Over the past decade, political philosophers and political theorists have had acommon purpose: to reflect on the merits of realism and idealism when theorizingabout the human condition and the nature of justice. We have settled that no oneis against being realistic or against being idealistic per se. The contributionsto this volume represent a conversation about what would make one attempt toarticulate ideals better than another.

Rawls ignited a debate in political theory when he introduced a division betweenthe ideal and nonideal parts of a theory of justice. In the ideal part of thetheory, one presents a positive conception of justice in a setting that assumesperfect compliance with the rules of justice. In the nonideal part, oneaddresses the question of what happens under departures from compliance. Criticsof Rawls have attacked his focus on ideal theory as a form of utopianism, andhave argued that political theory should be focused instead on providingsolutions to the manifest injustices of the real world. In this essay, I offer adefense of the ideal/nonideal theory distinction according to which it amountsto nothing more than a division of labor, and explore some scientific analogies.Rawls’s own focus on the ideal part of the theory, I argue, stemsfrom a felt need to clarify the foundations of justice, rather than a utopianneglect of real world problems.

A qualified pluralism is defended that recognizes value in a variety of forms ofpolitical theory and resists arguments that purport to show that one particularapproach should occupy a privileged position. Against realists, it is arguedthat abstract analyses of political values that bracket a wide range of factsabout people and their circumstances can be both coherent and important, whereasagainst those who think “ideal theory” or theidentification of ultimate principles should come first, it is argued that thecase for always giving priority to either one of these is weak.

The essay agues that there is little scope for ideal theory in politicalphilosophy, even under Rawls’s conception of its aims. It begins byidentifying features of a standard example of ideal theory in physics— the ideal gas law, PV=NRT and draws attention to thelack of these features in Rawls’s derivation of the principles ofjustice from the original position. A. John Simmons’s defense ofideal theory against criticisms of Amartya Sen is examined, as are furthercriticisms of both by David Schmidtz. The essay goes on to develop a conceptionof the domain of social relations to be characterized by justice that suggeststhat as a moving target it makes ideal theory otiose. Examination ofRawls’s later views substantiate the conclusion that ideal theory aspropounded in A Theory of Justice is a mistaken starting point in the enterpriseof political philosophy. Differences between the domains of ideal theory inmathematics, physics, and economics on the one hand, and political philosophy onthe other, reinforce this conclusion.

In this essay, I critically engage with a methodological approach in contemporary political theory — unconstrained utopianism — which holds that we can only determine how we should live by first giving an account of the principles that would govern society if people were perfectly morally motivated. I provide reasons for being skeptical of this claim. To begin with I query the robustness of the principles unconstrained utopianism purportedly delivers. While the method can be understood as offering existence proofs, because we can devise other situations in which morally flawless decision making would unearth alternative sets of principles, I argue that such proofs tell us surprisingly little about how we should live in general. Drawing on this point, I contend that normative models that wish away certain phenomena that are uncontroversially central to any account of politics cannot plausibly claim to tell us how we should live in political society. I conclude by offering a more positive sketch of why avoiding this brand of utopianism might not represent a problematic capitulation to the morally nonideal and suggest that theorizing in light of certain constraints may be a precondition of good normative theorizing itself.

Some understand utopia as an ideal society in which everyone would be thoroughlyinformed by a moral ethos: all would always act on their pure conscientiousjudgments about justice, and so it would never be necessary to provideincentives for them to act as justice requires. In this essay I argue that sucha society is impossible. A society of purely conscientiously just agents wouldbe unable to achieve real justice. This is the Paradox of PureConscientiousness. This paradox, I argue, can only be overcome when individualsare prepared to depart from their own pure, conscientious, judgments ofjustice.

This essay discusses and criticizes the claim that normative political theory canbe (justifiably and fruitfully) divided into two parts—a part havingto do with ideal theory which assumes full compliance and abstracts away fromissues having to do with implementation and, contrasting with this, a nonidealpart having to do with implementation and with rules and institutionsappropriate for conditions of partial compliance. On this conception of idealtheory, empirical facts about human behavior and motivation, connected to issuessurrounding compliance and implementation, are irrelevant to ideal theory,although such facts can be relevant to the nonideal part of normative theory. Iargue against this conception, holding instead that such empirical facts arerelevant to most or all of normative political theory, including“fundamental” normative principles.

Political philosophy appears to have recovered from its alleged death in themiddle of the last century, but now faces the realist charge that in the work ofJohn Rawls and those influenced by him it fails to be political in the rightway; it is merely “applied moral philosophy.” I dismissthe hyper-realist position of authors such as Raymond Geuss for taking animplausibly narrow view of politics. There is more merit in BernardWilliams’s claim that legitimacy, not justice, is the central problemof political philosophy. Yet we cannot understand the significance oflegitimation without referring to the moral values that are realized when itsucceeds. Thus, Williams fails to show that political normativity can bedetached entirely from ethics. Moreover the legitimacy requirements of a liberalstate, according to Williams, are substantively close to the requirements ofjustice according to Rawls. In light of the latter’s turn to“political liberalism,” they appear also to holdconvergent views about the status of the theories they are advancing. I concludeby suggesting that the “applied moral philosophy” chargewould apply only to philosophers who believe that general moral principles, likeutility or rights, can do all the work of political evaluation. Politics doesindeed have special features that impose distinctive justificatory requirementson its procedures and the outcomes they produce.

