Hostname: page-component-cd9895bd7-jn8rn Total loading time: 0 Render date: 2024-12-25T18:04:52.388Z Has data issue: false hasContentIssue false
  • English
  • Français

The Revisionist Difference Principle

Published online by Cambridge University Press:  01 January 2020

Andrew D. Williams
Andrew D. Williams*
Affiliation:
University of York Heslington, YorkY01 5DDUnited Kingdom
Get access Rights & Permissions [Opens in a new window]

Extract

John Rawls's famous difference principle is capable of at least four distinct statements, each of which occurs in A Theory of Justice. According to what I shall term the Crude Principle it is a necessary and sufficient condition for the justice of an institutional scheme which expands social and economic inequality that, subject to the satisfaction of more weighty principles, it increases the level of advantage of the least advantaged. Expressing this principle Rawls writes that,

Assuming the framework of institutions required by equal liberty and fair equality of opportunity, the higher expectations of those better situated are just if and only if they work as part of a scheme which improves the expectations of the least advantaged members of society. The intuitive idea is that the social order is not to establish and secure the more attractive prospects of those better off unless doing so is to the advantage of those less fortunate.

Type
Research Article
Information
Canadian Journal of Philosophy , Volume 25 , Issue 2 , June 1995 , pp. 257 - 281
Copyright
Copyright © The Authors 1995

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

1 Rawls, J. A Theory of justice, (Cambridge: Harvard University Press 1971), 75Google Scholar. See also p. 78 where he remarks that’ … inequality in expectation is permissible only if lowering it would make the working class even more worse off.’

2 A distribution is pareto efficient if there is no feasible pareto superior distribution, and is pareto-incomparable to another distribution if neither is pareto superior to the other. A distribution may be pareto superior in a weak and strong form. A distribution is weakly pareto superior to another distribution if at least one individual is better off, and no individual worse off, in the former than the latter. A distribution is strongly pareto superior to another if each individual is better off in the former than the latter. Note that whilst the Crude Principle prohibits only some weakly pareto superior distributions, egalitarian rules additionally prohibit some strongly pareto superior distributions.

3 However, for dissenting discussion of the first feature see Temkin, L. Inequality (New York: Oxford University Press 1993), ch. 9.Google Scholar

4 When levels are close-knit ‘it is impossible to raise or lower the expectations of any representative man without raising or lowering the expectations of every other representative man, especially that of the least advantaged.’ Under such circumstances, having secured an unequal distribution which maximally improves the expectations of the least advantaged, no pareto superior distribution is feasible. Its feasibility would require that, contrary to close-knittedness, the expectations of some other representative individual could be raised without affecting the expectations of the least advantaged. See Rawls, A Theory of Justice, 80.Google Scholar

5 Rawls, A Theory of Justice, 83Google Scholar. Elsewhere, however, Rawls employs subsidiary arguments which support the Crude rather than the Lexical Principle, since they treat benefit to the least advantaged as a necessary conditions of increases in inequality. For example, at 105, he claims that ‘A further merit of the difference principle is that it provides an interpretation of the principle of fraternity,’ which, he claims, corresponds to the’ … idea of not wanting to have greater advantages unless this is to the benefit of others who are less well off’ (105). For a further charge of inconsistency between the formulation of, and supplementary grounds for the difference principle, see Parfit, D. Reasons and Persons (Oxford: Oxford University Press 1984), 492-3.Google Scholar

6 Rawls, A Theory of Justice, 83Google Scholar

7 See Martin, Rawls and Rights (Lawrence: University Press of Kansas 1985), ch. 6Google Scholar, which is a longer version of sections 2-5 of a paper, ‘Two Interpretations Of The Difference Principle,’ co-authored with Prakash Shenoy.

8 Rawls explains that when ‘inequalities in expectations are chain-connected … if an advantage has the effect of raising the expectations of the lowest, it raises the expectations of all positions in between. For example, if the greater expectations of entrepreneurs benefit the unskilled worker they, they also benefit the semiskilled.' See Rawls, A Theory of Justice, 80.Google Scholar

For proof of the equivalence of the Simple and Revisionist Principles, given the assumption of chain connection, see the Appendix by P. Shenoy in Martin, Rawls and Rights. It can be grasped informally as follows. The Simple Principle requires that expansions in inequality maximally benefit the least advantaged. The Revisionist Principle requires expansions in inequality if they benefit at least one individual and disadvantage nobody. If close-knittedness and chain connection obtain, then expansions in inequality which maximally benefit the least advantaged also benefit at least one individual and disadvantage nobody for the following reasons. First, such expansions ex hypothesis maximally benefit the least advantaged. Second, since they increase the range they must also benefit the most advantaged. Third, since chain connection entails that if the range between the expectations of the more and less advantaged groups increases then the expectations of intermediate groups also increase, they must also benefit the intermediate groups.

