ISSA Proceedings 1998 – Normative Argumentation In A Pluralist World

In this paper I discuss a problem for normative reasoning which arises from the particular circumstances of a pluralist world. I attempt to clarify the nature of the problem and consider possible responses to it. I then make suggestions about the form and content which a solution to the problem must possess.
In section 1 I introduce three simple thought experiments as an aid to fixing the nature of the problem. In section 2 I distinguish universalist responses from those of a more extreme form, and indicate why universalist responses are preferable. In section 3 I suggest that the problem is a strictly normative one rather than a strictly moral one. In section 4 I point out the difficulties in some recent universalist theories. In section 5 I propose a universalist theory based on the materiality of human beings.

1. The pluralist context
In order to see why normative argumentation becomes problematic in a pluralist world, it is useful to conduct the following three thought experiments.
1. Imagine that the world contains only two human communities. They are geographically separated and their members never come into contact or even know of one another’s existence. Community A is deeply religious, and its members observe a strict sabbatarianism. They also believe that it is natural for women to be the subordinates of men, so that obedience is regarded as an appropriate relation between a woman and her husband, and women are barred from the same kind of participation in public life as men. Finally, they regard abortion as one form of murder, and treat it as such. Community B is wholly secular. Its members believe that they have a right to dispose of their leisure time as they see fit as long as they do not infringe the right of others to act similarly. They believe that women and men are equal and strive to ensure that women are represented in public office in just the same way as men. They believe that a woman has a right to control over her own body, and regard the choice of abortion at will as one manifestation of that right.
2. Imagine now that members of A and B do come into contact, but in a peripheral way. Perhaps they have occasion to trade and in that way they come to learn about their differing views about the world, but otherwise they continue to live their lives separately from one another.
3. Imagine finally that there continue to be A persons and B persons but that there are no longer two separated communities. There is just one geographical area, and A persons may live next door to B persons.

In example (1) there is, in one clear sense, disagreement between communities A and B. Their respective members hold beliefs which are the contradictories of one another. In another clear sense there is no disagreement. Since they do not even know of one another’s existence, there is no occasion when an A person makes a claim which a B person then goes on to deny.
In example (2) there is liable to be disagreement in the second sense as well as the first. A persons and B persons may well take issue with one another where they differ, so that one will deny what the other asserts. But if we imagine that contact between the communities is minimal, the disagreement may not issue in conflict of any further kind.
In example (3) there will not merely be disagreement in the two senses distinguished. There will be practical difficulties directly connected with the beliefs of A persons and B persons. In acting on the respective beliefs they hold, A persons and B persons will come into conflict. They will be respectively committed to realising states of their world which cannot jointly be realised, and those commitments will arise directly from their beliefs.

I refer subsequently to the state of affairs outlined in (3) as the third possibility. It is this third possibility which most closely mirrors the circumstances of much of the contemporary world. There is not just the abstract fact of unwitting attachment to contradictory propositions, nor just the fact of witting denial of the propositions asserted by someone else. There is, in addition, the fact of manifest doxastic dissension issuing in practical dissension. The content of the beliefs in imagined communities A and B was chosen to reflect the content of beliefs which, in the actual world, result in practical conflict between people.
The circumstance of the third possibility has no doubt been responsible for producing recent interest in the problem of divergent normative reasoning in a pluralist world, and that is what I wish to explore. We live in a de facto pluralist world, a world in which incompatible systems of thought as a matter of fact coexist, systems conflicting courses of action in virtue of their espousal of those systems. What intellectual resources are there for dealing with conflicts arising in that way? I leave aside here any adjudication on the question of the normative pluralism expressed by Isaiah Berlin, according to which there is a plurality of genuine and objective values which may simply come into conflict with one another, what he describes as ‘the permanent possibility of inescapable conflict between values’ (Berlin 1991: 80. See also Larmore 1994: 62-3).
Berlin’s conclusion is a drastic one. My concern is with how far we might deal with conflicting values and people’s attachment to them, how far we might proceed in some kind of neutral and objective evaluation of them, before reaching the point where we are forced to conclude that no further resolution is possible.

