1 Under the prevailing global regime, millions die from government violence every decade. Tens of millions die annually from easily curable diseases and malnutrition, to which they are exposed simply on account of poverty. And hundreds of millions live permanently on the edge of such death: politically impotent, voiceless, malnourished, and without reserves for the event of even a minor natural or social misfortune.

International inequalities are vast, and steadily increasing, as the poorest countries tend to have the lowest (often negative) rates of economic growth. The percapita GNP of the developed West is now some thirty times that of the Third World, and inequalities between the richest and the poorest countries, regions, groups, and persons are significantly greater still.

2 These two pages are offered as part of a discussion of conscientious refusal. It would thus be unfair to treat them as a considered proposal for how his first approximation should be adjusted so as to take account of global interdependence.

3 The parties are said to use the maximin rule, by which alternatives are ranked according to their worst possible outcomes (see TJ, 155ff). Since the parties focus here on representative groups (TJ, 64), institutional scheme A is preferred to B if and only if the lowest representative share under A is larger than the lowest representative share under B.

4 Here I am keeping fixed Rawls’s arguments to the effect that his principles are the maximin solution (cf. TJ, 152f and elsewhere).

5 KCMT, 550. Compare also BSS, 58, and especially Rawls, John ‘A Kantian Conception of Equality’ (Cambridge Review (1975), 94-9)Google Scholar, 98f.

6 Provided such institutional protection is feasible (TJ, 151f, 542f). Otherwise the lexical ordering of the two principles is to be suspended in favor of the general conception (stated at TJ, 303). In my argument, I neglect this complication, because Rawls asserts the special conception (under which the basic liberties have priority) as the long-range demand of justice (TJ, 152, 542). Moreover, even if the general conception did apply to today’s world (R₁) or to some of its poorer societies (R₂) this would still have very radical implications for the moral assessment of the current global basic structure - though showing this would require a somewhat different argument from the one in the text.

7 Rawls relies here on the seminal but somewhat dated account by Brierly, James L. The Law of Nations (Oxford: Clarendon Press 1963Google Scholar [1928]).

8 Rawls, John: ‘Fairness to Goodness’ (Philosophical Review) 84 [1975), 536-54)Google Scholar, 545

9 That Rawls invited us to picture the parties as heads of families (TJ, 128) does not undermine this commitment. For purposes of an account of justice between generations - abandoned since (BSS, 70 n. 11) - he merely made a motivational assumption about the individual persons represented by the parties.

10 This constraint is still compatible with the claim that the most fundamental right of persons is the right to live in a state that has the kind of state rights accorded by international law. If this were true, then the present order might indeed be the best we can hope for: At least the foremost right of persons is rather well fulfilled. Given what Rawls has said about the basic liberties, this piece of conservative ingenuity will not fit into his conception of justice. But it is popular with others, most notably Walzer, Michael Just and Unjust Wars (New York: Basic Books 1977)Google Scholar, esp. 53f; and ‘The Moral Standing of States,’ in Philosophy and Public Affairs 9 (1980), 209-29.

11 Here we should think not merely of individual transactions, like a treaty establishing a foreign military base or the agreement between a Western tourist and a Bangkok prostitute. There are more complex cases as when, in some poor country, local demand for grain and beans competes with foreign demand for cotton and coffee. Even if the coffee consumers are no more affluent than the relevant landowners and never have any dealings with the (much poorer) would-be consumers of grain, grossly unequal bargaining power may yet play a pernicious role: By displacing local food production, foreign demand for coffee tends to raise food prices, which in tum reduces the ability of the poorest locals to transform their need for food into effective market demand. On such scenarios for declining ‘exchange entitlements,’ cf. Sen, A.K. Poverty and Famines (New York: Oxford University Press 1981)Google Scholar.

12 Kant, Immanuel ‘Idea for a Universal History from a Cosmopolitan Point of View,’ in Kant on History, Beck, L.W. ed. (New York: Bobbs-Merrill 1963), 24Google Scholar

13 Rawls is sometimes aware of this, as I showed from TJ, 380.

14 Such a demonstration would be akin to the ‘slaveholder’s argument’ in Rawls (TJ, 167f). While it would justify the existing world order, it would not show who should be entitled to occupy the advantaged positions within that order.

