I am grateful to Horace W. Brock, Robert Goodin, Thomas Grassey, John Rawls, Robert Simon and the editors for comments on an earlier draft, which was prepared at the Center for Advanced Study in the Behavioral Sciences, Stanford.

1 Boston: Beacon Press, 1968.

2 Boston: Beacon Press, 1969.

3 References in this form are to page numbers of Understanding Rawls.

4 Marx, Karl Critique of the Gotha Programme, Marx-Engels Selected Works,vol. 2 (Moscow: Foreign languages Publishing House, 1962), p. 23.Google Scholar The continuation of Wolff's quotation may be helpful in explaining Marx's terms: ‘The capitalist mode of production, for example, rests on the fact that the material conditions of production are in the hands of non-workers in the form of property in capital and land, while the masses are only owners of the personal condition of production, of labour power. If the elements of production are so distributed, then the present-day distribution of the means of consumption results automatically.’

5 Wolff, The Ideal of the University, p. 110.

6 F. Engels, Socialism: Utopian and Scientific, in Marx-Engels, Selected Works, vol. 2 pp. 127-28.

7 Ibid., p. 136.

8 References in this form are to page number of Rawls, John A Theory of justice (Cambridge, Mass.: Harvard University Press, 1971).Google Scholar

9 Marx-Engels, Selected Works, vol. 2 p. 25.

10 The Poverty of Liberalism, p. 109.

11 Ibid., p. 111.

12 Ibid., p. 112.

13 London: Allen and Unwin, 1964.

14 New York: Basic Books, 1974, p. 198.

15 The Ideal of the University, p. 110.

16 Marx-Engels, Selected Works, vol.2, p. 24. This is, according to Marx, one of the conditions for society being able to inscribe on its banners ‘From each according to his ability, to each according to his needs!’

17 See Lindbeck, Assar The Political Economy of the New Left: An Outsider's View (New York: Harper & Row, 1971)Google Scholar for remarks on the inconsistency of being simultaneously against bureaucratic control and Problems as methods of coordination. In practice, no modern command economy relies entirely on sticks rather than carrots. ‘The patterns and problems of wage mechanisms [in China] do not depart radically from those in capitalist economies or those in the Soviet Union and other Communist countries.’Hoffman, Charles Work Incentive Practices and Policies in the People's Republic of China, 1953-1965 (Albany: State University of New York Press, 1967), pp. 113-14.Google Scholar

18 Ethics, 85 (1974), p. 9.

19 Ibid., p. 8.

20 ‘Justice as Fairness’, Philosophical Review 62 (1958), pp. 164-94. Reprinted in Laslett, P. and Runciman, W. G. (eds.), Philosophy, Politics and Society,Second Series (Oxford: Basil Blackwell, 1962), pp. 132-57.Google Scholar ‘Distributive Justice’, in Laslett, P. and Runciman, W. G. (eds.), Philosophy, Politics and Society, Third Series (Oxford: Basil Blackwell, 1967), pp. 58–82.Google Scholar

21 Laslett and Runciman, Second Series, pp. 138-39.

22 ‘A Refutation of Rawls’ Theorem on justice', journal of Philosophy 63 (1966), pp. 179-90.

23 Laslett and Runciman, Second Series, p. 143n.

24 Levine, Andrew ‘Rawls' Kantianism’, Social Theory and Practice 3 (1974), p. 57.CrossRefGoogle Scholar

25 Here is an exquisite example: ‘According to Rawls, the decision the individual makes in choosing the principle of justice is rational insofar as it serves his long-range best interests.’ Johnson, Oliver A. ‘The Kantian Interpretation’, Ethics 85 (1974), p. 64.CrossRefGoogle Scholar

26 Wolff, Robert Paul The Autonomy of Reason: A Commentary on Kant's Groundwork of the Metaphysic of Morals (New York: Harper Torchbooks, 1973), p. 226.Google Scholar

27 Wolff, ‘There's Nobody Here But Us Persons’, Philosophical Forum 5 (1975), p. 128.Google Scholar

28 Hart, H.L.A. ‘Rawls on liberty and Its Priority’, University of Chicago Law Review 40 (1973), pp. 534-55.CrossRefGoogle Scholar Reprinted in Daniels, Norman (ed.), Reading Rawls (New York: Basic Books, 1976), pp. 230-52.Google Scholar

29 ‘Insofar as our enterprises are inherently social, the public-private, interference-noninterference model of human relationships breaks down. The central problem ceases to be the regulation of each person's infringement on the sphere of other persons' actions, and becomes instead the coordination of the several actions and the choice of collective goals.’ The Poverty of Liberalism, p. 50.

30 Dahl, R. A. A Preface to Democratic Theory (Chicago: University of Chicago Press, 1956).Google Scholar

31 Bachrach, Peter The Theory of Democratic Elitism (Boston: Little Brown, 1967).Google Scholar

32 The Poverty of Liberalism, pp. 191-92.

33 Ibid., p. 194.

34 See my Liberal Theory of justice (Oxford: Clarendon Press, 1973), chapter 9; and Sartorius, Rolf E. Individual Conduct and Social Norms: A Utilitarian Account of Social Union and the Rule of Law (Encino, Calif.: Dickenson Publishing Company, 1975), pp. 120-29.Google Scholar However, Wolff misses our point (Barry, p. 103; Sartorius, p. 126) that if everyone really cares much more about getting to some threshold level than exceeding it (a piece of information Rawls imputes to everyone in the original position) a utilitarian calculation would decree that everybody should be got to that threshold as the first priority.