9 Rawls, ‘A Reply to Alexander and Musgrave,’ Quarterly Journal of Economics 88 (1974), 648CrossRefGoogle Scholar

10 See especially Rawls, A Theory Of Justice, 80Google Scholar. More recently, in Political Liberalism (New York: Columbia University Press 1993), 17, Rawls expresses a similar thought with different vocabulary. There he remarks that’ … the two principles of justice with the difference principle, with its implicit reference to equal division as a benchmark, formulate an idea of reciprocity between citizens.’

11 Rawls, A Theory of Justice, 80 (italics added)Google Scholar

12 Martin, R. Rawls and Rights, 96Google Scholar. Compare Rawls, A Theory Of justice, 81, Figure 10.Google Scholar

13 Rawls, A Theory of Justice, 80Google Scholar

14 The distinction between partially and fully just societies corresponds to that which Rawls makes between partial compliance and strict compliance theory. The former concerns how to act assuming every one acts justly, whilst the latter deals with how to act in circumstances in which injustice already exists. See, for example, Rawls, A Theory Of Justice, 244-5.Google Scholar

15 We might also ask why if Rawls believes that the Simple and the Revisionist Principles are both principles of ideal justice, his rejection of the latter as a principle of non-ideal justice is so resolute.

16 Rawls, A Theory Of Justice, 82CrossRefGoogle Scholar

17 Rawls, A Theory Of Justice, 82CrossRefGoogle Scholar

18 See ‘The Basic Structure As Subject,’ 282, in Political Liberalism, which reproduces an earlier paper of the same title that appeared in Goldman, A.I. and Kim, J. eds., Values And Morals (Dordrecht: Reidel 1978).CrossRefGoogle Scholar For another statement of the argument see Rawls, ‘A Well-Ordered Society,’ in Laslett, P. and Fishkin, J. eds., Philosophy, Politics and Society, Fifth Series (New Haven: Yale University Press 1979) 15–16.Google Scholar

19 Political Liberalism, 282. Given that Rawls is presenting an informal, moral argument, it is puzzling why he envisages it being considered by the parties to the original position, who are elsewhere described as moved not by such arguments but by the interests of the individuals they represent.

20 For the most sustained critique of the second stage of Rawls's argument see Cohen's, G.A. discussion of the appeal to incentives as a justification of inequality in ‘Incentives, Inequality And Community,’ in Petersen, G.B. ed., The Tanner Lectures On Human Values, Volume Thirteen (Salt Lake City: University Of Utah Press 1992) 262–329.Google Scholar

21 Rawls, Political Liberalism, 282Google Scholar (italics added)

22 It is worth noting, however, that Rawls's intuitive argument is considerably less forceful on the latter than the former construal. The central feature of the difference principle is the priority which it gives to the interests of the least advantaged. The former construal does at least secure such priority, but only if chain connection obtains. The latter construal, however, assumes that, but does not explain why, priority should be attached to the interests of the least advantaged in selecting a particular pareto-improvement upon equality. It remains obscure why Rawls does not infer a universal veto from an egalitarian benchmark but instead claims ‘Because we start from equal shares those who benefit least have a veto.’

23 To some Rawls remarks that a society’ … should operate only on the upward rising part of the contribution curve .. .’ and that ‘to realize the ideal of the harmony of interests … and to meet the criterion of mutual benefit … we should stay in the upward region of positive contributions’ may appear as counter-evidence to this conclusion, since maximizing the prospects of the least advantaged where chain connection fails may require departing from the region of positive contribution. However, I read them differently, namely as observations about the desirable properties which the Simple Principle possesses if chain connection obtains rather than as necessary conditions on an adequate principle. Nevertheless Rawls should, I think, have conceded that in retaining his commitment to the least advantaged even where chain connection fails he does forego one of the merits which the such a commitment possesses under more propitious circumstances. See, respectively, Rawls, A Theory Of Justice, 104-5Google Scholar, and Sec. 13, for Rawls's remarks and for background explanation.