2. Responses to de facto pluralism
We live in a world where people begin from differing assumptions, employ differing forms of reasoning and end up with differing conclusions. And all of this matters at the practical level. A range of responses to this dilemma is possible. At one extreme, we might long for a world in which people’s reasoning converges, where they all agree on what is of value and what not. At the other extreme, we might abandon any attempt to measure the diverging views against one another by retreating into some form of relativism. In the latter spirit, consider the pragmatist attempt to distinguish between fanaticism and a conscience worthy of respect. The criterion for this, according to Richard Rorty, ‘can only be something relatively local and ethnocentric -the tradition of a particular community, the consensus of a particular culture. According to this view, what counts as rational or as fanatical is relative to the group to which we think it necessary to justify ourselves – to the body of shared belief that determines the reference of the word “we” ’ (Rorty 1991: 176-7).
We might well feel that such responses are best fitted to some other circumstances than the ones we actually face: the former to a world where either community A or community B never existed; the latter to the lost world where community A and community B were entirely separate from each other. Neither really promises to negotiate the problem outlined in the third possibility of section 1 in a way which will produce an intellectually satisfying solution to conflict of values. For that, the former response would need to convince us that, from where we are now, there is some reasonable prospect or achieving consensus on currently disputed fundamentals. The second response, in its turn, would need to convince us either that no progress could be made on those disputes or that the very idea of progress in this context is a myth.
A distinct response consists in the Rawlsian view that a diversity of reasonable comprehensive doctrines is ‘not a mere historical condition that may soon pass away; it is a permanent feature of the public culture of democracy. Under the political and social conditions secured by the basic rights and liberties of free institutions, a diversity of conflicting and irreconcilable – and what’s more, reasonable – comprehensive doctrines will come about and persist if such diversity does not already obtain’ (Rawls 1993: 36). The burdens of judgement in a modern society, including the fact that people’s total experiences are very diverse, allow them to reach different views even when exercising their reason. ‘Different conceptions of the world can reasonably be elaborated from different standpoints and diversity arises in part from our distinct perspectives. It is unrealistic – or worse, it arouses mutual suspicion and hostility – to suppose that all our differences are rooted solely in ignorance and perversity, or else in the rivalries for power, status, or economic gain’ (Rawls 1993: 58). The major project then becomes that of determining at least a set of political arrangements which people can agree to from their conflicting comprehensive standpoints. But the standpoints themselves, as long as they are reasonable, are left untouched.

However, there is a prior question about which Rawls avowedly says little, and that is what conditions a doctrine must meet in order to qualify as reasonable. He tells us that reasonable comprehensive doctrines involve the exercise of theoretical reason to produce something consistent and the exercise of practical reason in determining priorities; and he claims that no tighter criterion is needed for the purposes of political liberalism (Rawls 1993: 59-60; cf. Rawls 1993: 37 n.38). This last claim may well be true, but there is then an unresolved question of whether the number of comprehensive doctrines for which an accommodation must be found can be reduced at all, whether it is possible to judge that some such doctrines are unreasonable and therefore open to criticism. No doubt a certain humility is appropriate when faced with a set of values which have held the allegiance of a large number of people over a significant period of time, but the possibility cannot be ruled out that some values, even if deeply and widely held, may be in some way deficient or wrong-headed (And it goes without saying that if we countenance that possibility, then we must countenance it in relation to our own values as well as other people’s).

In order to open up the possibility of judging some such doctrines to be unreasonable, it is necessary to establish the prior possibility of a standpoint from which that criticism could be made without seeming to favour any particular culture in its very operation. A number of theories respond to the problem of de facto pluralism by, in effect, attempting to define such a standpoint. They are  universalist in that they attempt to escape the limitations of a particular set of values by focusing on what is universal in human life. If, in all circumstances, there are certain things which we must value whatever else we value, this will provide us with a compelling starting point which is not local to any particular culture. It will also give us a criterion, relatively free from cultural bias, by which to judge the views of particular communities.

