2 Unless otherwise indicated, I shall employ the terms ‘revisionist liberalism,’ ‘the liberal theory of justice,’ or simply ‘liberalism’ interchangeably.

3 I would therefore distinguish egalitarian revisionist liberals, who give absolute priority to individual rights, and other proiessedly liberal egalitarians who self-consciously distance themselves from an exclusive focus on individual rights. For an example of the latter, see Ackerman, Bruce Social justice and the Liberal State (New Haven, CT: Yale University Press 1980)171-2Google Scholar.

4 Dworkin, Ronald ‘Liberalism,’ reprinted in A Matter of Principle (Cambridge. MA: Harvard University Press 1985), 181-204Google Scholar, at 192. Fried advocates the redistribution of social resources in the direction of greater equality, but neglects to specify the precise distributive principle which he favors. See his Right and Wrong (Cambridge, MA: Harvard University Press 1978), 119. Nagel defends a similar egalitarian principle of distribution. See his ‘Equality,’ in Mortal Questions (Cambridge: Cambridge University Press 1979).

5 Rawls, John A Theory Of justice (Cambridge, MA: Harvard University Press 1971), 505-6 (my emphasis)Google Scholar

6 Dworkin, Ronald Taking Rights Seriously (London: Duckworth 1977), 182Google Scholar (my emphasis)

7 For roughly similar views on the minimal prerequisites for being considered an equal from the standpoint of justice, compare Nagel, The View From Nowhere (Oxford: Oxford University Press 1985), 162, 205Google Scholar; and Fried, 118, 124.

8 Nozick, Robert Anarchy, State And Utopia (New York: Basic Books 1974), 49-50Google Scholar

9 It is sometimes suggested that this commitment to contractualism constitutes the most distinctive characteristic of liberal doctrine. See, for instance, Waldron, Jeremy ‘Theoretical Foundations Of Liberalism,’ The Philosophical Quarterly 37 (1986) 127-50CrossRefGoogle Scholar, esp. 128, 143, 149. A clear statement of the contractual condition on theories of justice is presented by Scanlon, T.M. ‘Contractualism and Utilitarianism,’ in Sen, A. and Williams, B. eds., Utilitarianism and Beyond (Cambridge: Cambridge University Press 1982) 103-28CrossRefGoogle Scholar.

10 For a critique of Nozick along these lines, see Varian, Hal ‘Distributive justice, Welfare Economics, And The Theory Of Fairness,’ reprinted in Hahn, F. and Hollis, M. eds., Philosophy And Economic Theory (Oxford: Oxford University Press 1979) 134-54Google Scholar, at 145. Varian argues that, while Nozick’s focus on the historical origins of an entitlement is valuable because it enhances people’s freedom to choose how to deploy their resources, this emphasis must be balanced by a principle which establishes an initial equality among people’s material endowments prior to their transactions (148).

11 Buchanan’s version of the social contract treats the existing distribution of society’s resources as a baseline, implying that each person’s initial holdings in a state of nature constitute a starting point for their contractual relationship. They then bargain over a rate of taxation. The result approximates the minimal state, with certain limited expenditures on public goods. See Buchanan, James and Brennan, Geoffrey The Reason Of Rules (Cambridge: Cambridge University Press 1985), 22Google Scholar. By contrast, Rawls’s egalitarian theory does not recognize ownership claims which are prior to the original position. Gauthier’s standpoint is similar to Buchanan’s, but he adds the provision that the distribution of resources, prior to the social contract, must have been acquired in conformity with individual rights. See his Morals By Agreemmt (Oxford: Clarendon Press 1986), 201,220. Harman argues that the social contract rightly reflects differences in bargaining power. (Harman attaches a peculiar meaning to the notion of right, in virtue of his defense of moral relativism.) He therefore defends inegalitarian restrictions on positive rights or claims for the redistribution of wealth. See his ‘Moral Relativism Defended,’ The Philosophical Review 84 (1975) 3-22, at 13. Among recent contractualists, only Brandt argues in favor of an egalitarian distribution with the aim of promoting each person’s welfare (defending a version of utilitarianism). But even he is uncertain that this outcome would result from the contractual situation which he defines. See his A Theory Of The Good And The Right (Oxford: Clarendon Press 1979), 200, 206.

