Abstract
In recent work in political philosophy there has been much discussion of two approaches to theorizing about justice that have come to be called ‘ideal theory’ and ‘non-ideal theory’. The distinction was originally articulated by Rawls, who defended his focus on ideal theory in terms of a supposed ‘priority’ of the latter over non-ideal theory. Many critics have rejected this claim of priority and in general have questioned the usefulness of ideal theory. In diagnosing the problem with ideal theory, they have frequently fingered for blame the idealization it involves. In this paper I focus on one particular, much-discussed idealization—full compliance—in order to defend it. Focusing on the assumption, I argue that Rawls’s work is not ideal in the way that it is usually thought to be, is less ideal than is widely recognized, and became less ideal over time. I also argue that critics who in effect claim that it is not realistic enough simply fail to understand Rawls’s central motivation. Finally, I defend the assumption by arguing that there is an important sense in which all theories of justice must assume full compliance. Such an assumption, I argue, is needed if we are to have a plausible basis on which to judge the normative attractiveness of a theory.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
To defend the full compliance assumption, however, is not to say that all objections that have been expressed about ideal theory are without merit.
It should be noted that while Rawls makes a number of other assumptions (e.g. that citizens have the ability to be fully cooperating members of society; see Rawls 2005, p. 18), I follow Swift and Stemplowska (2014) in thinking that these should be understood as assumptions of a different kind (‘simplifying’ ones) than those that define ideal theory. Swift and Stemplowska mark the distinction in this way: ‘simplifying assumptions make a complex problem more manageable, whether that problem is one of ideal or nonideal theory; assumptions of ideal theory, on the other hand, are those that allow us to focus on the particular question addressed by an ideal theory of justice’ (p. 126, n. 3).
Also: ‘The intuitive idea is to split the theory of justice into two parts. The first or ideal part assumes strict compliance and works out the principles that characterize a well-ordered society under favorable circumstances’ (Rawls 1999a, p. 216; emphases added).
So Rawls’s view is that some principles can be such that it is too difficult even for those with a sense of justice to comply with them, and so should be rejected for that reason, not that the sense of justice guarantees compliance such that those who find it too difficult to comply with any given principle lack the sense of justice. Swift and Stemplowska erroneously attribute the latter view to Rawls in (2014, p. 123).
The ideal, in other worlds, is for citizens to realize their capacity for a sense of justice, not just their capacity to comply—a well-ordered society would still be the goal. So the idea of well-orderedness plays roles in Rawls’s work that do not have to do with how realistic justice as fairness is. (I thank two anonymous referees for asking me to clarify this part of the discussion.)
Here my discussion passes over some tricky issues having to do with Rawls’s perplexing claim that certain conditions of life that he calls ‘natural limitations’ belong to non-ideal theory (Rawls 1999a, pp. 215–217). For discussion, see (Simmons 2010, pp. 12–17); and (Swift and Stemplowska 2014, pp. 113–115).
Marcus Arvan argues that there are three conditions of ideal theory, and consequently that there are three circumstances in which we might need non-ideal theory: where there are unfavorable conditions, partial compliance, or what he calls ‘no-circumstances-of-justice’. This last refers to the absence of ‘the normal conditions under which human cooperation is both possible and necessary’, such as roughly equal capacities, moderate scarcity and conflicting interests (Rawls 1999a, pp. 109–110). For Arvan’s argument, see (Arvan 2014, pp. 97–100). But it seems to me that it is not clear if anything that can properly be called ‘Rawlsian non-ideal theory’ can be applied where such circumstances do not obtain.
It is frequently said that in focusing on ideal theory, Rawls developed a theory of justice for societies in which there is no history of injustice; see, e.g., (Mills 2009, p. 179) and (Waligore 2016, pp. 48, 52). But this is plainly incorrect. Rawls’s whole point in developing his theory was to find a theory that societies could use to judge and, if necessary, reform their basic institutions, and he took it for granted that few if any societies were not in need of such reform.
This way of putting the point assumes that full compliance can be said to exist even in the absence of the full rectification of past injustice.
The most sustained argument to this effect can be found in (Simmons 2010, pp. 18–25).
I noted this in passing in (Matthew 2017, p. 254, n. 24). This important point is also recognized by (Swift and Stemplowska 2014, p. 123), and (Laurence 2018). (Despite expressing this important truth, however, Laurence does not seem to understand just how Rawls relaxes the assumption.) It is also implied by (Simmons 2010, p. 9).
