Hostname: page-component-586b7cd67f-rcrh6 Total loading time: 0 Render date: 2024-11-30T21:05:07.813Z Has data issue: false hasContentIssue false
  • English
  • Français

Maximin Justice, Sacrifice, and the Reciprocity Argument: A Pragmatic Reassessment of the Rawls/Nozick Debate

Published online by Cambridge University Press:  26 January 2009

Stephen W. Ball
Get access Rights & Permissions [Opens in a new window]

Extract

Theories of economic justice are characteristically based on abstract ethical concerns often unrelated to practical distributive results. Two decades ago, Rawls's theory of justice began as a reaction against the alleged ‘sacrifices’ condoned by utilitarian theory. One variant of this objection is that utilitarianism permits gross inequalities, severe deprivations of individual liberty, or even the enslavement of society's least well-off individuals. There are, however, more subtle forms of the objection. In Rawls, it is often waged without any claim that utilitarianism does in fact imply such gross deprivations in actual realworld circumstances. A second variant hinges, rather, on the milder claim that utilitarianism could condone such deprivations or sacrifices in some possible world—the objection being that utilitarianism improperly makes justice contingent, or uncertain, in this way. A third, still more abstract, variant would be that utilitarianism is flawed—not because of any practical distributive result, actual or hypothetical, but in theory—due to the way it treats individuals' interests, or the ‘concept of persons’ it presupposes.

Type
Research Article
Information
Utilitas , Volume 5 , Issue 2 , November 1993 , pp. 157 - 184
Copyright
Copyright © Cambridge University Press 1993

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 These three lines of objection are critically analysed in my essays ‘Economic Equality: Rawls Versus Utilitarianism’, Economics and Philosophy, ii (1986), 225–44CrossRefGoogle Scholar; ‘Choosing Between Choice Models of Ethics: Rawlsian Equality, Utilitarianism, and the Concept of Persons’, Theory and Decision, xxii (1987), 209–24CrossRefGoogle Scholar; and ‘Uncertainty in Moral Theory: An Epistemic Defense of Rule-Utilitarian Liberties’, Theory and Decision, xxix (1990), 133–60.CrossRefGoogle Scholar

2 Cf. Rawls, , A Theory of Justice, Cambridge, Mass., 1971, hereafter referred to as TJ, esp. pp. viii, 3, 26–9, 177–80, 586Google Scholar; with his ‘The Idea of an Overlapping Consensus’, Oxford Journal of Legal Studies, vii (1987), 14, 19–21Google Scholar. On the more positive theme of making basic rights and liberties more ‘secure’, cf. Rawls, , ‘Kantian Constructivism in Moral Theory’, Journal of Philosophy, lxxvii (1980), 561, 563Google Scholar; and ‘Justice as Fairness: Political not Metaphysical’, Philosophy & Public Affairs, xiv (1985), 226Google Scholar. Still more recently, Rawls is concerned to define and defend his own distinctive form of ‘individualism’: ‘The Priority of Right and Ideas of the Good’, Philosophy & Public Affairs, xvii (1988), 251, 264, 267, 268–9.Google Scholar

3 Illustrative is the recent Symposium on Rawlsian Theory, in Ethics, ic (1989)Google Scholar, in which utilitarianism is only obliquely mentioned, and dismissed, in scattered references to its ‘sacrifices’, the contingent status of utilitarian rights, its ‘merging of interests’, etc. In philosophy of law, the same anti-utilitarian themes have been echoed in the Dworkin literature: cf. my ‘Dworkin and His Critics: The Relevance of Ethical Theory in Philosophy of Law’, Ratio Juris, iii (1990), 340–84CrossRefGoogle Scholar, esp. § IV on ‘Law and Politics’.

4 The term ‘maximin’ will throughout be used (as a heuristically more graphic label) to designate the criterion of distribution, also known as the ‘difference principle’ (cf. TJ, pp. 75 ff., 302–3Google Scholar)—as distinguished from the principle, by the same name, for choice under uncertainty in Rawls's hypothetical choice model (the ‘original position’): p. 152. For a synopsis, with detailed critical analysis of how Rawls's overall distributive conception works, see my ‘Economic Equality: Rawls Versus Utilitarianism’, 225–44, esp. 226–7.Google Scholar

5 Arrow, Kenneth, ‘Some Ordinalist Utilitarian Notes on Rawls's Theory of Justice’, Journal of Philosophy, lxx (1973), 257Google Scholar; Nagel, Thomas, ‘Rawls on Justice’, in Daniels, N., ed., Reading Rawls, New York, 1974, p. 13Google Scholar (repr. from Phil Review, 1973 at p. 231CrossRefGoogle Scholar); Nozick, R., Anarchy, State, and Utopia, New York, 1974, pp. 192–7.Google Scholar

6 Nozick, , Anarchy, State, and Utopia, pp. 193–4Google Scholar. Recently Joshua Cohen has given a limited response to Nozick's problem by arguing that maximin should not threaten the self-respect of the more favoured man: ‘Democratic Equality’, Ethics, xcix (1989), 740Google Scholar. Cohen's argument, however, does not solve the problem here about sacrifice which appears to be Nozick's real concern.