12 Theory of Justice, 100

13 ‘What Is Equality? Part 2: Equality Of Resources,’ Philosophy and Pubic Affairs 10 (1981) 283-345, at 314

14 Ibid., 312-13

15 For Fried’s endorsement of the principle of redress, see 116, 135, 144. Fried, however, neglects to furnish guidelines concerning the limits of personal responsibility. Nagel invokes similar considerations in arguing for the redistribution of resources. See his ‘Poverty And Food: Why Charity Is Not Enough; in Brown, Peter G. and Shue, Henry eds., Food Policy (New York: The Free Press 1977) 54-64, at 57Google Scholar.

16 The idea that persons legitimately show favor towards the members of their society or that preferential principles of distribution are reserved for society is affirmed by Dworkin, ‘Liberalism,’ 192; Nagel (with some ambiguity), ‘Poverty And Food: Why Charity Is Not Enough,’ 55, 61; Rawls, ‘Justice As Fairness: Political Not Metaphysical,’ Philosophy And Public Affairs 14 (1985), 10Google Scholar; and Fried, 119, 131.

17 Fried, 118 (my emphasis)

18 Ibid., 130-1 (my emphasis)

19 This appears in Dworkin’s Taking Rights Seriously, 182. But Dworkin signals that he is still prepared to endorse the universalist grounds for rights in ‘What Is Equality?’ 345.

20 ‘Why Should Liberals Care About Equality?’ in A Matter of Principle 205-13, at 206 (my emphasis)

21 Ibid., 208. For his part, despite ample assurances that ‘the capacity for a sense of Justice is possessed by the overwhelming majority of mankind’ (A Theory of Justice, 505) Rawls continues to insist that relations of justice are only appropriate among the members of existing societies (’Justice as Fairness,’ 10). Here, once again, the discussion of justice begins with human beings and their general capacities, only to conclude with the ‘community’ and its ‘citizens.’ Several authors have pointed out that Rawls is inconsistent in limiting his theory of justice to the members of a single society, as opposed to humanity at large. In particular, I have benefited from the excellent treatment this issue receives from Richards, David ‘International Distributive Justice,’ in Roland Pennock, J. and Chapman, J.W. eds., Law, Economics, Society (New York: New York University Press 1982) 219-52Google Scholar, and Beitz, Charles Political Theory and International Relations (Princeton, NJ: Princeton University Press 1979), part IIIGoogle Scholar.

22 Here the remarks of an anonymous referee are entirely apposite, and so I shall reproduce them: ‘Further thought suggests that the critique may fit any account of justice, egalitarian or not, so long as it requires redistribution and the moral equality of persons, but restricts redistribution to a society. For example, a desert theory may rest on moral equality — all people have equal claims to develop merit with distribution then resting on merit — and so this theory will have similar problems if it restricts the operation to the principle within one society, overlooking the desert claims of those outside society who are as meritorious as those within it … the argument presents difficulty for all kinds of theories of justice.' It seems to me that libertarian writers have been quickest to grasp this point. For this and other reasons they have been at the forefront of a movement to eliminate the state as an entity restricting the allocation of resources among humanity at large.

23 The third condition which accounts for equality, the liberal commitment to contractualism, has been employed in the past to legitimate parochial schemes of distribution. Yet it seems neutral in the dispute between cosmopolitans and parochialists, since, in principle, a social contract might include humanity as a whole.

24 Beitz, Political Theory, 181f.

25 Beitz originally argued that only natural resources ought to be redistributed along egalitarian lines among the members of humanity. The global difference principle would only take effect if it could be shown that the world’s nations participate in appropriately reciprocal exchanges: a condition that Beitz thought had been met (Ibid., 143, 150). Later Beitz realized that making the global difference principle depend on the existence of these cooperative relationships or reciprocal ties was inconsistent with the aim of compensating persons for morally arbitrary factors. See his ‘Cosmopolitan Ideals and National Sentiment,’ The Journal of Philosophy 80 (1983) 591-600, at 595ff.