35 It is important to keep clear where the notion of a lottery makes sense and where it does not. We can perfectly well say that the people in the original position would agree that some things should be decided by lottery. Thus, to take David Lyons's example, doses of a vaccine in an epidemic might be distributed by lot if there are not enough to go around. ('Nature and Soundness of the Contract and Coherence Arguments', Reading Rawls, p. 155.) What does not make sense is the notion that the principle of distribution should be chosen by lot.

36 See Goodin, Robert E. The Politics of Rational Man (London: Wiley, 1976),Google Scholar chapter 4, for a useful discussion of this question. Gauthier, David ‘Reason and Maximization’, Canadian journal of Philosophy 4 (1974-5), pp. 411-33,CrossRefGoogle Scholar is also relevant. (See especially p. 428.)

37 Laslett and Runciman, Second Series, p.148. It must be confessed, however, that Rawls himself is not always as careful as he might be to keep the distinction clear. R.M. Hare, in ‘Rawls' Theory of Justice’ (in Reading Rawls) lists a number of problematic passages in A Theory of justice (p. 87, fn. 7). I believe that they can be sorted out on the lines given.

38 Luce, R.D. and Raiffa, H. Games and Decisions (New York: Wiley, 1957), p. 127.Google Scholar

39 Ibid., p. 127; and Harsanyi, John C. ‘On the Rationality Postulates Underlying the Theory of Cooperative Games’, The journal of Conflict Resolution 5 (1961 ), pp. 183-84.CrossRefGoogle Scholar

40 Indeed, rational maximizers will agree on a unique Pareto-optimal outcome even if it is not symmetric. Suppose that the payoffs from the three principles (this would have to be in real life rather than the original position) were (1, 1), (10, 2) and (1, 1 ). Principle II is as much the outcome for rational players as it would be if the payoffs from it were (2, 2) or (10, 10).

41 I think the reason why Wolff fails to see the relevance of his concession is that he makes it for a case where only one principle yields a Pareto-optimal outcome in relation to the ‘state of nature’. But if he agrees that a unique Pareto-optimal outcome will be chosen in the case where only one principle makes everybody better off than the ‘state of nature’, it is inconsistent to deny that a unique Pareto-optimal outcome will be chosen when more than one principle makes everybody better off than in the ‘state of nature’.

42 For the first see Beitz, Charles R. ‘Justice and International Relations’, Philosophy and Public Affairs 4 (1975), pp. 360-89;Google Scholar and pp. 452-58 of Amdur, Robert ‘Rawls’ Theory of Justice: Domestic and International Perspectives’, World Politics 29 (1977), pp. 438-61.CrossRefGoogle Scholar For the second, see Kavka, Gregory ‘Rawls on Average and Total Utility',Philosophical Studies 27 (1975), pp. 237-53;CrossRefGoogle Scholar and my reply in Philosophical Studies 31 (1977), pp. 317-25. Also see Hubin, D. Clayton ‘Justice and Future Generations', Philosophy and Public Affairs 6 (1976), pp. 70-83.Google Scholar

43 Strasnick, Steven ‘The Problem of Social Choice: Arrow to Rawls’,Philosophy and Public Affairs 5 (1976), pp. 241-73;Google Scholar and ‘Social Choice and the Derivation of Rawls’ Difference Principle’ journal of Philosophy 73 (1976), pp. 85-99. For comments see Goldman, Alan H.CrossRefGoogle Scholar, ‘Rawls's Original Position and the Difference Principle’ and Wolff, Robert Paul ‘On Strasnick's “Derivation” of Rawls's “Difference Principle” ‘, The journal of Philosophy 73 (1976), pp. 845-49CrossRefGoogle Scholar and 849-58 respectively. For a general discussion of Rawls in a ‘social choice’ framework, see Sen, A. K. ‘Welfare Inequalities and Rawlsian Axiomatics’, Theory and Decision 7 (1976), pp. 243CrossRefGoogle Scholar-62.

44 See Gauthier, David ‘Rational Cooperation’, Nous 8 (1974), pp. 53-65;CrossRefGoogle Scholar ‘Justice and Natural Endowment: Toward a Critique of Rawls’ Ideological Framework’, Social Theory and Practice, 3 (1974), pp. 4-26; and ‘The Social Contract as Ideology’, Philosophy and Public Affairs 6 (1977), pp. 130-64. A variant is Ronald Dworkin's contention in ‘The Original Position’ (Daniels, ed., Reading Rawls, pp. 16-53), that the distinctive feature of the contract notion — everyone having a veto — is necessary only for deriving a theory that makes rights fundamental as against one that makes duties or societal goals fundamental. Therefore, if ‘a contract would produce the result that some fundamental goal dictates [like utilitarianism] … the contract is supererogatory, because the final result is determined by the fundamental goal and the contract device adds nothing’ (pp. 43-44). This seems to me a weak argument. Why should not the case for utilitarianism be strengthened if it can be shown that it would be chosen in the original position? Moreover, it is questionable whether Dworkin's argument, if correct, does what is intended and shows that the contract framework is appropriate for Rawls's theory of justice. For on Dworkin's definition the difference principle is a goal-based and not a right-based notion.

45 Rawls, makes it more explicit that this is a fundamental building-block in the theory in a paper subsequent to A Theory of justice: ‘fairness to Goodness’, Philosophical Review 84 (1975), pp. 536-54.CrossRefGoogle Scholar

46 Honderich, Ted Three Essays on Political Violence (Oxford: Basil Blackwell, 1976), p. 41.Google Scholar

47 Cf. Rawls, ‘Fairness to Goodness’, p. 552.

48 This theme is elaborated in ‘Don't Shoot the Trumpeter— He's Doing His Best: Reflections on a Problem of Fair Division’, Theory and Decision, forthcoming.