¹ Rawls, John, A Theory of Justice (Oxford: Oxford University Press, 1971).Google Scholar

² Rawls, John, Political Liberalism (New York: Columbia University Press, 1993).Google Scholar

³ Steiner, Hillel, An Essay on Rights (Oxford: Blackwell, 1994).Google Scholar

⁴ Nozick, Robert, Anarchy, State, and Utopia (New York: Basic Books, 1974).Google Scholar

⁵ Dworkin, Ronald, ‘Foundations of Liberal Equality’, in Peterson, Grethe B., ed., The Tanner Lectures on Human Values, XI, 1990 (Salt Lake City: University of Utah Press, 1990), pp. 16–22Google Scholar. The distinction between these two strategies has been more frequently described as that between neutralist and perfectionist forms of liberalism. However, Dworkin's terms are preferable since a continuous or perfectionist liberalism might endorse policies of liberal neutrality if those are appropriate vehicles for realizing its conception of the good.

⁶ Raz, Joseph, The Morality of Freedom (Oxford: Clarendon Press, 1986).Google Scholar

⁷ Rawls, , Theory of Justice, pp. xv–xviGoogle Scholar

⁸ Rawls, , Political Liberalism, p. xviiiGoogle Scholar; see also pp. xxv, 4, 47, 133.

⁹ Rawls, , Political Liberalism, p. 16.Google Scholar

¹⁰ The capacity for a sense of justice is ‘the capacity to understand, to apply, and to act from the public conception of justice which characterties the fair terms of social cooperation’ and ‘a willingness … to act in relation to others on terms that they also can publicly endorse. ‘The capacity for a conception of the good is the capacity to form, to revise and rationally to pursue a conception of one's rational advantage or good.’ (Political Liberalism, p. 19).Google Scholar

¹¹ Rawls, , Political Liberalism, p. 46.Google Scholar

¹² Rawls, , Political Liberalism, p. 54.Google Scholar

¹³ Rawls, , Political Liberalism, pp. 5–6.Google Scholar

¹⁴ Rawls, , Political Liberalism, p. 224.Google Scholar

¹⁵ Non-public reasons may be shared by, and therefore be public to, a particular group or association, even though they are not public to political society as a whole; they are not necessarily reasons unique to individuals. That is why Rawls prefers to call them ‘non-public’ rather than ‘private’ reasons (Political Liberalism, p. 220).Google Scholar

¹⁶ Rawls, , Political Liberalism, p. 243, n. 32Google Scholar. Rawls takes ‘the normal case of mature adult women’ and considers the question of abortion in terms of three political values: ‘due respect for human life’, ‘the ordered reproduction of political society overtime’ and ‘the equality of women as equal citizens’. He moves swiftly to the conclusion that ‘any reasonable balance of these three values will give a woman a duly qualified right to decide whether or not to end her pregnancy during the first trimester’ and perhaps also ‘a right beyond this, at least in certain circumstances’. The reason he offers is that ‘at this early stage of pregnancy the political value of the equality of women is overriding’. What is breath-taking in all this is not Rawls's verdict on abortion but the simple assurance with which he declares that to be the verdict of ‘public reason’ and the ease with which he dismisses doctrinal views to the contrary as ‘unreasonable’.

¹⁷ Rawls, , Political Liberalism, p. 229.Google Scholar

¹⁸ Rawls, , Political Liberalism, p. 214.Google Scholar

¹⁹ For example, Rawls, , Political Liberalism, pp. 9, 12, 15, 24–5, 37.Google Scholar

²⁰ For example, Rawls, , Political Liberalism, pp. 40, 140, 160, 169.Google Scholar

²¹ I take it that Rawls does not adopt this easy mode of reconciliation because his concern is not to play with words but to show that his political conception of justice would be compatible with most of the comprehensive doctrines that are likely to be found in a modern democratic society. Recall that he invokes the idea of an overlapping consensus not to establish the rightness of the political conception of justice but to show that a modern society based on that conception has a good chance of being stable.

²² Rawls, , Political Liberalism, pp. 158–68.Google Scholar

²³ Rawls, , Political Liberalism p. 59.Google Scholar

²⁴ Similarly, if Rawls's distinction between simple and reasonable pluralism is to have the political significance he claims for it, that distinction must be based not upon his definition of the reasonableness of doctrines, but upon the reasonableness with which they are held. Rawls insists that the distinction between simple and reasonable pluralism has no bearing upon the content of his conception of justice; with whichever type of pluralism we sought to cope, we would still end up selecting the same principles of justice (Political Liberalism, pp. 64–5)Google Scholar. The distinction between the two forms of pluralism is significant only for the issue of stability.

