Freer than others

Bernard Williams

  • Inequality Examined by Amartya Sen
    Oxford, 207 pp, £19.95, September 1992, ISBN 0 19 828334 2

Every modern state and every modern political philosophy believes in equality of something. As Amartya Sen points out in this book, even libertarians, who think that there should be no politically imposed limits on what people may retain of what they gain without force or fraud, believe in the equal right to exert oneself in the market and not to be taxed. Those who think that more effortful or productive or responsible work deserves higher rewards think that this principle should be applied equally to all citizens. The important issue, then, as Sen has helpfully insisted over many years, is not whether we are in favour of equality, but rather: equality of what?

Even if all modern outlooks accept some kind of equality, many of them do not list it among their political ideals. There is a question, then, about what it is that makes some conceptions of equality rather than others into the focus of political programmes that aim to increase equality; programmes, for instance, which stand in the tradition of linking equality with liberty and fraternity. There is a question, too, of why it is that all modern states do profess some conception of equality. It is not enough to reply that they have to do so, because equality is (even now) a leading catchword of modern politics: this merely raises the question again, in a more cynical tone of voice. The answer seems to be that in the modern world, which has largely rejected mythical or merely traditional sources of authority, only some conception of equal consideration for each citizen can form the basis of uncoerced and informed allegiance to a government.

Not so many governments at the present time can be said without any hesitation or qualification to live up to the promise offered by that formula, and that fact itself leads back to the problem of defining equality as a political ideal, rather than as a mere assumption. The point of articulating and pursuing a political ideal of equality is not to indulge resentment or a managerial passion for uniformity, even though both have no doubt played some part in the history of egalitarianism. The basic aim, now more important than ever, is to find a practical conception of equality that can give people a genuine sense that they receive equal consideration from society and so have a stake in it. Only this has any hope of giving people a reason why they should obey and co-operate. Without such a reason, there are only coercion, mystification, habit, and the hope that people will be content with making the best of where they find themselves. The sort of equality that contains no aspiration and can be comfortably announced as being already here, such as an equal legal right to become a millionaire, is quite obviously not enough. It is extraordinary that anyone can have thought, as some followers of Lady Thatcher have thought and perhaps in a few cases still do think, that mere equality in the face of the market could realise enough of an idea of equal citizenship to make anything work, including the market itself.

Sen’s theory of equality does yield an aspiration that can speak to the problems of the modern state, although he does not himself discuss the most basic and general reasons for which a modern society might be interested in ideals of equality. He concentrates rather on what such an ideal should be. He properly reminds us that if we are going to develop ideas of equality, we had better have some notion of their purpose, and that for different purposes we may want to use different ideas of equality. For some purposes of economic understanding, inequality of income may be the relevant measure, but for broader political and social aims we need richer ideas. Sen is extremely aware of political issues, such as poverty, deprivation and injustice to women, but this book is not a work of political theory, and it does not start from the political questions that themselves create the demand for an understanding of equality. Sen is both an economist and a moral philosopher, but he approaches the problems of equality by a route that runs from economic theory.

The issue, as I said at the start, is, equality of what? Or as Sen also puts it, in the language of mathematical economists, in what space do we want equality to obtain? It is a fundamental point that equality in one space can, in virtue of the very same facts, mean inequality in another. To take one of Sen’s favourite examples, people have different needs with respect to food, because of their body weight, their age, their state of health, and so on. To give them all the same food will not generate the same degree of nutrition in each; equality in the space of food provision means inequality in the space of nutrition. Similarly, equality of money or other such resources does not mean equality in terms of what people can achieve: for many different reasons, people are not the same in their capacity to convert resources into worthwhile or satisfying activity.

Welfare economists have made this point in terms of what was traditionally their favourite measure, ‘utility’ – which means roughly the degree to which someone is satisfied with a given outcome or gets what he wants. An addition of the same resource does not yield the same increase in utility, either between different people, or indeed for the same person in different contexts. One application of this point is the familiar ‘diminishing marginal utility of money’, by which an additional hundred pounds means more to someone who has a little than to someone who has a lot. Sen, however, has been a powerful and influential critic of those who overestimate the usefulness and, beyond a certain point, the coherence of the concept of utility, and for many reasons (not for the most part stated here, but referred to as appearing elsewhere) he rejects utility as the measure of what he calls ‘basal equality’.