The Rawlsian theory of primary goods is itself the most obvious example of such a theory (Rawls 1972: 62, 92-3, 434; Rawls 1993: 75, 180-1, 298). Others are provided by Gewirth (1994); Kekes (1994); and Doyal and Gough (1991). Gewirth, for example, argues that freedom and well-being are prerequisites of all human action; that any agent must conclude that they have a right to them and that other agents have similar rights; and that the universal requirement of freedom and well-being can then be used to judge particular cultures, in terms of how far they make these provisions for everyone (Gewirth 1994: 22-43). Kekes argues in similar terms. He draws up a longer list of ‘primary values’, physical, psychological and social, the satisfaction of which is a prerequisite for a good life, and argues that these primary values ‘constitute a context-independent ground for settling some conflicts among values’ (Kekes 1994: 50). Doyal and Gough claim that ‘since physical survival and personal autonomy are the preconditions for any individual action in any culture, they constitute the most basic human needs – those which must be satisfied to some degree before actors can effectively participate in their form of life to achieve other valued goals’ (Doyal and Gough 1991: 54).[i]
Such universalist theories must pass two tests if they are to avoid cloaking cultural parochialism in merely apparent universality. They must avoid parochialism both in content and in form. That is to say, they must take a wide enough view of human behaviour to ensure that what they pick out as a universal feature of human life really is so, rather than being confined to our own or some other culture;[ii] and they must take care that, having found such a genuinely universal feature, they do not describe it in a way which is itself prejudicial from the standpoint of particular cultures. I take up the issues of form and content respectively in the following two sections.

3. The form of universalism
As an example of the need to avoid parochialism in form, consider Gewirth’s claim that all human agents have a right to freedom and well-being. He acknowledges the objection that the concept of rights is a local one and is particularly suited to a culture where there is great emphasis on the individual in contrast to the group.
He dismisses the objection on the grounds that ‘most moral and other practical precepts are addressed, directly or indirectly, to individuals’ (Gewirth 1994: 34). The argument from rights proceeds from that common assumption and so is held not to be of merely local or ethnographic validity. He argues further, and in a more explicitly moral way, that ‘the primary point of human rights is to protect individuals from unjustified threats to their freedom and well-being on the part of communities or cultures to which they may belong’ (Gewirth 1994: 35).
I leave aside here the contentious claim about most practical precepts being addressed to individuals.[iii] Suppose indeed that it is a universal truth that all human beings have rights of the kind specified by Gewirth. Even then, it would not follow that this was the appropriate form in which to couch a consideration which was to function as a criterion for assessing the rival claims of different cultures. Precisely because the concept of rights is so highly culture-specific and contested, it does not provide a sufficiently independent starting point for such assessment. We should have to argue to a proposition about rights rather than arguing from one, and that makes such a proposition unsuitable for the task in hand.
If this criticism of the form taken by Gewirth’s criterion is justified, someone might infer from it that we simply need a more universally acceptable form in which to couch the moral consideration which is to function as our criterion. But I want to argue for a stronger conclusion than that. I want to suggest a shift away from any specifically moral consideration as providing the required criterion. Moreover, I make this suggestion not because of the contentious nature of morality but because of the nature of the problem to which the criterion is meant to provide a solution.
Consider again the situation which gives rise to the problem. People hold varying and conflicting views about how they ought to behave; and where they live in juxtaposition, this issues in practical conflict which is itself an expression of the conflicting views. The problem then arises from a clash of practical attitudes and beliefs rather than from morality as such. That is an important difference, and it makes the problem both wider and narrower than a purely moral problem since not all practical thinking is moral and not all morality is practical.

