The Idea of Justice
Page 15
Kenneth Arrow himself joined others in pursuing ways and means of broadening the informational basis of social choice.10 In fact, Condorcet too had already pointed in that direction in the 1780s in very general terms.11 There is a close motivational link here with Condorcet’s passionate advocacy of public education and particularly women’s education: Condorcet was one of the first to emphasize the special importance of the schooling of girls. There is also a close connection with Condorcet’s deep interest in enriching societal statistics, and with his commitment to the necessity of continuing public discussion, since they all help to advance the use of more information in the procedures of public choice and in the exploration of social justice.12
I shall return to these issues after considering the nature and implications of the huge difference between the formulations of social choice theory, with its focus on arriving at a ranking of alternative social realizations, and the form of mainstream theories of justice that concentrate not on the discipline of assessing improvements or declines of justice, but on the identification of the perfectly just social arrangements in the form of ‘just institutions’.
t h e r e a c h o f s o c i a l
c h o i c e
t h e o r y
Because of the apparent remoteness of formal social choice theory from matters of immediate interest, many commentators have tended to see its applicability as being extremely limited. The uncompromisingly mathematical nature of formal social choice theory has also 94
v o i c e a n d s o c i a l c h o i c e contributed to this sense of the remoteness of the discipline of social choice from applicable practical reason. Certainly, actual interactions between the theory of social choice and the pursuit of practical concerns have tended to be significantly discouraged by what is seen as a big gulf between exacting formal and mathematical methods, on one side, and readily understandable public arguments, on the other.
Not surprisingly, social choice theory is seen by many commentators as being at some disadvantage, in terms of practical relevance, compared with philosophical analysis of social justice. Even though the writings of Hobbes or Kant or Rawls demand arduous deliberation and intricate reflection, their central messages have appeared, in general, to be much easier to absorb and use, compared with what emerges from the discipline of social choice theory. The mainstream philosophical theories of justice, therefore, appear to many to be much closer to the world of practice than social choice theory can aspire to be.
Is this conclusion right? I would argue that not only is this conclusion wrong, almost the exact opposite may be true, at least in an important sense. There are many features of social choice theory from which a theory of justice can draw a great deal, as will be discussed later, but I begin here by pointing to what is certainly one of the most important contrasts between social choice theory and mainstream theories of justice. As an evaluative discipline, social choice theory is deeply concerned with the rational basis of social judgements and public decisions in choosing between social alternatives. The outcomes of the social choice procedure take the form of ranking different states of affair from a ‘social point of view’, in the light of the assessments of the people involved.* This is very different from a search for the supreme alternative among all possible alternatives, with which
* As will be discussed presently, the individual rankings that serve as informational inputs in the process can be interpreted in many different ways, and that versatility is important for the reach of social choice theory and its ability to adapt the social choice format to varying problems of social assessment. See Social Choice Re-examined, edited by Kenneth J. Arrow, Amartya Sen and Kotaro Suzumura (London: Macmillan, 1997); Handbook of Social Choice and Welfare, vol. 1, edited by Kenneth J. Arrow, Amartya Sen and Kotaro Suzumura (Amsterdam and Oxford: Elsevier, 2002; vol. 2
forthcoming); The Handbook of Rational and Social Choice, edited by Paul Anand, Prasanta K. Pattanaik and Clemens Puppe (Oxford: Oxford University Press, 2009).
95
t h e i d e a o f j u s t i c e theories of justice from Hobbes to Rawls and Nozick are concerned.13
The distinction is important, for reasons that have already been discussed in earlier chapters. A transcendental approach cannot, on its own, address questions about advancing justice and compare alternative proposals for having a more just society, short of the utopian proposal of taking an imagined jump to a perfectly just world.
Indeed, the answers that a transcendental approach to justice gives –
or can give – are quite distinct and distant from the type of concerns that engage people in discussions on justice and injustice in the world (for example, iniquities of hunger, poverty, illiteracy, torture, racism, female subjugation, arbitrary incarceration or medical exclusion as social features that need remedying).
t h e d i s t a n c e b e t w e e n
t h e
t r a n s c e n d e n t a l a n d
t h e
c o m p a r a t i v e
Nevertheless, important as this elementary contrast is, the formal remoteness of the transcendental approach from functional judgements about justice does not in itself indicate that the transcendental approach cannot be the right approach. There might well be some less obvious connection, some relationship between the transcendental and the comparative that could make the transcendental approach the right way of proceeding to comparative assessments. That investigation must be undertaken, but the temptation to believe that any transcendental theory must carry within its body some justificatory grounds that would also help to resolve all comparative issues is not well founded. As it happens, some transcendental theorists not only concede that there is a gap here, but do so proudly enough, asserting the folly of going into the comparative sidetrack (and it is indeed a sidetrack in the purely transcendental perspective). Robert Nozick, for example, is content to demand that all libertarian rights be fulfilled (this is his transcendental picture), but dismisses the issue of trade-offs between failures in the fulfilment of different types of rights (he has little use for what he calls ‘utilitarianism of rights’).14 Similarly, it is not easy to see how the diagnosis of perfection in the frameworks of 96
v o i c e a n d s o c i a l c h o i c e Hobbes, or Locke or Rousseau would take us to decisive comparisons among imperfect alternatives.