He also rejects as the measure of equality the ‘primary goods’ that John Rawls has specified as the objects of distribution in his political theory. Rawls described these as multi-purpose goods that any reasonable human being in most social circumstances would want: they include money and ‘the means to self-respect’. Sen’s own proposal is that the space in which the most basic equality is to be established is that of freedom itself. Rawls’s primary goods, he claims, represent means rather than ends: the only point of money as a primary good is its power to increase one’s freedom to choose. The criticism should, perhaps, rather be that the description of the primary goods that Rawls gave in A Theory of Justice was misleading. One of Rawls’s primary goods was freedom itself, and Sen’s criticism, that the primary goods are only means to freedom, can hardly apply to that; but then Rawls’s description of a primary good does not apply to it very neatly, either.

If we can make freedom equal between different people, then we shall make equal the range of choices they have, and, with that, the range of ‘capabilities’ they possess for different kinds of human ‘functioning’. If we bring it about that disabled people get more resources, we increase the range of things that they can choose to do, and this is the sense in which we increase their freedom. In explaining these ideas, Sen makes a number of careful and important distinctions. In many cases, having a range of alternatives from which to choose is an instrumental good, in the sense that it enables one to find the most satisfactory option. This is so in the standard type of economic model centred on utility, in which the desired outcome is merely identified with the item one picks, and the range of choice one had serves only as the basket out of which one picks it. Sen points out, however, that in many connections, choosing is itself important. It often makes a difference whether one chose a certain outcome, or the outcome was simply delivered to one, even though the outcome is just as good in itself. Sen thus wants to give choice and action a real place in the theory of equality, and not leave them as merely the routes that lead to desirable outcomes, as they standardly have been left, not only in theory, but, too often, in the practice of the welfare state, which has tended to regard the disabled as beneficiaries rather than as people who want the chance to make their own choices.

If equality is basically to be understood as equality of freedom, then the supposed clash between equality and freedom, so famous from confrontations between Left and Right, must in some way be ill-defined. Sen indeed says that to put the problem in these terms reflects a category mistake. ‘They are not alternatives. Liberty is among the possible fields of application of equality, and equality is among the possible patterns of distribution of liberty’ (his emphasis). Sen does not go as far as Rousseau and some other philosophers who say that there cannot be a conflict between liberty and equality at all, on the ground that nothing which conflicts with equality can be genuine liberty, and nothing that conflicts with liberty can be genuine equality. To the extent that equality at the most basic level is being extended, it must indeed be true in Sen’s view that someone’s freedom (of some kind) is being increased, since freedom is what is being more equally distributed. However, this does not mean that we can think solely in terms of an increase in freedom, and forget about conflicts between freedom and equality.

Suppose that the freedom of some poor and disabled people to get around is increased by special provision, and this is paid for by an increase in redistributive taxation. (Given present attitudes in Britain, the example is distinctly utopian, but that does not affect the argument.) As a result of the tax increase some higher tax-payers’ range of choices, and hence, in Sen’s sense, their freedom, is diminished: they can no longer afford both the Bentley and Gstaad this year. Now what is being bought with this bit of their freedom is an increase in someone else’s freedom. However, it is not necessary to Sen’s argument that the increase in the disabled person’s freedom would be greater than the rich person’s loss of freedom. One person’s freedom is being set against another’s, because freedom is the currency of equality, but it is not necessarily the case, merely in terms of freedom, that one person’s gain will more than cancel out the other’s loss: all that is necessary is that at the end we should be nearer to equality. In such a case there can be a real conflict between freedom and equality – even though the equality is itself equality of freedom.

Sen recognises that in such transactions there is a danger, as with equality over other spaces, of ‘levelling down’, but argues that there are other values to be taken into consideration. He accepts, for instance, that it would not be sensible to use large resources in order to increase marginally the capabilities of disadvantaged people, if the cost of this were to reduce severely the productivity of advantaged people; that would be an unacceptable loss in efficiency. He does not say much about values other than economic efficiency in their relation to equality of freedom. Thus he does not say much about the situation just discussed, in which the loss is of freedom. Again, he does not say very much about fairness, for instance in relation to questions of giving people who are more skilled better rewarded positions. He discusses this in terms of incentives and efficiency, but many people think that it is not merely inefficient, but actually unfair, not to give some such rewards (though Rawls dissents on the ground that no one deserves their talents, which is true but doubtfully relevant). It would be interesting to see how Sen would bring together freedom, the basic currency of equality, with desert and similar differential ideas of fairness; outside the area of punishment, does anyone deserve more freedom than another?

