¹ Dunn, John, Interpreting Political Responsibility (Cambridge: Cambridge University Press, 1990), 193.Google Scholar

² Cohen, G. A., Rescuing Justice and Equality (Cambridge, MA: Harvard University Press, 2008), 268.CrossRefGoogle Scholar

³ It is easy to forget that it is not just critics of ideal theory who have sought to limit the pluralism of political theory; it is also those ideal theorists who have argued that ideal theory needs to be done first.

⁴ There is a difficult issue concerning whether we should think of justice as a master value — in effect whether we should regard “justice” as an umbrella term that shelters any weighty consideration relevant to our evaluation of social arrangements — or whether we should think of it as simply one important value among many. Some realists have objected to the primacy that Rawlsians give to justice in evaluating social institutions, but that criticism makes sense only if we regard justice as one value among several. I will sidestep this underlying issue and simply assume that a theory of justice has an important place in our normative thought, even if there are other independent values the theorization of which raises the same methodological issues, and even if justice does not have the kind of primacy that Rawls attributes to it.

⁵ See Waldron, Jeremy, Law and Disagreement (Oxford: Oxford University Press, 1999),CrossRefGoogle Scholar especially 1–4, 149–63. See also Galston, William, “Realism in Political Theory,” European Journal of Political Theory 9 (2010): 385–411, at 391;CrossRefGoogle Scholar Mason, Andrew, “Rawlsian Theory and the Circumstances of Politics,” Political Theory 38 (2010): 658–83,CrossRefGoogle Scholar especially 658–64; Sleat, Matt, Liberal Realism: A Realist Theory of Liberal Politics (Manchester: Manchester University Press, 2013), esp. chap. 2.CrossRefGoogle Scholar

⁶ Mason, “Rawlsian Theory and the Circumstances of Politics,” 664; Matt Sleat, “Realism, Liberalism and Non-ideal Theory Or, Are there Two Ways to Do Realistic Political Theory?” Political Studies, DOI: 10.1111/1467-9248.12152, p. 10.

⁷ Some define ideal theory more broadly so that it includes theorizing that involves no concern with action guidance, but for reasons that will become clearer I prefer to distinguish ideal theory “proper” from theorizing about ideals that has no inherent concern for guiding action. For discussion of the nature of ideal theory, see especially John Simmons, A., “Ideal and Nonideal Theory,” Philosophy and Public Affairs 38 (2010): 5–36.CrossRefGoogle Scholar See also Stemplowska, Zofia, “What’s Ideal About Ideal Theory?” Social Theory and Practice 34 (2008): 319–40;CrossRefGoogle Scholar Robeyns, Ingrid, “Ideal Theory in Theory and Practice,” Social Theory and Practice 34 (2008): 341–62;CrossRefGoogle Scholar Hamlin, Alan and Stemplowska, Zofia, “Theory, Ideal Theory, and the Theory of Ideals,” Political Studies Review 10 (2012): 48–62.CrossRefGoogle Scholar

⁸ How it is formulated will depend in part upon precisely what role an ideal theory is thought to play in guiding political action: see Section IV for further discussion.

⁹ For discussion of the distinction between hard and soft constraints, see Pablo Gilabert and Holly Lawford-Smith, “Political Feasibility: A Conceptual Exploration,” Political Studies 60 (2012): 808–825, at 813; Holly Lawford-Smith, “Understanding Political Feasibility,” Journal of Political Philosophy 21 (2013): 243–59, at 252.

¹⁰ See John Rawls, Political Liberalism, paperback edition (New York: Columbia University Press, 1996), 54–58 — though, of course, Rawls seems to think that the burdens of judgment affect the justification of comprehensive moral doctrines in a more fundamental way than the justification of principles of justice. See also Sleat, Liberal Realism, 133–36; David Schmidtz, “Nonideal Theory: What It Is and What It Needs to Be,” Ethics 121 (2011): 772–96, at 781–83.