The story is more complex with Kant or Rawls, since their elaborate reasoning about the identification of the transcendental solution does offer clues to some – though not all – comparative issues as well. For example, Rawls’s formulation of the difference principle, a component of his second principle of justice, gives us ground enough to rank other alternatives in terms of the respective advantages of the worst-off.15 And yet this cannot be said about the other part of Rawls’s second principle, in which different violations of fair equality of opportunity would have to be assessed by criteria on which Rawls does not give us anything like a definitive guidance. The same can be said of the violations of liberties, which would negate the fulfilment of the first principle, since liberties are of different types (as Rawls himself discusses), and it is not at all clear how different violations of liberties would be comparatively assessed. There are different ways of doing this, and Rawls does not privilege any one way over others.
Indeed, he says relatively little on this question altogether. And that is, of course, fine for Rawls’s purpose, since a transcendental identification does not demand that this further comparative issue be addressed. A transcendental theory need not be what was called, in the Introduction, a ‘conglomerate’ theory (resolving transcendental and comparative issues simultaneously), and even though there is more articulation in Rawlsian reasoning about comparative questions than in many other transcendental theories, a big chasm still remains. A conglomerate theory is not needed by Rawls for his principles of justice (identifying perfectly just institutions), and he does not offer one.
But does not a transcendental identification in itself tell us something abou
t comparative issues, even when those issues are not explicitly confronted? Are there not some analytical connections here?
Are we being led astray by artificial separations that do not exist?
These doubts demand serious investigation. There are two questions, in particular, to address. First, could it be the case that transcendental identification of the perfectly just social arrangement will automatically tell us how to rank the other alternatives as well? In particular, can the answers to transcendental queries also take us, indirectly, 97
t h e i d e a o f j u s t i c e to comparative assessments of justice as a kind of ‘by-product’? In particular, could comparisons of ‘distance from transcendence’ at which the different societal arrangements stand be the basis of such comparative assessment? Could the transcendental approach be ‘sufficient’ for yielding much more than what its formal content suggests?
Second, if there is a query about sufficiency here, there is also one about necessity. Could it be the case that the transcendental question (‘what is a just society?’) has to be answered first, as an essential requirement, for a cogent and well-founded theory of comparative justice, which would otherwise be foundationally disjunctive and frail? Is the transcendental approach, aimed at identifying a perfectly just state, necessary for comparative judgements of justice as well?
Implicit beliefs in the sufficiency or the necessity (or both) of a transcendental approach for comparative assessment clearly have had a powerful role in the widespread conviction that the transcendental approach is crucial for the entire theory of justice.16 Without denying the practical relevance of, or intellectual interest in, comparative judgements, the transcendental approach has appeared to many theorists to be a central requirement of a well-grounded theory of justice.
The hypotheses of sufficiency and of necessity would, therefore, need closer examination to determine the substantive place of transcendental theories in the political philosophy of justice.
i s t h e t r a n s c e n d e n t a l
a p p r o a c h
s u f f i c i e n t ?
Does a transcendental approach produce, as a by-product, relational conclusions that are ready to be drawn out, so that transcendence may end up giving us a great deal more than its overt form articulates?
In particular, is the specification of an entirely just society sufficient to give us rankings of departures from justness in terms of comparative distances from perfection, so that a transcendental identification might inter alia entail comparative gradings as well?
The distance-comparison approach, even though it has some apparent plausibility, does not actually work. The difficulty lies in the 98
v o i c e a n d s o c i a l c h o i c e fact that there are different features involved in identifying distance, related, among other distinctions, to different fields of departure, varying dimensionalities of transgressions, and diverse ways of weighing separate infractions. The identification of transcendence does not yield any means of addressing these problems to arrive at a relational ranking of departures from transcendence. For example, in a Rawlsian analysis of the just society, departures may occur in many different areas, including the breaching of liberty, which, furthermore, can involve diverse violations of distinctive liberties (many of which figure in Rawls’s capacious coverage of liberty and its priority). There can also be violations – again, possibly in disparate forms – of the demands of equity in the distribution of primary goods (there can be many different departures from the demands of the ‘difference principle’).
There are many different ways of assessing the extent of each such discrepancy and of appraising the comparative remoteness of actual distributions from what the principles of full justice would demand.
We have to consider, further, departures in procedural equality (such as infringements of fair equality of public opportunities or facilities) which figure within the domain of Rawlsian demands of justice (in the first part of second principle). To weigh these procedural departures against infelicities of emergent patterns of interpersonal distribution (for example, distributions of primary goods), which also figure in the Rawlsian system, would require distinct specification – possibly in axiomatic terms – of relative importance or significance (or ‘trade-offs’
as they are sometimes called in the somewhat crude vocabulary of multidimensional assessment). But these valuations, helpful as they would be, lie beyond the specific exercise of the identification of transcendence and are indeed the basic ingredients of a ‘comparative’
rather than a ‘transcendental’ approach to justice. The characterization of spotless justice, even if such a characterization were to emerge clearly, would not entail any delineation whatever of how diverse departures from spotlessness would be compared and ranked.