Determinations of equality and inequality in the space of freedom demand, as the examples show, some ways of ‘measuring’ increases or decreases in freedom. No sensible person should demand highly determinate or quantitative measures, and Sen has for many years been a leader in trying to persuade his fellow economists that some reasonable comparisons in the actual world are worth a great deal more than highly sophisticated operations on quantities that exist only in mathematical models. In the matter of counting or weighing freedoms or capabilities, however, we do need some guidance, and it can be fairly complained that Sen does not give us very much. It is an obvious point that one can count capabilities, choices, and so forth in any way one likes. To use an example I have put to Sen before, someone who introduces a new washing powder introduces also indefinitely many new choices (such as choosing between buying some arbitrary other good and buying this washing powder) and at the same time takes others away (such as the chance of making an informed choice of washing powder without worrying about this one). Counting, clearly, will get us nowhere. To this line of questioning Sen gives a robust reply to the general effect that any criterion can give rise to some such difficulties, and you just have to use good judgment in the face of actual circumstances.

The considerations one uses in actual circumstances rely, unsurprisingly, not on numbers, but on weight: some capabilities (or freedoms or possibilities of choice) are more important than others. In very many cases, however, the importance of the freedom is directly related to the importance of the functioning in question. The capacity to walk is important in the first instance because walking is important. If I can walk, there are many more things that I can do, and choices come with this, of where and when to walk. But if there is a real question that centres on the choices – a question whether I can walk where and when I choose – this is naturally understood as a further matter, one that comes up only if I can walk. To put both questions under the language of freedom runs the risk of mixing together two different kinds of political concern: it is one thing to able to walk (not to be paralysed, for instance), and another to be free (e.g. from police interference) to walk where and when I want.

There are other cases in which Sen’s emphasis on freedom seems to pick on a consequence, rather than the centre, of some undesirable state. Stressing, as Franklin Roosevelt did, the importance of ‘freedom from’ such things as malaria, he lays the weight on counterfactual choice, the kinds of life people could choose to lead if they did not have malaria. It is importantly true that malaria is not just unpleasant but disabling. On the other hand, does that fact in itself pick out what is so obviously bad about it? ‘If only I had not been ... I could have chosen a richer life than the one I have’ can be truly filled in many different ways, and not all the fillings have anything like the same political or social significance.

In many of Sen’s examples, there is no doubt at all that the state of disadvantaged people would be improved if resources could be devoted to relieving their disadvantage: malnutrition, disease, ignorance, insecurity. Equality would be advanced if their state were better. But – as Sen himself admirably brings out in some of his technical discussions of comparabilities – you can arrive at this conclusion on almost any account of equality. With regard to these disadvantages, any reasonable story about the way human beings should live will deliver much the same result, and the special emphasis on freedom seems unnecessary, and in some cases, as I have already suggested, secondary. In other cases, on the other hand, the emphasis on freedom makes a considerable difference, but its results are also contestable. This is true with Sen’s admirable discussion of gender discrimination in various parts of the world. Many of the statistics about women’s disadvantage refer, once more, to such uncontroversial evils as malnutrition and early death, but others, relating to women’s levels of education and chances of employment, raise ideologically disputed questions of what capabilities should be developed by women.

It is no criticism of Sen that he should take a stand in favour of women’s rights to self-development. What is unclear, rather, is the extent to which he thinks that these dimensions of freedom and capability can themselves be theoretically derived. Does his theory say only that freedom and capability are the proper basis of claims to equality? Or does the theory deliver also the conclusion that the demand for equality of educational opportunity for women follows from any adequate account of human capabilities and potentialities? I should not be surprised or disappointed if he wanted to say the second. But then his theory will need to be supplemented by materials which at the moment it does not offer or even promise, in particular a theory of false consciousness which will explain why many women have failed to understand their own capabilities.

As it is, Sen’s theory does stand rather oddly to the politics with which he is so evidently concerned. Much of the disadvantage that he mentions, which a move to greater equality would hope to reduce, is so uncontroversially awful that the refined arguments about the primacy of freedom seem unnecessary: whatever space you are working in – whether it is that of utility, resources, primary goods or freedom – you will get the same answer. In other cases, the results of the approach are much more controversial, at least in terms of local cultural problems, and then one must ask how far Sen’s theory licenses us politically to treat such cases as being just like the uncontroversial cases (which is what, in terms of his theory, they are: all the cases equally involve the restriction of basic freedoms). But that needs a further political dimension of the theory: a dimension in which we can understand such things as false consciousness and the ideological misrepresentation of basic human capacities, and which will help us to discuss (among other things) the relation of Western agencies to people who do not necessarily share Western views.

In his work on many topics, notably famine, and also on several subjects discussed in this book, such as the definition of poverty, Sen’s acute analysis and his remarkable powers of making subtle and relevant distinctions combine with his astonishing range of information to make instruments suitable for immediate political application. The theory of equality as the equality of freedom does not seem quite to do this. Its distinctions seem to yield either more than we need for political purposes, or less. Perhaps this is only to say that we need more weapons than this compelling and elegantly argued book can offer. Granted the depth and the growth of inequality in this country, to look no farther, it is hardly surprising.