The Idea of Justice

by Amartya Sen

  Fourth, equality is itself not the only value with which a theory of justice need be concerned, and it is not even the only subject for which the idea of capability is useful. If we make the simple distinction between aggregative and distributive considerations in social justice, the capability perspective with its pointer to an important way of assessing advantages and disadvantages has implications for both aggregative and distributive concerns. For example, an institution or a policy may well be defended not on the grounds that it enhances capability equality, but for the reason that it expands the capabilities of all (even if there is no distributional gain). Equality of capability, or more realistically reduction of capability inequality, certainly has claims on our attention, but so has the general advancement of the capabilities of all.

  Through denying the case for single-minded concentration on capability equality, or for that matter on capability-based considerations in general, we do not denigrate the critically significant role of capabilities in the idea of justice (discussed earlier, particularly Chapters 11–

  13). The reasoned pursuit of a very important element in social justice, which does not crowd out everything else, can still have a crucial role in the enterprise of enhancing justice.

  298

  e q u a l i t y a n d l i b e r t y c a p a b i l i t y a n d

  p e r s o n a l

  l i b e r t i e s

  As discussed in Chapter 2, in departing from John Rawls’s focus on primary goods in the Difference Principle in addressing distributional issues, and in bringing in the far-reaching role of capabilities in that exercise, there is no hidden intention of disputing Rawls’s reasoning on other issues. Those issues include the priority of liberty, which forms the subject matter of the first principle in Rawls’s theory of justice.

  Indeed, as I have already argued (in Chapter 2, ‘Rawls and Beyond’), there are good grounds for giving personal liberty some kind of a real priority (though not necessarily in the extremist lexicographic form chosen by Rawls). 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. Liberty is indeed useful, like income and other primary goods, but that is not all that is involved in the importance of liberty, since it touches our lives at a very basic level and it demands that others should respect these deeply personal concerns that everyone tends to have.

  This distinction is crucial to bear in mind when we compare the competing claims of primary goods and capabilities for one limited purpose in the assessment of justice, to wit, how to evaluate general distributive concerns, based on comparisons of overall individual advantages. That is, of course, the subject matter of Rawls’s difference principle, but it is just one part of a larger theory of Rawlsian justice.

  When it is claimed, as I have, that capabilities can do the job of judging the overall advantages of different people better than primary goods, then that is precisely what is being affirmed – and not anything more. There is no claim here that the capability perspective can take over the work that other parts of Rawlsian theory demand, particularly the special status of liberty and the demands of procedural fairness. Capabilities cannot do that work any more than primary goods can. The contest between primary goods and capabilities is in a limited arena, in a specified domain, concerned with the assessment of overall advantages that the individuals respectively have.

  299

  t h e i d e a o f j u s t i c e Since I am broadly in agreement with the Rawlsian reasoning underlying the first principle, that is, the importance of the priority of personal liberty shared equally by all, it is perhaps useful to consider whether this priority must be as absolute as Rawls makes it out to be.

  Why must any violation of liberty, significant as it is, invariably be judged to be more crucial for a person – or for a society – than suffering from intense hunger, starvation, epidemics and other calamities? As was discussed in Chapter 2 (‘Rawls and Beyond’), we have to distinguish between giving some priority to liberty (not treating it merely as one of the components in the large bag of ‘primary goods’, since liberty is so central to our personal lives), and the ‘extremist’ demand of placing a lexicographic priority on liberty, treating the slightest gain of liberty – no matter how small – as enough reason to make huge sacrifices in other amenities of a good life – no matter how large.

  Rawls argues persuasively for the former, and yet chooses, in the formulation of the difference principle, the latter. But as was discussed in Chapter 2, the mathematics of differential weighting allows many intermediate possibilities between no extra weight on liberty and complete priority of liberty over all else. We can be ‘Rawlsian’ in the former sense, as far as the ‘priority of liberty’ is concerned, without signing up for the latter.

  The exact extent of priority that may be given, in a particular case, to personal liberty would certainly be a good subject for public reasoning, but Rawls’s main success here seems to me to lie in showing why personal liberty has to be given a pre-eminent place in public reasoning in general. His work has helped to generate the understanding that justice in the world in which we live demands a very special concern with liberties that all can share.* The important point to note here is that liberty has a place in a just social arrangement that goes well beyond recognizing liberty to be a part of personal advantage, in the way income or wealth is. Even as the role of substantive freedoms in the form of capabilities is emphasized in the present work (departing

  * Sharing is very important here, rather than liberty being demanded for some but not for others. Mary Wollstonecraft’s criticism of Edmund Burke’s support for American independence, without raising the question of the liberty of slaves, was discussed earlier (in Chapter 5, ‘Impartiality and Objectivity’).