²⁵ Rawls, , Political Liberalism, pp. 157, 169, 218.Google Scholar

²⁶ Compare the argument of Cohen, Joshua, ‘Moral Pluralism and Political Consensus’, in Copp, D., Hampton, J. and Roemer, J. R., eds, The Idea of Democracy (Cambridge: Cambridge University Press, 1993), pp. 270–91Google Scholar. Rawls acknowledges Cohen as the source of his distinction between simple and reasonable pluralism (Political Liberalism, pp. xxxi, 36).Google Scholar

²⁷ Rawls, , Political Liberalism, p. 13.Google Scholar By a ‘democratic society’, Rawls seems to mean nothing more precise than the kind of society characteristic of Western liberal democracies. Elsewhere Rawls formulates this clause as ‘certain fundamental ideas seen as implicit in the public political culture of a liberal society’ (my emphasis); ‘The Law of Peoples’, in Shute, Stephen and Hurley, Susan, eds, On Human Rights: The Oxford Amnesty Lectures 1993 (New York: Basic Books, 1993), pp. 41–82, 220–30, at p. 221.Google Scholar

²⁸ Cf. Rorty, Richard, Contingency, Irony and Solidarity (Cambridge: Cambridge University Press, 1989), pp. 59–60.CrossRefGoogle Scholar

²⁹ See also Rawls's defence of abstraction in political philosophy (Political Liberalism, pp. 43–6)Google Scholar, particularly his remark that ‘We turn to political philosophy when our shared political understandings, as Walzer might say, break down, and equally when we are torn within ourselves’ (p. 44).

³⁰ In ‘The Law of Peoples’, Rawls argues for the toleration of well-ordered non-liberal societies. However, a well-ordered society is not merely one that exhibits a de facto consensus. Rawls observes that a political culture may be ‘distorted and corrupt’ (p. 229)Google Scholar and holds that a regime may be morally inadequate because of ‘the nature of the public political culture and the religious and philosophical traditions that underlie its institutions’ (p. 77; also pp. 74–5).

³¹ Rawls, , Political Liberalism, pp. xix, 46, 100.Google Scholar

³² In his ‘Law of Peoples’, Rawls seems to affirm the antifoundational character of his thinking by allowing that non-liberal societies, which are organized according to conceptions of justice different from his own, need not be considered unjust on that account. He also allows that those societies, to be just, need not embrace the liberal's political conception of persons as free and equal citizens (pp. 65, 66, 69). Yet Rawls seems careful to extend toleration to non-liberal societies only on terms which do not seriously compromise personhood. For one thing, his model of a well-ordered non-liberal society is a highly consensual one in which a people as a whole is committed to a comprehensive doctrine; their collective pursuit of that doctrine does not therefore involve ‘imposing’ it upon their individual members. For another, Rawls extends international toleration only to ‘well-ordered’ non-liberal societies and those are societies which are guided by a ‘common good conception of justice’ and which respect individual human rights.

³³ Suppose that X promises Y that he will give something desirable to Z. That arrangement causes two difficulties for the benefit theory. First, the direct and intended beneficiary of X's promised duty is Z, but it is not clear that Z has a right against X since X's duty is owed to Y, the promisee, rather than to Z, the beneficiary. Secondly, even though Y is not the direct and intended beneficiary of the promise, it is Y who possesses the right that X shall perform his duty since Y is the promisee and, as such, it is he who is entitled to waive or to cancel X's duty or to insist upon its performance. Thus it would seem both that beneficiaries need not possess rights and that non-beneficiaries can possess rights. For further discussion of this case and of the benefit and choice theories generally, see my Rights (Basingstoke: Macmillan, 1994), chap. 2.Google Scholar

³⁴ A person has a claim-right when he is owed a corresponding duty (e.g. the right to have a debt repaid, the right not to be assaulted); he has a liberty-right when he has no duty to the contrary (e.g. the right to express his views or to dress as he pleases); he has a power when he is formally enabled to effect a transaction (e.g. his right to marry or to vote); he has an immunity when he is not subject to another's power (e.g. his right not to have his property sold by another). If A is the right-holder and B is the other party in the relation, then, when A has a claim-right to x, B has a corresponding duty to provide A with x; when A has a liberty-right to do x, B has a corresponding ‘no-claim-right’ that A shall not do x; when A has a power, B has a corresponding liability; and when A has an immunity, B has a corresponding disability or ‘no-power’. Cf. Hohfeld, Wesley N., Fundamental Legal Conceptions as Applied in Judicial Reasoning (New Haven, Conn.: Yale University Press, 1919).Google Scholar