The problem is wider, because it is replicated wherever people hold varying views about how to behave, whether those views are specifically moral or not. For example, they may hold varying views about their own or other people’s interests, about what it would be prudent to do rather than what it would be moral to do, and they may attempt to act to realise those interests in ways which issue in practical conflict. The problem is narrower, because not all of morality is concerned in a direct way with practical conclusions. Some moral thinking is concerned with assessment, for example of character or disposition, in ways which stop short of any immediate connection with action. In those circumstances, there can be disagreements, but they more closely resemble the circumstances of example (2) in section 1 rather than the circumstances of the third possibility.
Accordingly, I suggest that we cease to see the problem as one about conflict in moral reasoning and instead see it as a problem about conflict in normative reasoning, where that term is used to denote any reasoning connected with decisions about what to do, in contrast to theoretical reasoning which is connected only with what is true. Normative reasoning therefore includes moral reasoning but also, for example, reasoning about what is in one’s own or someone else’s interests. That shift in the way of seeing the problem dictates a similar shift in the search for a solution. We should cease to ask: Is there some universal feature of human life which provides material for a culturally independent moral criterion by which to judge the rival claims of different cultures? Instead, we should ask: Is there some universal feature of human life which carries a culturally independent relevance to reasons for acting?
The concept of a reason for acting is a much better candidate for possessing the required neutrality of form for some universal consideration to take. It is already possessed by any deliberative agent as a necessary part of their conceptual equipment, and it is not in itself contentious or contestable. Of course, deliberative agents disagree about what reasons for acting there are, as well as what kind of reasons there are. But they do not and could not disagree in using the idea of a reason for acting. This different starting point is therefore preferable for dealing with the problem set by the third possibility in section 1: it is independent of particular cultures and it is of universal application.

4. The content of universalism
It is another and more complicated question whether anything in human life possesses the required universality to provide a reason for acting for all human agents. Is there anything which, regardless of cultural context, is necessarily germane to all human agents? Is there any universal prerequisite of human agency, irrespective of the particular goals which a human agent has? It will be plain that freedom or autonomy are favoured candidates for that role. There is a problem, however. Freedom and autonomy are themselves morally saturated notions, and the danger is that as soon as we begin to fill out their content we find that we are once again using a concept in a way which will not be universally assented to in all cultures.
The point is illustrated by Philip Pettit’s recent sponsorship of freedom as being ‘capable of commanding the allegiance of the citizens of developed, multicultural societies, regardless of their more particular conceptions of the good’ (Pettit 1997: 96). But it is the ideal of freedom specifically as nondomination which Pettit believes can play this role, and he has to face the objection that such an ideal is not neutral and will not command universal allegiance. His response is that traditions which reject that ideal and display a tendency to subject oneself to, for example, those with a priestly role, involve ‘the suppression of a deep and universal human desire for standing and dignity….Embrace the life of a sect who abase themselves before some self-appointed guru and you will see little in the idea of freedom as nondomination. Embrace the life of a contemporary, pluralistic society and you will see much’ (Pettit 1997: 96-7).
The sponsorship of freedom or autonomy specifically in the form of nondomination looks like a clear departure not just from neutrality but also from a universal starting point. Freedom or autonomy in that form is certainly not a necessary condition of all human agency. Separately from whether such a state of affairs is desirable, it is plain that even a slave is capable of many instances of human agency. Moreover, Pettit’s response ignores the fact that sects which worship gurus often exist within a contemporary pluralistic society, and that is precisely what gives rise to the problem in the third possibility of section 1. We are not given here a reason for embracing the ideal of nondomination, only the assertion that for anyone who has embraced it a certain kind of problem will not arise.