  300

  e q u a l i t y a n d l i b e r t y from Rawls), there is no necessity there to deny the special role of liberty.*

  t h e p l u r a l f e a t u r e s o f f r e e d o m Given the importance of freedom in different forms in theories of justice, I must now go into a closer examination of the contents of liberty and freedom, which has been a veritable battleground in the literature. The terms ‘freedom’ and ‘liberty’ are used in many different ways, and something more must be said on their respective domains.

  One distinction in particular, between the opportunity aspect and the process aspect, was explored in Chapter 11 (‘Lives, Freedoms and Capabilities’). The plurality of aspects of freedom can be approached and identified in other ways as well, besides the already discussed distinction between the opportunity and process aspects. Freedom to achieve what one reasonably wants to achieve relates to a variety of factors, and they can have varying relevance to different concepts of freedom.

  The question whether a person can bring about the objects of her reasoned choice is crucial to the idea of freedom that is being pursued here, of which the notion of capability is a part.† But the effectiveness of preference can occur in different ways. First, a person can bring about the chosen result through her own actions, yielding that particular outcome – this is the case of direct control. But direct control is not necessary for effectiveness. Second, there is the broader

  * The priority of liberty plays an important part in the social choice result presented in my ‘The Impossibility of a Paretian Liberal’, Journal of Political Economy, 78

  (1970). John Rawls comments illuminatingly on this connection in his essay, ‘Social Unity and Primary Goods’, in Amartya Sen and Bernard Williams (eds), Utilitarianism and Beyond (Cambridge: Cambridge University Press, 1982). I shall return to this issue later on in the chapter.

  † In seeing freedom in terms of the power to bring about the outcome one wants with reasoned assessment, there is, of course, the underlying question whether the person has had an adequate opportunity to reason about what she really wants. Indeed, the opportunity of reasoned assessment cannot but be an important
part of any substantive understanding of freedom. As was discussed in Chapter 8, ‘Rationality and Other People’, this is a central question in assessing the rationality of preference and choice.

  301

  t h e i d e a o f j u s t i c e consideration of whether a person’s preferences can be effective –

  whether through direct control or through the help of others. Illustrations of the ‘indirect power’ to bring about the preferred results vary from such simple cases of acting through an attorney or loyal friends or relations, to more complex ones in which a doctor takes decisions for a person to bring about a result that the patient would actually choose, given enough knowledge and understanding: the issue of effective power. The importance of effective power through indirect control calls for some discussion here, particularly since it is so common to see freedom as being nothing other than control, and being given the choice to do certain things oneself.

  Many of the freedoms that we exercise in society work through some process other than direct control.8 For example, a wounded and unconscious victim of an accident may not take the decisions about what is to be done to him, but in so far as the doctor chooses a course which the doctor knows the patient would have preferred had he been conscious, there is no violation of the patient’s freedom – indeed, there is an affirmation of that freedom in the sense of ‘effective power’, if the doctor’s choice is guided by what the patient would have wanted.9 This is a distinct issue from the welfare of the patient as she

  – the doctor – reads it, which could also guide the doctor. Even though respecting the freedom of the patient may often have the same requirements as the advancement of the well-being of the patient, the two need not coincide. For example, a doctor may respect the unconscious patient’s well-known rejection of medicines derived from cruel experiments on animals, even though in the doctor’s view the well-being of the patient would have been enhanced by the use of precisely that medicine. The guidance of well-being can differ – possibly quite sharply – from the demands of the effective freedom of the patient.

  The idea of effective freedom can be extended to more complex cases of societal arrangements, for example where the civic authorities looking after regional epidemiology arrange to eliminate local epidemics (what the people, it is known, want). The idea of effectiveness would apply to the group and its members, and effective freedom here takes a social – or a collaborative – form, but it is still a case of effectiveness without any individual having any specific control over 302

  e q u a l i t y a n d l i b e r t y the societal decision. The distinction is between the local authorities undertaking some policy on the grounds that this is what the people want and would, given the option, choose, and the authorities undertaking that policy on the grounds that this would enhance, in the view of the administrators, the welfare of the people in the locality. The second is, of course, a worthy enough reason, but it is not quite the same reason as the first (even though the two arguments have causal connections since consideration of well-being may plausibly influence the choice – or would-be choice – of the people involved).

  A different kind of distinction would be that between being able to get some result precisely because of having that preference, perhaps in conformity with the preferences of the others involved (for example, a person wanting the elimination of epidemics in unison with others in that region – a preference that, ultimately, may guide public policy), and a person being able to get what one wants due to good luck. It may just turn out, for one reason or another, that precisely what this person wants actually does occur. There is fulfilment here, but not necessarily any effectiveness of one’s preferences since there may be no influence of one’s priorities on what occurs (it may not at all be the person’s wanting the result, individually or jointly, that brings the result about). There is not only no control here (direct or indirect), but not even any exercise of power, through whatever means, to produce a result in line with one’s preferences. One succeeds with one set of preferences but does not necessarily do so with another.