24 See A Theory of Justice, Sec. 26, ‘Some Reasons for the Maximin Criterion,’ American Economic Review 64 (197 4) 141-6, and ‘Reply to Alexander and Musgrave,’ Quarterly Journal of Economics 88 (1974) 633-55. I also draw upon Strasnick, S. ‘Review of Robert Paul Wolff's Understanding Rawls,’ Journal of Philosophy 76 (1979) 496–510.Google Scholar

25 The presence of (b) does not entail that of (c), since it is consistent with (b) that the maximin solution itself involves the possibility of unacceptable outcomes. For illustration see Kavka, G.S. Hobbesian Moral and Political Theory (Princeton: Princeton University Press 1986), 143.Google Scholar

26 Neither does this objection criticize Rawls in the same manner as utilitarians such as Arrow and Harsanyi, who deny that condition (a) is present in the Original Position. They controversially argue for a subjectivist conception of probability according to which, even if the parties lack objective estimates of probabilities, it is justifiable to arbitrarily assume that each possible outcome is equally probable and, thus, that uncertainty does not exist.

27 Since A Theory of Justice Rawls himself appears to recognize such a problem. Thus in his ‘Reply to Alexander and Musgrave’ he distinguishes between two pair-wise comparisons between his Two Principles and, respectively, (i) the Principle of Average Utility and (ii) a Mixed Conception, in which the First Principle is preserved, but the Second Principle is modified by the replacement of the Difference Principle with the Principle of Average Utility. He concludes that if the latter modification eliminates the grave outcomes associated with utilitarianism then his central argument may need to be supplemented. Thus Rawls remarks that ‘The two principles always secure acceptable conditions for all, while the utility criterion may not. This test is not so clear when the utility principle is substituted for maximin [i.e. the Difference Principle], as in the second pair-wise comparison; but I believe that once we introduce more refined considerations based on publicity and stability, and the like, this test still has force’ (652).

28 For other sympathetic treatments of the Principle see Barry, B. Theories of Justice (Hemel Hempstead: Harvester-Wheatsheaf 1989), 229-31CrossRefGoogle Scholar, Davidson, A. ‘Is Rawls a Kantian?’ Pacific Philosophical Quarterly 20 (1985), 71-4Google Scholar, and Rae, D. ‘A Principle Of Social Justice,’ in Laslett, P. and Fishkin, J. eds., Philosophy, Politics And Society, Fifth Series (New Haven: Yale University Press 1979).Google Scholar

29 Martin, Rawls and Rights, 97Google Scholar. Note that, as T. Pogge has pointed out, (b) becomes problematic if applied to distributions in which the same representative groups do not exist. Even if this is not possible when groups are defined as deciles or quintiles, it may be true when they are defined in other ways, such as those mentioned in Rawls, A Theory of Justice, 98.Google Scholar

30 Thus he writes that, ‘… where chain connection does not hold, it is then possible that pareto efficiency and egalitarianism could lead to an outcome that is different from the outcome preferred by the maximin [i.e. Simple] version of the difference principle …. Nonetheless, we can say that for Rawls … the application of the difference principle in such cases is problematic. We believe that Rawls would be reluctant to see the difference principle applied in precisely those situations.’ See Martin, Rawls and Rights, 95-6Google Scholar. Note that if it implies Rawls would not merely be reluctant but unwilling to apply the difference principle, then Martin's remark appears mistaken. Recall that in A Theory of justice, 80, Rawls states that even if ‘these conditions [i.e. chain connection and close-knittedness] may not hold …. We are still to maximize the expectations of the most advantaged.’

31 Here I assume that the Simple Principle is shorthand for the Lexical Principle.

32 Martin, Rawls and Rights, 98-9Google Scholar. Note that Martin assumes that for the absence of surplus inequality there must be compensating benefits for all, and not merely for the least advantaged.

33 See Rae, D. ‘Maximin Justice And An Alternative Principle Of General Advantage,’ American Political Science Review 69 (1975), 647, n. 44.CrossRefGoogle Scholar

34 On page 235 of The Marality Of Freedom Raz suggests two reasons for thinking so. Unless such principles were valid, (a) indifference would exist between wasteful and non-wasteful ways of establishing equality, and (b) no benefit to an individual would matter unless other individuals existed. Both conclusions he thinks absurd.