26 It seems to me that this consequence can be deduced to follow from Beitz’s views. Beitz remarks that, ‘assuming that Rawls’s arguments for the two principles are successful, there is no reason to think that the content of the principles would change as a result of enlarging the scope of the original position so that the principles would apply to the world as a whole’ (Political Theory, 151). Beitz also signals that the principles are concerned with the interests of individuals and not communities: ‘It seems obvious that an international difference principle applies to persons in the sense that it is the globally least advantaged representative person (or group of persons) whose position is to be maximized’ (152). Furthermore, he notes that ‘because the difference principle applies in the first instance to persons, it would also require intrastate inequalities to be minimized if necessary to maximize the position of the (globally) least advantaged group’ (153). Thus, the relevant principles of justice do not alter depending on whether we are referring to a single society or the globe as a whole. The principle simply addresses individual claims for material goods. It is not concerned with the claims of separate states and collectives. All that is then required is the plausible empirical assumption that international population transfers can efficiently enhance the material prospects of the world’s least advantaged members. In this vein, several authors have argued that Rawls’s framework supports the idea of open international borders. For useful discussions, see King, Timothy ‘Immigration from Developing Countries: Some Philosophical Issues,’ Ethics 93 (1983) 525-36CrossRefGoogle Scholar; Hudson, James L. ‘The Philosophy of Immigration,’ The Journal of Libertarian Studies 8 (1986) 51-62Google Scholar; Carens, Joseph H. ‘Aliens and Citizens: The Case for Open Borders,’ The Review of Politics 49 (1987) 251-73CrossRefGoogle Scholar.

27 The effect on the provision of these public goods is likely to be even more dramatic if the state proved responsive to the literal recommendations of public finance economists and attempted to apply a unanimity condition on non-redistributive to public expenditures.

28 For comprehensive treatments of the problems of collective action, see Hardin, Russell Collective Action (Baltimore: John Hopkins University Press 1982), ch. 3-5Google Scholar; and Comes, Richard and Sandler, Todd The Theory of Externalities, Public Goods, and Club Goods (Cambridge: Cambridge University Press 1986), parts 1 and 3Google Scholar.

29 The reciprocity thesis features prominently in Rawls’s discussion of justice. See his ‘Justice as Fairness,’ 10. But similar ideas are invoked by Fried (119, 131) and Dworkin, ‘Why Should Liberals Care About Equality?’ (211) Recently, Brian Barry has endorsed a similar view, reversing his earlier advocacy of a cosmopolitan theory of distributive shares. See his ‘Humanity and Justice In Global Perspective,’ in Roland Pennock, J. and Chapman, J.W. eds., Law, Economics, Society (New York: New York University Press 1982) 219-52Google Scholar, at 232, as compared with The Liberal Theory Of Justice (Oxford: Clarendon Press 1973), 133.

30 Though one should not underestimate the difficulties. Beitz maintains that under present circumstances, where the economies of nations remain interdependent, reciprocal exchanges are international in character (Political Theory, 152; ‘Cosmopolitan Ideals,’ 596). Even if this claim is false, however, Beitz notes that the range of reciprocal relationships are not confined to any single, ongoing society. Thus, reciprocal exchanges are too broad to underwrite the liberal focus on existing societies. Somewhat differently, Brian Barry has pointed out that if cooperation or reciprocity is understood strictly in terms of rational advantage, the condition is not met within existing societies. Rational maximizers would, according to Barry, be better off expelling the least well-off and foregoing their marginal social product. See his Theories Of Justice, Vol. I (Hertfordshire: Harvester-Wheatsheaf 1989), 243. Hence, this concept of reciprocity would be too narrow to underwrite the revisionist focus on existing societies. My own skeptical view is that any attempt to adduce objective criteria, or necessary and sufficient conditions for the purpose of identifying societies and prospective societies is likely to be unpersuasive. In an adequate global theory of justice, the unit of society should be defined subjectively according to people’s level of attachment to the political units of their choosing.

31 Someone could argue that these reciprocal exchanges are irrelevant from a moral point of view, since historical relations of reciprocity reflect morally arbitrary factors. This point has been made by David Richards (289-90), and more recently embraced by Beitz in ‘Cosmopolitan Ideals.’

32 Rawls writes that: ‘One feature of justice as fairness is to think of the parties in the initial situation as rational and mutually disinterested …. In the absence of strong and lasting benevolent impulses, a rational man would not accept a basic structure merely because it maximized the algebraic sum of advantages irrespective of its permanent effect on his own basic rights and interests. Thus it seems that the principle of utility is incompatible with the conception of social cooperation among equals for mutual advantage’ (Theory of Justice, 13-14; see also 176).