³⁵ To simplify matters, I set aside the case of rights which correlate with disabilities, as does Steiner, (Essay on Rights, p. 73).Google Scholar

³⁶ Notice that for Steiner the terms ‘freedom’ and ‘liberty’ are importantly different. ‘Freedom’ is an empirical concept – to be free to do x is to be, as a matter of fact, unprevented from doing x. ‘Liberty’, which Steiner uses in a Hohfeldian sense, is a normative concept – to be at liberty to do x is to have no duty not to do x (and to be at liberty not to do x is to have no duty to do x).

³⁷ I do not stop to examine Steiner's analysis of freedom which is a separate subject in its own right. However, for those who find his conceptions of freedom and unfreedom rather too parsimonious, the simplest response may lie in a recourse to MacCallum's well-known triadic analysis of freedom, according to which a complete statement concerning freedom must be of the form: X (the agent) is free/unfree from Y (the obstacle) to do Z (the action). Steiner's conception of freedom can be expressed in terms of that triad, but he seems to think that we should be able to make simple diadic statements about freedom which make unnecessary any reference to the middle term (Y). In other words, we should be able to say simply that X is either free or unfree to do Z. If, however, we reintroduce an essential reference to the middle term (Y), we can say, for example, that X was free from physical restraints to commit murder, but that he was not similarly free from moral or legal obligations or from the threat of legal punishment. Spelling things out in that way removes any sense of absurdity from observing that X did something that he was (in some respects but not others) unfree to do.

³⁸ Steiner, , Essay on Rights, p. 39.Google Scholar

³⁹ Nor does the test of compossibility necessarily favour the choice theory of rights over the benefit theory. That x is in my interest is normally considered a necessary rather than a sufficient condition of my having a right to x; we might add, like Raz, Joseph (Morality of Freedom, p. 166)Google Scholar, that I have a right to x only if my interest in having x is such as to justify another's having a duty to provide me with.x. We might add the further condition that interests can generate duties only in so far as those duties are compossible, in which case they will be paired with compossible rights. Thus we would have an interest-based but compossible set of rights; whether those rights were in any way objects of choice (i.e. accompanied by powers and liberties) would seem immaterial to their compossibility. Certainly compossibility is not a necessary feature of every possible version of the benefit theory of rights, but neither is it a necessary feature of every possible version of the choice theory.

⁴⁰ Steiner, , Essay on Rights, p 2.Google Scholar

⁴¹ Steiner, , Essay on Rights, p. 198.Google Scholar

⁴² Steiner, , Essay on Rights, p. 199.Google Scholar

⁴³ Dworkin, Ronald, Taking Rights Seriously (London: Duckworth, 1978), pp. xi, 90–4, 364–8Google Scholar; A Matter of Principle (Oxford: Clarendon Press, 1985), pp. 335–72Google Scholar. In one important respect Steiner goes beyond the metaphor. While a trump card beats any card of other suits, a higher trump card also beats a lower trump card. But within a set of compossible rights, there would be no occasion for one right to ‘beat’ another.

⁴⁴ Dworkin, , Taking Rights Seriously, pp. 191–2.Google Scholar

⁴⁵ Steiner is willing to bite the bullet and accept that we ought to respect rights even when our doing so will result in moral catastrophes (pp. 199–201). However, he appeals to the implications of the choice theory to ameliorate that harsh consequence. If there are circumstances in which continued respect for my rights will produce a moral catastrophe, I myself have the power to waive the problematic rights and so forestall the catastrophe. If, however, I insist that others continue to respect my rights, that is what they must do – whatever the consequences.