¹¹ In one sense of this slippery term, realists are claiming that the notion of justice is essentially contestable. Even though the reasons that people give in favor of their own particular interpretations have force, they are not logically compelling. See Andrew Mason, Explaining Political Disagreement (Cambridge: Cambridge University Press, 1993), especially chap. 2.

¹² Sleat, Liberal Realism, chap. 2.

¹³ For the importance of distinguishing between the different roles that ideal theory might play, for example, in providing a gauge or standard, a model or template, or a target, see Jennan Ismael, “A Philosopher of Science Looks at Idealization in Political Theory,” this volume.

¹⁴ Rawls, Political Liberalism, xlvii-xlix.

¹⁵ Rawls’s category of reasonable conceptions of justice is narrower than one might initially think, for it seems to exclude libertarian conceptions and some communitarian ones: see Mason, “Rawlsian Theory and the Circumstances of Politics,” 661.

¹⁶ This shows that an ideal theory can acknowledge the need for legal institutions even in a fully just society, for they may be necessary for securing full compliance. Indeed, Rawls himself maintains that legal institutions would be needed to motivate full compliance in such a society because without them people will lack the assurance that others are doing their part. See Rawls, John, A Theory of Justice, rev. ed. (Oxford: Oxford University Press, 1999), 211.Google Scholar

¹⁷ Onora O’Neill contrasts idealizations such as these with abstractions: “[a]bstraction . . . is a matter of bracketing, but not of denying, predicates that are true of the matter under discussion”; in contrast, an assumption “idealises when it ascribes predicates . . . that are false in the case in hand, and so denies predicates that are true of that case.” (Onora O’Neill, Towards Justice and Virtue. A Constructive Account of Practical Reasoning [Cambridge: Cambridge University Press, 1996], 40–41.) The assumption of convergence is rather different from that of full compliance. The latter is also sometimes regarded as an idealization, but that is misleading: it is not an assumption introduced for the purpose of simplifying the construction of the theory, but rather it is part of what makes a theory ideal in the relevant sense: a society could not be fully just unless its institutions realize the relevant principles of justice and its members comply with the principles of justice that apply to them.

¹⁸ Rawls, Theory of Justice, 218, 6. It can be misleading to regard idealizations as falsehoods in the way that O’Neill does. In ideal theories, they are best understood as a kind of counterfactual reasoning. Rawls is not asserting that all members of society are able to manage their own affairs or that societies are closed systems; instead he is asking what principles would apply to a society in which all its members had capacities above a particular threshold, joined it at birth, and exited it only at death.

¹⁹ For relevant discussion, see especially Laura Valentini, “On the Apparent Paradox of Ideal Theory,” Journal of Political Philosophy 17 ( 2009), 332–55; Colin Farrelly, “Justice in Ideal Theory: A Refutation,” Political Studies 55 (2007), 844–64.

²⁰ We might also distinguish between action guidance and practical relevance. Even if an ideal theory could not provide us with reasons for action here and now but is likely to provide us with reasons for action in the foreseeable future, then that would show that it has practical relevance. Indeed, if we cannot justifiably rule out the possibility that the theory will provide reasons for action in the foreseeable future, then that would seem to be enough to show that it has practical relevance.

²¹ See Sleat, “Realism, Liberalism and Non-ideal Theory.”

²² See Matt Sleat, “What Is a Political Value? Political Philosophy and Fidelity to Reality,” this volume.

²³ See Mason, “Rawlsian Theory and the Circumstances of Politics.”

²⁴ Cf. Bernard Williams, In the Beginning Was the Deed (Princeton, NJ: Princeton University Press, 2005), 8–10.

²⁵ See Estlund, David, “What Good Is It? Unrealistic Political Theory and the Value of Intellectual Work,” Analyse and Kritik 33 (2011): 395–416.Google Scholar

²⁶ See Hamlin and Stemplowska, “Theory, Ideal Theory and the Theory of Ideals,” 52–58; Cohen, Rescuing Justice and Equality, 268. See also Adam Swift, “The Value of Philosophy in Nonideal Circumstances,” Social Theory and Practice 34 (2008): 363-87, especially 366–68; Estlund, “What Good Is It?”