If we wish to retain freedom or autonomy as the culturally-neutral and genuinely universal consideration then we must avoid any contentious or merely local conceptualisation. It is possible to do this, but doing so carries a price. Thus, Doyal and Gough begin with a minimal definition according to which to be autonomous ‘is to have the ability to make informed choices about what should be done and how to go about doing it’ (Doyal and Gough 1991: 53). But as they themselves acknowledge, this description of  autonomy is tantamount to a description of agency itself. If that is correct, then it cannot at the same time be treated as a separate prerequisite of agency. The danger in looking for something which is genuinely universal and genuinely tied to agency itself, in all forms and in all circumstances, is the danger of disappointment: that all we can unearth is a tautology.
The consideration of health suffers from some of the same drawbacks as freedom or autonomy. Doyal and Gough tell us, for example, that ‘physical health can be thought of transculturally in a negative way. If you wish to lead an active and successful life in your own terms, it is in your objective interest to satisfy your basic need to optimise your life expectancy and to avoid serious physical disease and illness conceptualised in biomedical terms. This applies to everyone, everywhere’ (Doyal and Gough 1991: 59). One difficulty which this claim in its universalised form has to meet is dramatised in the case of the philosopher Brentano, who said that he welcomed his blindness since it enabled him to concentrate on his philosophy. That suggests that placing a priority on health does not have the universal reach which they might think. Their reply is that ‘Such arguments ignore the fact that Brentano had to possess enough physical health to acquire the conceptual tools necessary to respond to his disablement in the enhanced way he claimed’ (Doyal and Gough 1991: 316 n6).
The tensions between avoiding lack of universality and avoiding lapse into tautology are once again apparent. There are people whose aspirations are not for anything which they or their own culture would really call an active and successful life: they might, for example, aspire to achieve a certain contemplative state. There are people who want to achieve particular goals even at the cost of the loss of longevity or of risking falling prey to disease. The needs specified by Doyal and Gough will not apply in such cases. At that point the temptation is therefore to move in the direction of tautology: at least such people need a sufficient degree of health for their particular aspirations to be met. That is correct, but it has the consequence that their needs will be different from those originally specified and may in fact be extremely minimal in the relevant respects. They may in fact amount to no more than this: that if there is something you aspire to do, you need not to be dead, you need to be alive long enough to do it and in a fit state to do it.

5. Materiality
I now want to suggest that the connection between agency and materiality does not suffer the same defects as those suffered by the considerations linked with agency in section 4.[iv] We can take the idea of agency in its most general and uncontentious sense, the idea simply of human beings doing things in the world, and truthfully assert that agency in that uncontentious sense always has material prerequisites. It is a consequence of our being the kind of creatures we are that our survival from one moment to the next depends on the satisfaction of a range of material needs, a range which expands as we conceive of more extended forms of action whose execution takes more than a moment. Thus, my scratching my nose in a moment depends on an uninterrupted supply of oxygen to my brain; my posting a letter tomorrow depends on my receiving sustenance sufficient to support my continued biological functioning; my completing a philosophy paper in the next month depends on my having shelter and clothing to protect me from the elements. In addition, the latter two actions also depend on the availability of a range of material objects which I can employ in various ways, and that will be typical of most actions above a low level of triviality. But the fact of material prerequisites for my own existence as an agent is absolutely exceptionless.
This consideration, then, is universal in content. All human agents must satisfy certain material needs as a precondition of exercising their agency in any matter whatsoever. We are not speaking here of a local truth which might fail to hold in some other place. We can be entirely confident that we shall not come across a culture where people can carry out their plans of action without needing to meet material requirements. If we came across creatures of whom that was true, it would be no mere verbal matter to refuse to apply the term ‘human being’ to them. They would be so fundamentally different from us that there would be substantive reasons for such a refusal.
Has this material consideration been stated in a culture-neutral form, that is, in terms which do not covertly and illicitly favour one culture over another? It is tempting to think so, on the grounds that no culture can or does deny these obvious facts or dispute the way in which they are expressed. In discussing Bruno Snell’s account of Homeric images of the person Bernard Williams says: ‘We do indeed have a concept of the body, and we agree that each of us has a body. We do not, pace Plato, Descartes, Christianity, and Snell, all agree that we each have a soul. Soul is, in a sense, a more speculative or theoretical conception than body’ (Williams 1993: 26). Of course, one could imagine someone resisting these claims, but they would have to adopt an extreme position in order to do so in a form which challenged the cultural neutrality of the claim that satisfaction of material need is a precondition of all action. For example, it would not be sufficient to object that we are essentially souls and are merely temporarily trapped in our bodies. For as long as we are so trapped, the claims of the previous paragraph stand. It might instead be said that we merely have the illusion of being embodied and that what is real is our souls. But then while we suffer from such an illusion we have no alternative but to accompany it with the further illusion that we have to act so as to meet our illusory bodily needs. The objection then begins to look less like a rejection of the terms in which the claimed universality is expressed and more like an objection to its presumed philosophical status. But even a whole culture which took this position would have to feed and clothe itself in order to advocate it or to do anything else.