⁴⁶ Steiner, , Essay on Rights, p. 258.Google Scholar

⁴⁷ The conceptual character of Steiner's assault upon the right of bequest implies that a legal right of bequest is no less problematic than a natural right. Steiner argues that legal systems are able to create legal rights of bequest only by using ‘legal fictions’ (Essay on Rights, pp. 256–8).Google Scholar

⁴⁸ Steiner concedes that there might be non-rights-based arguments for establishing bequest as a legal practice, for example, utilitarian arguments which defend bequest in terms of its beneficial incentive effects; but his own position leaves no significant moral space for such arguments to have effect. The estate of a deceased person belongs to the category of unowned things; every person has a natural right to an equal portion of unowned things; consequently, any attempt to establish bequest as a legal practice would violate that natural right and any justification offered for bequest would be trumped by it. Legal rules providing for bequest could be legitimate only in the unlikely event that everyone waived their natural right to an equal portion of the estates of the dead.

⁴⁹ Just who owns a child would seem a potentially complicated matter for Steiner since he includes in the labour that produces the child not merely the procreative efforts of its natural parents but also the labour that goes into raising the child (Essay on Rights, p. 246)Google Scholar. In addition, in so far as a portion of the child consists of purely natural resources, everyone has an equal claim to those (p. 248, n. 25).

⁵⁰ Steiner qualifies his statement that, during its zygotic, foetal and minority phases, the child is at its parents' ‘disposal’. The child consists, in part, of purely natural resources and everyone has an equal claim to those resources (Essay on Rights, p. 248, n. 25)Google Scholar. Thus, the child's parents could be denied the right to kill the child because, in so doing, they would be violating other people's equal claim to the child's genetic endowment. Yet, in dealing with other natural resources, Steiner is quite happy to see them divided up amongst people so that each person acquires an exclusive right over a particular bundle of natural resources, including the right to destroy some or all of that bundle. People can also acquire, through transactions with others, exclusive non-original rights over natural resources. In so far as children are natural resources, nothing in Steiner's argument seems to prevent their being dealt with in exactly the same way. But, even if we could safeguard the child by appealing to everyone's equal right to the child as a natural resource, that would still seem the wrong argument. Surely killing a child is wrong primarily because it wrongs the child and only secondarily, if at all, because it deprives others of something.

⁵¹ Steiner, , Essay on Rights, pp. 202–3, 223, 282.Google Scholar

⁵² Steiner, , Essay on Rights, pp. 81, 101.Google Scholar

⁵³ Steiner, , Essay on Rights p. 262.Google Scholar

⁵⁴ Steiner, , Essay on Rights p. 39.Google Scholar

⁵⁵ Rawls, Political Liberalism, p. 291.Google Scholar

⁵⁶ Rawls, , Political Liberalism, p. 298.Google Scholar

⁵⁷ Rawls, , Political Liberalism, pp. 325–31, 356–63Google Scholar; see also Rawls, , Theory of Justice, pp. 224–7, 277–9.Google Scholar

⁵⁸ One significant substantive difference is that, for Rawls, the basic liberties are inalienable (Rawls, , Political Liberalism, pp. 365–7)Google Scholar, while it would seem that, for Steiner, nothing is inalienable.

⁵⁹ Rawls, , Political Liberalism, pp. 265–9, 281–5.Google Scholar

⁶⁰ Rawls, , Political Liberalism, pp. 269–70.Google Scholar

⁶¹ Rawls, , Political Liberalism, pp. 268, 282–4.Google Scholar

⁶² Rawls, , Political Liberalism p. 271Google Scholar

⁶³ A further important consideration is Rawls's remark that ‘the first principle covering the equal basic rights and liberties may easily be preceded by a lexically prior principle requiring that citizens’ basic needs be met, at least insofar as their being met is necessary for citizens to understand and to be able fruitfully to exercise those rights and liberties. Certainly any such principle must be assumed in applying the first principle.’ (See Rawls, , Political Liberalism, p. 7Google Scholar; see also pp. 166, 228–30, and ‘Law of Peoples’, pp. 56, 223 n. 15, 225 n. 26)Google Scholar. However, he does not elaborate on how significantly that concern for basic needs is likely to affect the shape of his just society.

⁶⁴ In addition to working throughout Political Liberalism with a conception of states as separate and closed communities, Rawls makes clear in his ‘Law of Peoples’ that he believes that the norms governing international life should be thought of as norms governing relations between peoples rather than individuals.

⁶⁵ Rawls, , ‘Law of Peoples’, pp. 56, 75–7, 225, 228–30.Google Scholar

⁶⁶ Strictly, Steiner's parcelling out of deceased estates in equal measure to all living adults should be described as ‘distributive’ rather than ‘redistributive’ since, rather than adjusting individuals' existing holdings, it takes resources that have become unowned and adds them to each individual's ‘initial’ holdings.