33 Ibid., 253-4

34 As he writes, ‘I suppose, then, that for the most part each person holds two relevant positions: that of equal citizenship and that defined by his place in the distribution of wealth …. This position [of equal citizenship] is defined by the rights and liberties required by the principle of equal liberty and the principle of fair equality of opportunity’ (Ibid., 96-7). Dworkin’s own commitment to an individualistic standard of well-being is well expressed in his ‘What is Equality?’ Here he actually takes Rawls to task (340f.) for being insufficiently individualistic, on the grounds that the difference principle assesses people’s well-being in relation to their membership in a group or collective (the least well-off class of persons). Nagel and Fried do not, to my knowledge, present theories of society or any indication of the relationships which they think are critical in associations appropriately described that way — apart from the vague notion of reciprocity. This raises problems for their respective accounts of justice, since, as we earlier noted, each author endorses the idea that priority is rightly reserved for the members of society in a scheme of distributive shares. It is not clear what the basis for that provision could possibly be from their accounts.

35 ‘Liberalism,’ reprinted in A Matter of Principle (Cambridge, MA: Harvard University Press 1985) 181-204, at 191 (my emphasis)

36 Ibid., 203

37 Not all collectivists are similarly agnostic about the good. Some urge that political argument be placed on a footing where people make claims to the effect that their life-style more exactly approximates a basic truth about humans, or possesses some other intrinsically valuable property. A realism about the good appears to be shared by Alasdair MacIntyre in his After Virtue (Notre Dame, IN: University of Notre Dame Press 1981 [2nd ed. 1984]) ch. 18; Michael Sandel in his Liberalism and the Limits of Justice (Cambridge: Cambridge University Press 1982), 175-83; and Joseph Raz, ‘Rights-Based Moralities,’ in Waldron, Jeremy ed., Theories of Rights (Oxford: Oxford University Press 1984) 182-200Google Scholar, at 188. In my view, none of these arguments makes adequate provision for the inevitable rejoinders tendered by agnostics and other skeptics.

38 ‘The Priority Of The Right, and Ideas of The Good,’ Philosophy And Public Affairs 17 (1988) 251-302, at 274

39 This assumes that everyone — including whatever persons immigrate to that society — will continue to embrace the principles of justice. This may be an optimistic belief, but objections to cosmopolitanism should dispense with an appeal to the possibility that various non-ideal circumstances will prevail. In the first case, such objections hold no weight if and when circumstances change, so that the members of humanity acquire a sense of justice. Second, it leaves intact the aim of promoting circumstances in which all persons come to share the appropriate sense of justice.

40 It may be worth pointing out that the collectivist account defended here differs from communitarian theories such as Michael Sandel’s. Sandel assumes that persons are under an obligation to their community as the result of a ‘social desert’ claim which that community enjoys against individual talents and endowments. That is to say, the members of the community are under a natural, contract-independent obligation to their particular political institutions. See his Liberalism And The Limits Of Justice, 101-3. I doubt whether any such natural political obligations are defensible. Thus, the collectivist account which I defend simply empowers persons to favor their community within the confines of a global theory of distributive shares. This is on condition that the members of a society come to identify their ends with a wider group: a contingent matter. It does not claim that they are under some special obligation to any particular community by nature. I consider the issues surrounding the relationship between particular communities and a universal theory of justice in a forthcoming book The Rights of Individuals, The Prerogatives of Communities.

41 For discussion of the following issues I am grateful to Tom Baldwin.

42 I therefore take issue with Michael Walzer’s highly original and stimulating discussion of these issues. Walzer makes the plausible claim that the members of a political community require control over their territorial boundaries in order to retain the character of their community. Consequently, he urges that political communities have natural rights to their territorial resources. Under this right Walzer implicitly lumps the following powers: (i) to occupy a surface area of the earth; (ii) to limit immigration to that area; (iii) to exploit the territory’s natural resources; and (iv) to enjoy the benefits that can be traced to the technological assets of the community. See his Spheres of Justice (Oxford: Basil Blackwell 1983), ch. 2. It seems to me that the aim of ensuring self-determining communities does not support the strong thesis that communities should enjoy their present levels of wealth. Where self-determination and the preservation of cultures is concerned, territorial integrity surely plays a far more indispensable role than simple material opulence. A global contractual theory of justice is therefore likely to distinguish between powers (i) and (ii), relating to territorial claims, and powers (iii) and (iv) pertaining to wealth.

43 I argue this at greater length in an essay ‘Liberalism And The Decline Of Culture,’ forthcoming in Ethics.

44 For a notable exception see Bruce Ackerman, Social Justice in the Liberal State, sect. 45.