²⁷ Edward Hall, “Political Realism and Fact-Sensitivity,” Res Publica 19 (2013): 173–81, at 174; see also 175–76.

²⁸ Anderson, Elizabeth, The Imperative of Integration (Princeton, NJ: Princeton University Press, 2010), 6.Google Scholar

²⁹ Hall’s thesis seems to rule out the possibility of deontological principles that give us pro tanto reasons for prohibiting certain kinds of actions independently of what costs would be incurred in doing so. But it is primarily directed against the fact-insensitive ultimate principles that are not tailored to fit any practical purposes that G. A. Cohen thinks underpin rules of regulation that guide our decision making.

³⁰ Galston, “Realism in Political Theory,” 393.

³¹ For Wittgenstein’s remarks on rule-following, see Ludwig Wittgenstein, Philosophical Investigations, trans. G. E. M. Anscombe, 3d ed. (Oxford: Blackwell, 1967), paragraphs 138–242.

³² See Elizabeth Anderson, “Reply to Critics of The Imperative of Integration,” Political Studies Review 12 (2014), 376–82, at 379.

³³ In this context it is worth noting that G. A. Cohen responds to Anderson’s own critique of luck egalitarianism, which argues against luck egalitarianism partly by considering what the effects would be of implementing it, by maintaining that her argument does not undermine the luck egalitarian principle itself but merely shows that we need to consider the costs in terms of other values of trying to implement it and then balance the value of implementing it against those costs. See Elizabeth Anderson, “What Is the Point of Equality?” Ethics 109 (1999): 287–337; Cohen, Rescuing Justice and Equality, 271.

³⁴ Hall, “Political Realism and Fact-Sensitivity,” 178.

³⁵ For relevant discussion, see Gheaus, Anca, “The Feasibility Constraint on the Concept of Justice,” Philosophical Quarterly 63 (2013): 445–64;CrossRefGoogle Scholar Andrew Williams, “Justice, Incentives and Constructivism,” Ratio 21 (2008): 476-93, at 490–92; Swift, Adam and Stemplowska, Zofia, “Ideal and Nonideal Theory,” in Estlund, David, ed., Oxford Handbook of Political Philosophy (Oxford: Oxford University Press, 2012), 384–85.Google Scholar

³⁶ See Andrew Mason, “What is the Point of Justice?” Utilitas 24 (2012): 525–47, at 527–28.

³⁷ See David Miller, “A Tale of Two Cities; or, Political Philosophy as Lamentation,” in his Justice for Earthlings: Essays in Political Philosophy (Cambridge: Cambridge University Press, 2013), 237–39.

³⁸ Simon Hope emphasizes the idea that reflecting upon what justice requires of us individually or collectively is a form of practical reason rather than theoretical reason. See Simon Hope, “Idealization, Justice, and the Form of Practical Reason,” this volume.

³⁹ It seems to have highly counterintuitive consequences. See Gheaus, “The Feasibility Constraint on the Concept of Justice,” 452.

⁴⁰ Cohen, Rescuing Justice and Equality, 251.

⁴¹ Cohen, Rescuing Justice and Equality, 252.

⁴² For the distinction between ultimate principles and rules of regulation, see Cohen, Rescuing Justice and Equality, chaps. 6–7, esp. 263ff.

⁴³ This would not undermine Cohen’s thesis that ultimate principles are not grounded in facts of any kind. As he emphasizes, when a fact is a condition of the application of a principle, it is not part of the justification of it. See Cohen, Rescuing Justice and Equality, 331–36.

⁴⁴ See Amartya Sen, “What Do We Want from a Theory of Justice?” The Journal of Philosophy 103 (2006): 215–38; The Idea of Justice (Cambridge, MA: Harvard University Press, 2009). See also Anderson, The Imperative of Integration, 3.