A related point reinforces the claim that the materiality consideration can be posited in a culture-neutral way. An obvious distinction can be made between things which are important to us because we invest them with importance and things which are important to us whether we think they are or not. An example of the former would be the pain of social opprobrium arising from having children out of wedlock. That is something dependent on social attitudes. An example of the latter would be the pain associated with falling off a cliff. Our materiality is of the latter kind. It is important whatever we think about it, and whether we think about it or not. But that makes it, in itself, an objective consideration, and to that extent beyond the reach of any particular culture (though of course there can be crucial cultural variation in the way that objective fact is perceived and theorised).

Consider now how this universality of form and content extends the reach, as it were, of the claim that any agents must concern themselves with the meeting of their material needs. We have examined theories which attempt to establish what things agents must concern themselves with for living a good life or furthering their rational plan of life. Even leaving aside the clear contentiousness of the contents of a good life and the arguable contentiousness of the contents of a rational plan of life, the materiality consideration extends well beyond these theories. It picks out what is a prerequisite of any life at all. Indeed, it goes beyond what any agents must concern themselves with and speaks to what any human beings must concern themselves with. Imagine, for example, someone who has no interest in acting at all but aspires simply to experience certain states. Then exactly the same considerations will apply: they must concern themselves with the meeting of their material needs for this aspiration too to have any chance of success. As well as being able to speak to these outer reaches of human life, as it were, the material consideration can also do all of the work of the other and more contentious favoured considerations, and in a way it subsumes them. Thus, suppose that your primary value is autonomy, the living of your life with the shape you have chosen to give it, rather than that allotted by someone else or by inanimate circumstances. Then you have strong grounds for being concerned with your materiality, because the securing of your material survival is itself a precondition of achieving such autonomy. But if your primary value is something quite different from autonomy, for example a life of service and dedication to the wishes of your master, then exactly the same will be true. That is, whatever your goals, you have a reason to concern yourself with your material circumstances.
Some caveats should be entered about what is established in this paper. We began by asking whether it was possible to arrive at some starting point for judging different cultures which was itself universal and not biased at the outset in favour of or against particular cultures. The suggestion now is that the materiality of human beings is such a starting point, since it is necessarily relevant to the practical reasoning, moral or otherwise, of all agents in all circumstances. It is no more than a starting point. I have suggested that it meets the formal conditions which any candidate for this role must meet, but it is another matter altogether actually to put it to work in the assessment of the values of different cultures. For that, we should have to construct a theory similar in nature to Gewirth’s, which arrived at some metric for judging the adequacy of different societies’ arrangements for meeting the material needs of its members. That would be a colossal and complex task.
The fact that it is a further task might allay fears which would otherwise arise about the stress here on materiality. For example, it might be felt that such a stress must betoken subscription to the Promethean character of both liberalism and Marxism in their inappropriate perception of the relation between human beings and external nature as one of mastery and control (cf. Benhabib 1992: 69). Or it might even be felt that it must betoken subscription to a crass materialism which simply judges cultures according to the extent of their theoretical or practical commitment to maximising material consumption.
Subscription to either of these positions would be incompatible with the use of the material consideration as a neutral and universal arbiter among different cultures’ values. So far, the only information which can be legitimately used for that task are the facts that human beings are material creatures; that the satisfaction of their material needs is a precondition of their acting; and that they themselves must act so as to secure the satisfaction of those needs. There may be many objectionable beliefs and values which come to be associated with those facts, including an insensitive and unduly utilitarian attitude towards the natural world, but the facts themselves are not in dispute. It is therefore a matter for further negotiation what follows from them.
That said, such negotiation is precisely what should occur. If the point of the exercise is to enable us to make comparative judgements about differing values, then an impartiality which is appropriate at the outset would be entirely inappropriate at the end of the process. Judgements have to be made and criticisms levelled. But that will be possible when we have a much fuller theory of materiality. The material considerations do not just pass the minimal test of possessing the rather abstract properties required of a consideration which is to serve in adjudication of rival views. They impinge on our lives in a series of complicated ways which touch on our vital interests in a pervasive way, whatever those interests are taken to be. (I attempt to set out some arguments to establish that point in Graham 1998.) Indeed, it is precisely because we have other aims,  beyond the mere maintenance of material existence, that we need to take account of the relations we must enter into in order to maintain ourselves in a condition where we can pursue those other aims. That is why we have to take our materiality so seriously, whatever our values.