The absence of such comparative implications is not, of course, an embarrassment for a transcendental theory itself, seen as a freestanding achievement. The relational silence is not, in any sense, an
‘internal’ difficulty; indeed, some pure transcendentalists would be 99
t h e i d e a o f j u s t i c e utterly opposed even to flirting with gradings and comparative assessments, and may quite plausibly shun relational conclusions altogether. They may point in particular to their understanding that a ‘right’ social arrangement must not, in any way, be understood as a
‘best’ social arrangement, which could open the door to what is sometimes seen as the intellectually slippery world of graded evaluations in the form of ‘better’ or ‘worse’ (linked with the relationally superlative ‘best’). The absoluteness of the transcendental ‘right’ –
against the relativities of the ‘better’ and the ‘best’ – may or may not have a powerfully reasoned standing of its own (I refrain from going into that issue here).* But it does not, of course, help at all – and that is the central point here – in comparative assessments of justice and therefore in the choice between alternative policies.
To be sure, members of any polity can imagine how a gigantic and totally comprehensive reorganization might be brought about, moving them at one go to the ideal of a fully just society. A no-nonsense transcendental theory can serve, in this sense, as something like the grand revolutionary’s ‘one-shot handbook’. But that marvellously radical handbook would not be much invoked in the actual debates on justice in which we are ever engaged. Questions on how to reduce the manifold injustices that characterize the world tend to define the domain of application of the analysis of justice; the jump to transcendental perfection does not belong there. It is also worth noting here the general analytical point, already noted in the Introduction, that the diagnosis of injustice does not demand a unique identification of ‘the just society’, since a univocal diagnosis of the deficiency of a society with, say, large-scale hunger, or widespread illiteracy, or rampant medical neglect, can go with very different identifications of perfectly just social arrangements in other respects.
Even if we think of transcendence not in the gradeless terms of
‘right’ social arrangements, but in the graded terms of the ‘best’ social arrangements, the identification of the best does not, in itself, tell us much about the full grading, such as how to compare two non-best alternatives, nor does it specify a unique ranking with respect to which
* See, however, Will Kymlicka, ‘Rawls on Teleology and Deontology’, Philosophy and Public Affairs, 17 (Summer 1988).
100
v o i c e a n d s o c i a l c h o i c e the best stands at the pinnacle; indeed, the same best may go with a great many different rankings at the same pinnacle.
To consider an analogy used earlier, the fact that a person regards the Mona Lisa as the best picture in the world does not reveal how she would rank a Picasso against a Van Gogh. The search for transcendental justice can be an engaging intellectual exercise in itself, but –
irrespective of whether we think of transcendence in terms of the gradeless ‘right’ or in the framework of the graded �
��best’ – it does not tell us much about the comparative merits of different societal arrangements.
i s t h e t r a n s c e n d e n t a l
a p p r o a c h
n e c e s s a r y ?
Consider now the hypothesis that the identification of the best, or the right, is necessary, even if not sufficient, to rank any two alternatives in terms of justice. In the usual sense of necessity, this would be a somewhat odd possibility. In the discipline of comparative judgements in any field, relative assessment of two alternatives tends in general to be a matter between them, without there being the necessity to beseech the help of a third – ‘irrelevant’ – alternative. Indeed, it is not at all obvious why in making the judgement that some social arrangement X is better than an alternative arrangement Y, we have to invoke the identification that some quite different alternative, say Z, is the very
‘best’ (or absolutely ‘right’) social arrangement. In arguing for a Van Gogh over a Picasso we do not need to get steamed up about identifying the most perfect picture in the world, which would beat the Van Goghs and the Picassos and all other paintings in the world.
It might, however, be thought that the analogy with aesthetics is problematic since a person might not even have any idea of a perfect picture, in a way that the idea of ‘the just society’ has appeared to many to be clearly identifiable, within transcendental theories of justice.
(I will argue presently that the existence of a best – or inviolate –
alternative is actually not guaranteed even by as complete a ranking of relative achievements of justice as possible, but I proceed, for the moment, on the presumption that such an identification can be made.) 101
t h e i d e a o f j u s t i c e The possibility of having an identifiably perfect alternative does not indicate that it is necessary, or indeed useful, to refer to it in judging the relative merits of two other alternatives; for example, we may indeed be willing to accept, with great certainty, that Mount Everest is the tallest mountain in the world, completely unbeatable in terms of stature by any other peak, but that understanding is neither needed, nor particularly helpful, in comparing the peak heights of, say, Mount Kilimanjaro and Mount McKinley. There would be something deeply odd in a general belief that a comparison of any two alternatives cannot be sensibly made without a prior identification of a supreme alternative. There is no analytical connection there at all.