True, at one point in Chap. 8 of Theory Rawls writes that ‘a well-ordered society endures over time’ (Rawls 1999a, p. 398). This suggests that in supposing well-orderedness there, he also supposes stability. This is consistent with one possible interpretation of the stability problem in Part Three (according to which the sole question at issue is that of the relative stability of justice as fairness and utilitarianism), but I believe that Rawls’s discussion there as a whole doesn’t take it for granted that a well-ordered Rawlsian society endures.
This is an admittedly idealized form of stability that, presumably, actual stable societies need not quite fully resemble.
This point eludes Heidi M. Hurd, and this leads her to a deeply misguided—and uncharitable—critique of Rawls; see her (1995).
Rawls has often been misunderstood on this point, with commentators being of the view that in PL a well-ordered society must be presupposed to exist before his investigation of that book’s central question can get going. For examples of this error, see (Chambers 2006: esp. pp. 82–83; Copp 1996; Freeman 1994: esp. pp. 638–639).
It should be noted that Levy does not deny that the assumption of full compliance might be a useful idealization for certain purposes (see Levy 2016, pp. 314, 317, 324); he simply denies that it is a necessary one.
Rawls: ‘It is essential to observe, however, that this assumption still permits the consideration of men’s capacity to act on the various conceptions of justice. The general facts of human psychology and the principles of moral learning are relevant matters for the parties to examine. If a conception of justice is unlikely to generate its own support, or lacks stability, this fact must not be overlooked. For then a different conception of justice might be preferred’ (1999a, pp. 125–126).
Schmidtz also fails to get this, at one point even writing: ‘If indeed “an important feature of a conception of justice is that it should generate its own support”… then a serious investigator does not stipulate that a conception has that pivotal feature (when in the real world it patently does not). A serious investigation checks’ (2018, p. 341, quoting Rawls). But as I have been arguing, Rawls assumes strict compliance precisely as a way of checking.
Estlund (2011) explicitly discusses the strains on pp. 226–229.
In Rawls’s case that selection procedure is the original position device.
Cohen arguably misinterprets Rawls when he understands stability in terms of the ‘propensity’ of principles of justice to last (although a stable Rawlsian society does endure, what is important for Rawls is that principles are able to gain citizens’ allegiance), but I believe that his arguments can probably be reformulated to apply to a correct understanding of Rawlsian stability.
This is, of course, not to say that those who favor the Rawlsian view have no responses to these criticisms. For criticism of Estlund, see (Wiens 2016); but see also (Estlund 2016). As for Cohen, though several critics have rejected his more general constructivism argument, it is much harder to find critiques of his stability argument. It is discussed in (Kaufman 2018: Chap. 3), but he is only concerned to argue that Cohen’s argument depends on a misunderstanding of Rawls.
This presumption, it may be argued, does not always hold, and this possibility may be seized on as the basis for an objection: namely, that for some distributional patterns, and more generally for some theories, something less than full compliance is sufficient to fully produce the relevant pattern (or other results), and so enable evaluation. I grant that something less than full compliance may be sufficient to fully produce the result, but I doubt that much less than full compliance can be sufficient. Even for Rawls, full compliance means ‘(nearly) everyone’ complies with the principles (2001, p. 13; emphasis added).
For my purposes I understand the ‘normative attractiveness’ of a theory in terms that exclude compliance considerations, though I am aware that some regard compliance considerations as at least partly constitutive of what makes a theory normatively attractive.
To be sure, since Rawls regards the two principles as normatively attractive partly because of compliance considerations, had he held such considerations to be irrelevant to the evaluation of theories, he would presumably have defended some other set of principles.
Although above I said that Rawls thought that his quest to know what a perfectly just achievable society looks like ‘requires’ the assumption of full compliance, I should acknowledge that it is not entirely clear that he regarded the assumption as quite mandated by his concern that theories be realistic.
The view is similar to the view that Estlund calls ‘complacent realism’, and thus faces a similar problem; see Estlund (2014, pp. 115, 132).
References
Anderson, Elizabeth. 2010. The Imperative of Integration. Princeton, NJ: Princeton University Press.
Arvan, Marcus. 2014. First Steps Towards a Nonideal Theory of Justice. Ethics & Global Justice 7: 95–117.
Chambers, Simone. 2006. The Politics of Equality: Rawls on the Barricades. Perspective on Politics 4: 81–89.
Cohen, G. A. 2008. Rescuing Justice and Equality. Cambridge, MA: Harvard University Press.