NOTES
i. Elsewhere, Gough says that physical health and autonomy are ‘universal rerequisites for any person ‘successful participation in whatever form of life she finds herself in, or chooses to live in’ (Gough 1996: 82).
ii. Rawls’s list of primary goods is an uneasy combination of universal and culture-specific features of human life. For discussion of that aspect of his theory, see Graham 1996: 141-3.
iii. I argue for the existence and practical importance of irreducibly collective actions in Graham (forthcoming).
iv. The attempt to connect freedom with materiality has a long history. James Harrington, for example, argues the need for a person to have material resources if they are to be free: ‘The man that cannot live upon his own must be a servant; but he that can live upon his own may be a freeman’ (Harrington 1992: 269, cited in Pettit 1997: 32). But that attempt suffers from the disadvantage which I have discussed above, that freedom is a morally saturated notion, and the required neutrality is lost as soon as it becomes clear that it is being interpreted in a particular way which favours one understanding of freedom rather than some other. For a related and contemporary connection between autonomy and materiality, see Christman 1994.

REFERENCES
Benhabib, S. (1992). Situating the Self. Cambridge: Polity Press.
Berlin, I. (1991). The Crooked Timber of Humanity. New York: Knopf.
Christman, J. (1994). Distributive Justice and The Complex Structure of Ownership. Philosophy and Public Affairs, 23, 225-250.
Doyal, L. and Gough, I. (1991). A Theory of Human Need. Basingstoke: Macmillan.
Gewirth, A. (1994). Is Cultural Pluralism Relevant to Moral Knowledge? Social Philosophy and Policy 11, 22-43.
Gough, I. (1996). Justifying Basic Income? A Review of P. Van Parijs, Real Freedom for All. Imprints, 1, 72-88.
Graham, K. (1996). Coping with the Many-Coloured Dome: Pluralism and Practical Reason. In: D. Archard (Ed.), Philosophy and Pluralism (pp. 135-46), Cambridge: Cambridge University Press.
Graham, K. (1998). Being Some Body: Choice and Identity in a Liberal Pluralist World. In: B. Brecher, J. Halliday and K. Kolinskà (Eds.), Nationalism and Racism in the Liberal Order (pp. 176-192, Ch. 12), Aldershot: Ashgate.
Graham, K. (forthcoming). Collective Responsibility. In: T. van den Beld (Ed.), Moral Responsibility and Ontology. Dordrecht: Kluwer.
Harrington, J. (1992). Commonwealth of Oceana and a System of Politics,. J G A Pocock (Ed.), Cambridge: Cambridge University Press.
Kekes, J. (1994). Pluralism and the Value of Life. Social Philosophy and Policy 11, 44-60.
Larmore, C. (1994). Pluralism and Reasonable Disagreement. Social Philosophy and Policy 11, 61-79.
Pettit, P. (1997). Republicanism: A Theory of Freedom and Government. Oxford: Clarendon Press.
Rawls, J. (1972). A Theory of Justice. Oxford: Oxford University Press.
Rawls, J. (1993). Political Liberalism. New York: Columbia University Press.
Rorty, R. (1991) The Priority of Democracy to Philosophy. In: Objectivity, Relativism and Truth: Philosophical Papers, vol. 1 (pp. 175-96), Cambridge: Cambridge University Press. Williams, B. (1993). Shame and Necessity. Berkeley: University of California Press.