⁴⁵ See Dunn, Interpreting Political Responsibility, 197. See also Judith N. Shklar, “The Liberalism of Fear,” in J. N. Shklar, Political Thought and Political Thinkers, ed. S. Hoffman (Chicago: University of Chicago Press, 1998), 10–11.

⁴⁶ Judith J. Thomson, “A Defence of Abortion,” Philosophy and Public Affairs 1 (1971): 47–66.

⁴⁷ For more on the nature of analogical reasoning, and defenses of it, see Mason, Explaining Political Disagreement, 30–35; Cass Sunstein, “On Analogical Reasoning,” Harvard Law Review 106 (1993): 741–91, especially 773–81.

⁴⁸ Galston, “Realism in Political Theory,” 402–3. The title of David Miller’s paper, “Political Philosophy for Earthlings,” in his Justice for Earthlings: Essays in Political Philosophy (Cambridge: Cambridge University Press, 2013), alludes to much the same point.

⁴⁹ Miller, “Political Philosophy for Earthlings,” 31.

⁵⁰ See Mason, “What is the Point of Justice?” 539, 546.

⁵¹ Simmons, “Ideal and Nonideal Theory,” 21ff; 34–36. Rawls maintains that ideal theory provides “the only basis for the systematic grasp” of the problems raised by noncompliance that are treated within nonideal theory, and that ideal theory “is the fundamental part of the theory of justice and essential for the nonideal part as well” (Rawls, Theory of Justice, 8, 343).

⁵² Simmons, “Ideal and Nonideal Theory,” 22.

⁵³ Suppose for example that a society has reasons of justice for putting in place a package of austerity measures, but adopting that package would make it highly likely that a fascist government will end up in power. We don’t need an ideal theory of justice to believe with justification that it would be better not to implement the austerity measures. I owe this example to Jerry Gaus.

⁵⁴ Nor do we even have the kind of knowledge that would enable us reliably to distinguish hard from soft constraints in our determination of what constitutes perfect justice. See also Hamlin and Stemplowska, “Theory, Ideal Theory and the Theory of Ideals,” 59; Swift and Stemplowska, “Ideal and Nonideal Theory,” 379–80.

⁵⁵ Simmons, “Ideal and Nonideal Theory,” 24.

⁵⁶ For discussion of the distinction between what people can’t do and what they won’t do and its significance for normative theory, see David Estlund, Democratic Authority: A Philosophical Framework (Princeton, NJ: Princeton University Press, 2008), 264–70; “Utopophobia,” Philosophy and Public Affairs 42 (2014): 113–34.

⁵⁷ See Rawls, Political Liberalism, xix; Justice as Fairness: A Restatement (Cambridge, MA: Harvard University Press, 2001), 13.

⁵⁸ Miller thinks that Rawls’s feasibility condition also excludes principles with implications that we would find intolerable: "for Rawls the limits of political possibility are set not just by political laws, but by the range of outcomes that we — people in modern liberal societies — would regard as fundamentally unacceptable” (Miller, “Political Philosophy for Earthlings,” 33).

⁵⁹ Buchanan, Allen, Justice, Legitimacy, and Self-Determination. Moral Foundations of International Law (Oxford: Oxford University Press, 2004), 60–62.Google Scholar

⁶⁰ Simmons, “Ideal and Nonideal Theory,” 29–30.

⁶¹ See Paul Feyerabend, Against Method 3d ed. (London: Verso, 1993), especially chap. 1.

⁶² Miller seems to deny that this is a legitimate form of political theorizing: see “Political Philosophy for Earthlings,” 46–47.

⁶³ See Schmidtz, “Nonideal Theory,” 774.

⁶⁴ Indeed, it seems to me that these are the reasons why G. A. Cohen thought we should strive to identify the correct fundamental principles of justice; it’s not that he thought that the project of devising the best “rules of regulation” for a society was unimportant.