these issues of differentiation here, since they are not central to the approach of this book.

  196

  p l u r a l i t y o f i m p a r t i a l r e a s o n s And being rich, my virtue then shall be

  To say there is no vice but beggary.*

  It is hard to deny that our positions and predicaments can influence our general attitudes and political beliefs about social differences and asymmetries. If we take self-scrutiny very seriously, it is possible that we may be hard-minded enough to seek more consistency in our general evaluative judgements (so that, for example, our judgements on the rich do not radically vary depending on whether we are ourselves rich or poor). But there is nothing to guarantee that this type of exacting scrutiny will always occur, since we are capable of much self-indulgence in our views and opinions of things in which we are directly involved, and this may restrain the reach of our self-scrutiny.

  In the social context, involving fairness to other people, there would be some necessity to go beyond the requirements of rationality in terms of permissive self-scrutiny, and to consider the demands of

  ‘reasonable conduct’ towards others. In that more demanding context, we must pay serious attention to the perspectives and concerns of others, as they would have a role in the scrutiny to which our decisions and choices can be sensibly subjected. In this sense, our understanding of right and wrong in society has to go beyond what Adam Smith called the dictates of ‘self-love’.

  Indeed, as Thomas Scanlon has persuasively argued, ‘thinking about right and wrong is, at the most basic level, thinking about what could be justified to others on grounds that they, if appropriately motivated, could not reasonably reject’.2 While survival under one’s own engaged scrutiny is central to the idea of rationality, taking serious note of critical scrutiny from the perspectives of others must have a significant role in taking us beyond rationality into reasonable behaviour in relation to other people. There is clearly room here for the demands of political and social ethics.

  Is Scanlon’s criterion different from the demands of Rawlsian fairness, through the device of the ‘original position’, which was examined earlier? Certainly, there is a strong connection between the two.

  Indeed, the Rawlsian ‘veil of ignorance’ in the ‘original position’

  (whereby no one knows who he or she is going to be in the real world)

  * William Shakespeare, King John, II. 1. 593–6.

  197

  t h e i d e a o f j u s t i c e was devised by Rawls to make people see beyond their personal vested interests and goals. And yet there are substantial differences between Rawls’s firmly ‘contractarian’ approach, focusing ultimately on mutual benefits through agreement, and Scanlon’s broader analysis of reasoning
(even though Scanlon rather muddies the water by insisting on calling his own approach ‘contractualist’).

  In Rawlsian analysis, when the representatives of the people congre-gate and determine what principles must be seen as ‘just’ for guiding the basic institutional structure of the society, the interests of the different persons all count (in an anonymous way, since no one knows, thanks to ‘the veil of ignorance’, who exactly anyone is actually going to be). As Rawls characterized the original position in his Theory of Justice, the parties or their representatives do not unleash any specific moral views or cultural values of their own in the deliberations of the original position; their task is merely to best advance their own interests and the interests of those whom they represent. Even though all the parties pursue their respective interests, the contract on which a unanimity is meant to emerge can be seen, in the Rawlsian perspective, as the best for the interests of all, taken together, under the ‘veil of ignorance’ (since the veil prevents anyone from knowing who exactly he or she is going to be).* Impartial aggregation through the use of ‘the veil of ignorance’, it must be emphasized, need not be an unproblematic search, since it is not at all clear what would be chosen in that kind of devised uncertainty. The absence of a unique solution, unanimously chosen by all the parties, corresponds to the absence of a unique social aggregation of the conflicting interests of different people. For example, the Rawlsian distributional formula of prioritiz-

  * See John Harsanyi, ‘Cardinal Welfare, Individualistic Ethics, and Interpersonal Comparisons of Utility’, Journal of Political Economy, 63 (1955). There are other claimants to a solution, for example maximizing an equity-adjusted sum-total of utilities, as proposed by James Mirrlees (‘An Exploration of the Theory of Optimal Income Taxation’, Review of Economic Studies, 38, 1971). See also John Broome, Weighing Lives (Oxford: Clarendon Press, 2004). I shall not go further into this question here, but it is addressed in my Collective Choice and Social Welfare (San Francisco, CA: Holden-Day, 1970, and Amsterdam: North-Holland, 1979); On Economic Inequality (Oxford: Clarendon Press, 1973, expanded edition, jointly with James E. Foster, 1997); and ‘Social Choice Theory’, in Kenneth Arrow and Michael Intriligator (eds), Handbook of Mathematical Economics (Amsterdam: North-Holland, 1986).