35 Raz, The Morality of Freedom, 235-6Google Scholar

36 Raz, The Morality of Freedom, 236Google Scholar. Note that diminishing principles assert the diminishing moral importance of certain benefits, such as happiness. This assertion is distinct from that of the diminishing personal utility of particular resources, which claims that the more of a resource one consumes, the less benefit one derives from each additional unit.

37 See Raz, The Morality of Freedom, 239–40Google Scholar, for a defense of the presupposition.

38 Raz, The Morality of Freedom, 241Google Scholar

39 Alternative forms of happiness are incommensurable when they cannot be ranked upon the same scale as either better than, worse than, or equal to each other in value. The mark of incommensurable alternatives is therefore the absence of a transitive ranking. For illustration and further discussion see, respectively, The Morality Of Freedom, 243 and ch. 13.

40 Raz, The Morality of Freedom, 242Google Scholar

41 Raz, The Morality of Freedom, 243Google Scholar

42 A revisionist might, therefore, observe that Raz's claim that ‘It is hard to think of a fundamental principle which is non-diminishing’ (ibid., 244) is false.

43 The following remarks would also apply if a Razian critic of revisionism attempted to claim that the revisionist's concern for the production of primary goods should be diminishing, and that, therefore, his recourse to a supplementary egalitarian principle to regulate their distribution is superfluous.

44 Raz's view that happiness is of diminishing moral significance is not, however, equivalent to the Simple Principle since his view (i) concerns the distribution of happiness rather than primary goods, and (ii) asserts the existence of a pro tanto reason, rather than a conclusive reason, to attach priority to the least advantaged. The reason, unlike the Simple Principle, may be over-ridden by the opportunity costs experienced by, or the numbers of, more advantaged individuals.

45 Egalitarians must also justify the metric they employ to identify different individuals' levels of advantage. But this, the problem of interpersonal comparison, is not the question I have in mind here.

46 Martin, Rawls and Rights, 220, n. 12.Google Scholar

47 For discussion of various such measures see Sen, A. On Economic Inequality (Oxford: Clarendon Press 1973), ch. 2CrossRefGoogle Scholar and Temkin, Inequality.

48 Note that this manner of application is not merely an ad hoc adjustment of the Revisionist Principle, designed to by-pass the previous intransitivity objection. It is suggested by Martin's treatment of the principle of efficiency as lexically prior to the principle of equality within his discussion of the difference principle. See Martin, Rawls and Rights, 97.Google Scholar

49 See Sen, A. Choice, Welfare and Measurement (Oxford: Blackwell 1982), 4.Google Scholar

50 Thomas Pogge has also pointed out a further potential problem. When construed to attach priority to egalitarian considerations the Revisionist Principle selects D1 from (D1, D2, D3) yet it selects D3 if D2 (40, 40, 40) is replaced with D4 (20, 20, 20).

51 Cohen, G.A. ‘Marxism And Contemporary Political Philosophy, or Why Nozick Exercises some Marxists more than he does any Egalitarian Liberals,’ Canadian journal Of Philosophy, Supplementary Volume 16, sec V.Google Scholar

52 Under different circumstances not only may distinct principles diverge but also the very same principle may be used for different purposes. Hillel Steiner's discussion of the children's fable The Little Red Hen illustrates well how a concern for producers' entitlements was put first to progressive and then to reactionary use. See Steiner, H. ‘The Fruits Of One's Labour,’ Sec. 1 in Milligan, D. and Miller, W. Watts eds., Liberalism, Citizenship And Autonomy (Aldershot: Avebury 1992) 79–87.Google Scholar

53 For further non-hypothetical contractarian defenses of the attachment of priority to the least advantaged see: Cohen, J. ‘Democratic Equality,’ Ethics 96 (1989) 727-52CrossRefGoogle Scholar; Nagel, T. ‘Equality,’ in Mortal Questions (Cambridge: Cambridge University Press 1979);Google Scholar Scanlon, T.M. ‘Contractualism and Utilitarianism,’ Sen, A. and Williams, B. eds. Utilitarianism and Beyond (Cambridge: Cambridge University Press 1982).Google Scholar

For helpful and illuminating commentary I am very grateful to E.P. Casal, M.G. Clayton, G.A. Cohen, J.N. Gray, R. Martin, D.L. Miller, M. Philp, T. Pogge, J. Raz, and T.M. Wilkinson.