  For example, a person’s religious practice may happen to be in conformity with what the state wants to enforce, and the person may thus see his religious preferences fulfilled, without those preferences having any particular role in the state decisions. It may look as if there is nothing substantial that can be called ‘freedom’ in the person’s piece of good luck, and in terms of bringing about a particular result –

  whether through direct or indirect control – this scepticism about the presence of freedom is well justified, since the person here is just in a favourable situation, rather than being effective in getting whatever he wants.* And yet the person’s freedom to live as he would like can

  * Philip Pettit takes this view and sees freedom only in ‘content-independent’ terms (so that one’s effectiveness must be independent of what exactly the person wants).

  303

  t h e i d e a o f j u s t i c e contrast sharply with the predicament of someone else who subscribes to some heterodox beliefs and may face obstacles to his practice (in another age, he could have been unlucky enough to face the Inquisition). There is a freedom of some importance in being able to follow one’s preferred lifestyle, despite there being no real freedom of choice here (that is, irrespective of the content of one’s preference). When, for example, Akbar pronounced and legalized his freedom-favouring decision that no one ‘should be interfered with on account of religion, and anyone is to be allowed to go over to a religion that pleases him’, he guaranteed the effective freedom of a great many people – indeed, a majority of his subjects who earlier faced discrimination on grounds of not being Muslim – and yet those subjects would have had no power to stop him had Akbar chosen differently.

  This distinction relates to one to be discussed presently, involving the contrast between capability in general and capability without dependence, emphasized in a specific approach to freedom (to be discussed presently), called the ‘republican’ view, developed particularly by Philip Pettit. But I hope the preceding discussion has done something to establish the need to see freedom in plural terms, rather than seeing it as having only one feature.

  c a p a b i l i t y , d e p e n d e n c e

  a n d

  i n t e r f e r e n c e

  Some people use the terms liberty and freedom quite interchangeably, and treat the two as if they were much the same. In Rawls’s arguments for priority of liberty there is, however, a special concern with freedom in personal lives, and especially freedom from intrusive interference by others, including the state. Going beyond what people can – taking everything into account – actually do, Rawls also investigates the importance of people being at liberty to lead their own lives as they would like, and in particular the liberty not to be messed around by See his Republicanism: A Theory of Freedom and Government (Oxford: Clarendon Press, 1997), and ‘Capability and Freedom: A Defence of Sen’, Economics and Philosophy, 17 (2001).

  304

  e q u a l i t y a n d l i b e r t y the interference of others. And that, of course, is the classic territory of John Stuart Mill’s pioneering work, On Liberty.10

  In some theories of freedom, for example what is called ‘republican’

  or ‘neo-Roman’ theory, liberty is defined not just in terms of what a person is able to do in a certain sphere, but also includes the demand that others could not have eliminated that ability of this person even if they wanted to do so. In this view, a person’s liberty may be compromised even in the absence of any interference, simply by the existence of the arbitrary power of another which could hinder the freedom of the person to act as they like, even if that intervening power is not actually exercised.11

  Philip Pettit has argued against the view of freedom as capability on ‘republican’ grounds, since a person may have the capability to do many things that are dependent on the ‘favour of others’, arguing that to the extent the person’s actual choices (or achievements) are dependen
t in this way, he is not really free. As Pettit explains: ‘Imagine that you have a disposition to choose between A and B that is content-independently decisive but that your enjoyment of such decisive preference depends on the goodwill of those around you . . . You may be said to have decisive preferences but their decisiveness is favour-dependent.’* Certainly being free to do something independently of others (so that it does not matter what they want) gives one’s substantive freedom a robustness that is absent when the freedom to do that thing is conditional either on the help – or on the tolerance – of others, or dependent on a coincidence (‘it so happens’) between what the person wants to do and what the other people who could have stopped it happen to want. To take an extreme case, it can certainly be argued

  * Philip Pettit, ‘Capability and Freedom: A Defence of Sen’, Economics and Philosophy, 17 (2001), p. 6. I am commenting here not on the ‘defence’ part of Pettit’s arguments, but on his critique of my focus on capability, suggesting that it should be extended in the direction of the ‘republican’ view, so that capabilities that are favour-dependent do not count as real freedoms. Pettit sees this as a natural extension of the idea of capability and its defence (as presented by me): ‘Under my reading, Sen’s theory of freedom coincides with the republican approach in this emphasis on the connection between freedom and non-dependency’ (p. 18). I see the relevance of that connection, but have to argue that both concepts – the republican and the capability-based views of freedom – have value since they reflect distinct aspects of the inescapably plural idea of freedom.