7 I would like to thank Allan Gibbard for critical comments (on my use of the terms ‘sacrifice’, etc.) in response to an earlier form of my analysis of Rawls's reciprocity argument, concentrated in Sections II–V of the present essay. Since writing this essay, I note that Gibbard's recent book on rationality and moral psychology contains a number of interesting suggestions of relevance to the present analysis. See Gibbard, A., Wise Choices, Apt Feelings: A Theory of Normative Judgment, Cambridge, Mass., 1990Google Scholar. On the above topic, cf. the observations that, among alternative cooperative schemes which are ‘mutually advantageous’, some are ‘more advantageous’ than others for a given personwhile other schemes contain ‘poor terms’, and morality generally requires (or is intuitively perceived as requiring) ‘genuine sacrifice’: pp. 264, 298, 308, and cf. n. 23 below. In general, Gibbard's book may be viewed as highlighting the importance of the topic of reciprocity, vis-à-vis (feelings about) fairness, in socioeconomic theory.

8 On the general idea of the ‘representative man’, see Rawls, , TJ, p. 64Google Scholar. (Rawls clearly means to refer to the gender-neutral ‘person’, and my pronoun usage throughout the present essay adequately reflects this.)

9 Nozick, who poses the general problem about ‘groups’, does not delineate these variants, and does not consider the problem as merely preliminary: Anarchy, State, and Utopia, p. 190.Google Scholar

10 I would like to thank Richard B. Brandt for helpful conversation with me on this point.

11 TJ, p. 103.Google Scholar

12 Ibid., p. 15, italics added, and the same idea about desert (or lack thereof) is reiterated at pp. 103–4. In the p. 15 version, Rawls credits Allan Gibbard as the originator of this intuitive line of argument. In conversation with me, however, Professor Gibbard has expressed reservations as to exactly what intuitions are involved here, and more general skepticism as to the theoretical need for this whole argument. The present essay addresses both of these concerns.

13 ‘Some Reasons for the Maximin Criterion’, American Economic Review, lxiv (1974), 144.Google Scholar

14 Cf. ‘Kantian Constructivism in Moral Theory’, 517, 528.Google Scholar

15 Ibid. 550–1.

16 ‘Political not Metaphysical’, cf. 232, 235–6.Google Scholar

17 ‘The Idea of an Overlapping Consensus’, 12.Google Scholar

18 Nozick, , Anarchy, State, and Utopia, pp. 196–7.Google Scholar

19 Ibid. p. 197n.

20 The outcome of Nozick's interpretative argument, later in the same note, ibid.

21 TJ, p. 74Google Scholar, italics added. Rawls contrasts his ‘democratic’ conception, designed to counteract the economic effects of both natural and social contingencies, as against the ‘liberal’ view which counteracts only the second (arbitrary) factor, and ‘natural aristocracy’, i.e. a free-market conception which counteracts neither. For further analysis of these conceptions, see my ‘Choosing Between Choice Models of Ethics’, 210–11, and esp. 213, at n. 8Google Scholar. For the ‘relative stability’ argument, against utilitarianism, cf. TJ, § 76. In general, instability arguments are of course not unheard-of in political and economic theory. Cf. Jürgen Habermas on the ‘threshold of tolerance’ for economic conditions visà-vis the stability of moral beliefs in ‘legitimacy’: Legitimationsprobleme im Spätkapitalismus, Frankfurt a.M., 1973, pp. 90, 132 ff., 177Google Scholar; or the theme that there can be ‘no [political] loyalty without [i.e., belief in moral] legitimacy’: in Habermas, Jürgen, Moralbewuβtsein und kommunikatives Handeln, Frankfurt a.M., 1983, p. 72.Google Scholar

22 TJ, p. 87Google Scholar. Though Rawls is talking here in the context of ‘background’ institutions, he presumably would not consider the maximin principle itself to be satisfied either. One aspect of bad background conditions would presumably be inter-class ‘visibility’, explained below, which tends to aggravate the socially disruptive consequences of violating maximin.

23 Once again I would like to thank Allan Gibbard for making this general suggestion to me in conversation. Certainly any suggestion here by Gibbard should be taken seriously, since Rawls himself credits Gibbard as the progenitor of the favoured-man argument, as we noted earlier (viz., n. 7). More specifically, Gibbard's suggestion corresponds to the interpretation attaching direct moral significance to the idea of ‘willing’ cooperation—as delineated and rejected below.

Again, since writing this essay, I note several related points in Gibbard, 's new book, Wise Choices, Apt Feelings: A Theory of Normative Judgment, Cambridge, Mass., 1990Google Scholar. On the one hand, there is similar reference in passing to the idea of ‘glad cooperation’, p. 270 (italics added). On the other hand, there appear to be also psychological mechanisms which would make people ‘willing to accept poor terms’: cf. p. 298.