Copp, David. 1996. Pluralism and Stability in Liberal Theory. The Journal of Political Philosophy 4: 191–206.
Espindola, Juan, and Moises Vaca. 2014. The Problem of Historical Rectification for Rawlsian Theory. Res Publica 20: 227–243.
Estlund, David. 2011. Human Nature and the Limits (If Any) of Political Philosophy. Philosophy & Public Affairs 39: 207–237.
Estlund, David. 2014. Utopophobia. Philosophy & Public Affairs 42: 113–134.
Estlund, David. 2016. Reply to Wiens. European Journal of Political Theory 15: 353–362.
Farrelly, Colin. 2007. Justice in Ideal Theory: A Refutation. Political Studies 55: 844–864.
Freeman, Samuel. 1994. Political Liberalism and the Possibility of a Just Democratic Constitution. Chicago-Kent Law Review 69: 619–668.
Hamlin, Alan, and Zofia Stemplowska. 2012. Theory, Ideal Theory and the Theory of Ideals. Political Studies Review 10: 48–62.
Hurd, Heidi M. 1995. The Levitation of Liberalism. Yale Law Journal 105: 795–824.
Kang, Hye-Ryoung. 2016. Can Rawls’s Nonideal Theory Save His Ideal Theory? Social Theory and Practice 42: 32–56.
Kaufman, Alexander. 2018. Rawls’s Egalitarianism. Cambridge: Cambridge University Press. Ch. 3.
Laurence, Ben. 2018. Constructivism, Strict Compliance, and Realistic Utopianism. Philosophy and Phenomenological Research 97: 433–453.
Levy, Jacob. 2016. There is No Such Thing as Ideal Theory. Social Philosophy & Policy 33: 312–333.
Matthew, D. C. 2017. Rawls and Racial Justice. Politics, Philosophy and Economics 16: 235–258.
Matthew, D. C. 2015. Rawlsian Affirmative Action: A Reply to Robert Taylor. Critical Philosophy of Race 3: 324–343.
Mills, Charles W. 2009. Rawls on Race/Race in Rawls. Southern Journal of Philosophy Supplement 47: 161–184.
Mills, C. W. 2005. ‘Ideal Theory’ as Ideology. Hypatia: A Journal of Feminist Philosophy 20: 165–184.
Rawls, John. 2005. Political Liberalism. Expanded edn. New York, NY: Columbia University Press.
Rawls, John. 2001. In Justice As Fairness: A Restatement, ed. Erin Kelly. Cambridge, MA: Harvard University Press.
Rawls, John. 1999a. A Theory of Justice. Revised Ed.. Cambridge, MA: Harvard University Press.
Rawls, John. 1999b. The Law of Peoples. Cambridge, MA: Harvard University Press.
Schmidtz, David. 2018. Ideal Theory. In Oxford Handbook of Distributive Justice, ed. Serena Olsaretti, 327–345. Oxford: Oxford University Press.
Sen, Amartya. 2006. What Do We Want from a Theory of Justice? Journal of Philosophy 103: 215–238.
Sen, Amartya. 2009. The Idea of Justice. Cambridge, MA: Harvard University Press.
Simmons, A. John. 2010. Ideal and Nonideal Theory. Philosophy & Public Affairs 38: 5–36.
Swift, Adam, and Zofia Stemplowska. 2014. Rawls on Ideal and Nonideal Theory. In The Blackwell Companion to Rawls, ed. Jon Mandle and David Reidy, 112–127. Oxford: Wiley-Blackwell.
Valentini, Laura. 2009. On the Apparent Paradox of Ideal Theory. Journal of Political Philosophy 17: 332–355.
Vandenbroucke, Frank. 2001. Social Justice and Individual Ethics in an Open Society: Equality, Responsibility, and Incentives, Ch. 8. New York, NY: Springer.
Waligore, Timothy. 2016. Rawls, Self-respect, and Assurance: How Past Injustice Changes What Publicly Counts as Justice. Politics, Philosophy & Economics 15: 42–66.
Wiens, David. 2016. Motivational Limitations on the Demands of Justice. European Journal of Political Theory 15: 333–352.
Acknowledgements
I would like to thank Louis-Philippe Hodgson for reading and providing comments on a previous draft of this paper.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Matthew, D.C. Rawls’s Ideal Theory: A Clarification and Defense. Res Publica 25, 553–570 (2019). https://doi.org/10.1007/s11158-018-09412-1
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11158-018-09412-1