15 See Beitz, 165, for the idea of minimal interaction (trading apples and pears). Peter Danielson suggests a theory of global justice as rational cooperation; see ‘Theories, Intuition, and the Problem of World-Wide Distributive Justice,’ in Philosophy and the Social Sciences 3 (1973), 331-40.

16 Barry raises this objection against Beitz: ‘I do not think that [a global difference principle] can plausibly be said to be advantageous to rich as well as poor countries’ (Barry, Brian ‘Humanity and Justice in Global Perspective,’ in Pennock, and Chapman, eds., Ethics, Economics, and the Law [New York: New York University Press 1982], 232)Google Scholar. He can be construed as asserting that the transition to a just scheme must benefit even those now unjustly advantaged. This is the construal to which David Richards responds (’International Distributive Justice,’ ibid., 277f):

when John Stuart Mill ... criticiz[ed] injustices to women, he conceded that men, as a class, would suffer some losses when they surrendered their unjust domination, just as slaveowners did when slavery was ended; Mill’s argument is quite clear that the gain is not one of actual reciprocal advantage to men (indeed, they lose), but the gain in justice when men regulate their conduct by principles they would reasonably accept if they were women on the receiving end.

But Barry may also be interpreted as claiming that the new institutional scheme, once in place, must be a mutually beneficial one. He could then reply to Richards that, whereas men are better off with sex equality than if they did not interact with women at all, the U.S. (or the developed West) would be worse off in a Rawlsian world order than in splendid isolation. Barry’s objection, so construed, arises not only in the global context. It can equally be raised within a state - on behalf of a wealthy province, for example, or by a group of such provinces seeking to exclude a poor one. Quite apart from the issue of globalization, it is then crucial for Rawls’s theory that this objection be met, as I hope to have done in the text.

17 cf. Rawls’s distinction between ‘perfectly just’ and ‘just throughout’ (TJ, 79).

18 As argued in Beitz 136-42. Beitz suggests a global resource redistribution principle, which he thinks should apply even in the absence of any international cooperation.

19 Rawls, John ‘Reply to Alexander and Musgrave,’ in Quarterly Journal of Economics 88 (1974), 634Google Scholar. Twelve conditions for well-orderedness are listed ibid., 634-6.

20 And if it meant this, then Rawls’s principles would be entirely irrelevant, because most of the conditions for well-orderedness are not satisfied by any existing national society either.

21 The later Rawls, that is, who is withdrawing from the topic of global justice, and no longer believes his conception of justice to be appropriate to all national societies. The spatio-temporal domain that he takes his conception to be applicable to seems to have undergone steady shrinkage toward the central instance (the U.S. between 1960 and 1979). Those uneasy with this shrinkage have another reason for following my interpretation: G does not sidestep cultural divergence, but accommodates it.

22 cf. JFPM, 246-51, and Rawls, John ‘The Idea of an Overlapping Consensus,’ in Oxford Journal of Legal Studies 7 (1987), 1-25CrossRefGoogle Scholar.

23 Cf. Beitz, Charles ‘Cosmopolitan Ideals and National Sentiment,’ in Journal of Philosophy 80 (1983) 591-600CrossRefGoogle Scholar, esp. 596.

24 There is another important reason for believing that an international discussion of the topic of global justice can be a catalyst for moral progress: Many persons in the West acquiesce in the foreign policy of their government not because they believe that- thanks to cultural diversity, perhaps- it is justified, but because they have no settled moral beliefs about this matter one way or the other. However, in view of the enormity of prevailing inequalities and deprivations, this is an issue that one ought to try to address. if anything academics might do can really make a substantial moral difference, then a discourse about global justice might well be it.

25 Rawls did allow for some variation in national constitutions, which are determined through yet another thought-experiment - a hypothetical ‘constitutional convention’ featuring parties behind a thinner veil of ignorance (TJ, §31).

26 Of course, such opportunities for permitting greater diversity of forms of life exist on the national level as well - and Rawls’s first principle would then seem to favor a federalist constitution. Curiously, Rawls never addresses the important issue of an institutional scheme’s degree of political centralization. Is this left to the parties in the constitutional convention?