Moreover, Gibbard elsewhere insightfully observes that Rawls's sense of ‘reasonable’ is distinct from (Rawls's sense of) ‘rational’: ibid., p. 261, n. 6. But does this not imply that people may find it rational to accept and cooperate with even ‘poor terms’ which they regard as ‘unreasonable’? If so, this appears to reinforce the problem of how the idea of ‘glad’ cooperation could help Rawls.

24 In general, much of Rawls's argument against utilitarianism hinges on this same type of equivocation as to where ‘uncertainty’ exists. See my ‘Uncertainty in Moral Theory: An Epistemic Defense of Rule-Utilitarian Liberties’, 143 ff.Google Scholar

25 TJ, pp. 535 ffGoogle Scholar. For more detail and a critical analysis on the envy problem, see my ‘Economic Equality: Rawls Versus Utilitarianism’, 233–7.Google Scholar

26 In Kant's book on practical anthropology: Bk. 2, § 60, and Bk. 3, § 75. For page references in the German text: Anthropologie in pragmatischer Hinsicht, Bibliothek, Philosophische, Bd. 44, Hamburg, 1980, pp. 155, 187Google Scholar. Kant adduces the example of a rich man angered when his servant breaks one of his rare glass goblets. Similarly, in his lectures on ethics, and again in the context of interpersonal comparisons, Kant observes that the beggar at the door is often happier than the king on his throne: Eine Vorlesung Kants Über Ethik, ed. Menzer, Paul, Berlin, 1924, p. 182 (section entitled ‘command of oneself’).Google Scholar

27 See Brandt, Richard B., ‘Problems of Utilitarianism: Real and Alleged’, in Bowie, N., ed., Ethical Theory in the Last Quarter of the Twentieth Century, Indianapolis, 1983, pp. 102 ff.Google Scholar

28 I have argued for this approximation thesis, i.e., that both yield similar degrees of economic (in) equality, in my essay, ‘Economic Equality: Rawls Versus Utilitarianism’. Cf, Rawls's casual assumption, cited at n. 13 above, that the more favoured will receive ‘less’ with maximin than with the utility principle.

29 Nozick, , Anarchy, State, and Utopia, pp. 195–6Google Scholar—although ‘maximax’ is not Nozick's term here. Nor does Nozick use the term ‘maximin’ to designate Rawls's difference principle; he denotes that principle as ‘minimax’: p. 190.

30 Ibid., esp. pp. 163, 169, 219–20. Cf. Nagel, T., ‘Libertarianism Without Foundations’, in Paul, J., ed., Reading Nozick, Oxford, 1982, p. 199, n. 13Google Scholar. Nozick's suggestion here that ‘patterned’ principles like Rawls's are generally unstable (cf. p. 164n.) is reminiscent of Rawls's own (in)stability charges against Nozickian principles (recall n. 21 above), and provides a good example of what we do not assume in the above ‘visibility’ argument (at n. 25 above). The basic pragmatic point about economic accumulation was well expressed by Rousseau, in his ‘Discourse on Political Economy’Google Scholar. The state is to prevent extreme economic inequality, not by removing treasures, but rather by removing all means of their accumulation; and the state protects the poor not by building hospitals, but by safeguarding citizens from becoming so poor: ‘C'est done une des plus importantes affaires du gouvernement de prévenir l'extrême inégalité des fortunes, non en enlevant les trésors à leurs possesseurs, mais en ôtant à tous les moyens d'en accumuler, ni en bâtissant des hôpitaux pour les pauvres, mais en garantissant les citoyens de le devenir.’ This essay, ‘Discours Sur L'Économie Politique’, in Du Contrat Social: ou principes du droit politique, Paris, 1985, p. 236.Google Scholar

31 It has been objected here that my treatment of Rawls's ‘moral argument’ in the preceding paragraphs overlooks that fact that Rawls has much more extensive moral argumentation for maximin and/or against utilitarianism. One must recall, however, that the present essay is concerned only with the ‘moral argument’ connected directly with the favoured-man argument, and in this context, the idea of reciprocity. (Recall specifically the opening paragraphs of Section IV above, where we delineated, and deferred discussion of, the ‘moral phase’ of the favoured-man/reciprocity argument until now.) This is a significant (and as I have argued, neglected) part of Rawls's overall moral argument, but it is certainly not intended that the present essay (or indeed any single essay) could deal with all of Rawls's argument. I have addressed other important strands of Rawls's anti-utilitarian argumentation in my essays cited above, esp. at note 1 above.

32 I would like to thank a critical referee for detailed guidance in shortening this essay and delimiting the scope of my argument. As to what cannot be concluded or precluded here, please recall also notes 1 and 12 above.

33 In his new book (1993) which largely reiterates his essays, which we have taken into account, Rawls emphasizes a ‘liberal’ conception of what is ‘reasonable’ or ‘reciprocal’ as a matter of political compromise, or available consensus; however, this does not explicitly address Nozick, nor solve the general problem about ‘sacrifice’, etc., that has been argued in this essay. See my review of Rawls's book, to appear in this journal.