The debate between proponents of ideal and nonideal approaches to politicalphilosophy has thus far been framed as a meta-level debate about normativetheory. The argument of this essay will be that the ideal/nonideal debate can behelpfully reframed as a ground-level debate within normative theory.Specifically, it can be understood as a debate within the applied normativefield of professional ethics, with the profession being examined that ofpolitical philosophy itself. If the community of academic political theoristsand philosophers cannot help us navigate the problems we face in actualpolitical life, they have not lived up to the moral demands of their vocation. Amoderate form of what David Estlund decries as“utopophobia” is therefore an integral element of a properprofessional ethic for political philosophers. The moderate utopophobe maintainsthat while devoting scarce time and resources to constructing utopias maysometimes be justifiable, it is never self-justifying. Utopianism is defensibleonly insofar as it can reasonably be expected to help inform or improvenon-utopian political thinking.

Contra John Rawls, G. A. Cohen argues that the fundamental principles of justiceare not constrained by the limits of our nature or the nature of society, evenat its historical best. Justice is what it is, even if it will never berealized, fully or at all. Likewise, David Estlund argues that since our innatemotivations can be justice-tainting, they cannot be a constraint on the rightconception of justice. Cohen and Estlund agree that if the attempt to implementa certain conception of justice is likely to result in widespread harm orinjustice, then it should not be implemented, but that this does not entail thatthe conception itself is false. I argue that (i) there is no way to judge thesoundness of a principle of justice independently of all psychological facts,and the effects that the principle is likely to have if it is implemented; (ii)a principle of justice that, if implemented, makes it hard or impossible forindividuals committed to justice to lead happy and worthwhile lives, even if thecircumstances are favorable to living justly, cannot be sound; (iii) without theconstraints noted in (i) and (ii), there can be no reason to reject racist,sexist, or other wrongheaded principles of justice that have been advanced assound over the years, principles that even Cohen and Estlund would reject. Inshort, justice is justice only if kept within the limits of human nature.

Ideal theorists in political philosophy seek to describe a perfect politicalsociety, and to evaluate political principles by reference to their consequencesin a world where everyone complies with the principles. I argue that idealtheory is not needed to set goals for practical inquiries, nor to definejustice, nor to enable rankings of injustices. Nor is it useful to theorizeabout very different kinds of society that might occur in the far future. Idealtheory tempts us to make each of three kinds of error: it tempts us to proposenorms that no specific agent can act on, to posit crazy exaggerations of moralvirtues, and to place too much trust in abstract philosophical reasoning. Abetter approach to normative questions is to rely on analogical argumentsstarting from uncontroversial intuitions about concrete scenarios.

Political theory is not a purely theoretical enterprise; it is intended to bepractical and action-guiding. To perform this role, the requirements ofpolitical theory must be possible, and the standard of possibility it employsmust be appropriate to the political domain. Because human beings vary in theircapacity for morality and justice, a reasonably just society, as Rawlsunderstands it, must not be expected. Despite his concerns to the contrary, thepossibility of a just polity is not needed to ward off resignation and cynicism.There is a principled path between a politics of complacency that thwartsfeasible progress and a politics of utopian aspiration that ends by inflictingharm in the name of doing good.

This essay seeks to defend the claim that political philosophy ought to beappropriately guided by the phenomenon of politics that it seeks to both offer atheory of and, especially in its normative guise, offer a theory for. It doesthis primarily through the question of political values. It begins by arguingthat for any value to qualify as a value for the political domain, it must beintelligible in relation to the constitutive features of politics as a humanactivity. It then examines the extent to which the preconditions for therealization of values in practice ought to figure in our considerations as towhether they are values that fit or belong to our social world. We canunderstand these parts of the essay as responding to two related questions,respectively: (i) Is this a political value at all? — which is toask, is it a value that is appropriate for the political realm?; and then (ii)Is this a political value for us? The final section responds to the often-madecomplaint that political philosophy ought not to make any concessions to theactual world of politics as it really is, arguing that attending to therealities of politics, and in particular the constitutive conditions ofpolitical activity, gives meaning to the enterprise as the theorization ofpolitics (and not something else). Furthermore those same conditions provide thelimits of intelligibility beyond which ideals and values can no longer be, inany meaningful sense, ideals and values for the political sphere.

This essay offers an account of feasible actions. It criticizes the conditionalaccount of feasibility and offers instead what I call the constrained account offeasibility. The constrained account is superior, I argue, on account of how itdeals with the problem of motivational failure to act and with collectiveaction. According to the constrained account, roughly put, an action is feasiblewhen the agent or agents performing it know how to perform it and areappropriately responsive to incentives. The essay shows that some collectiverequirements for action that appear feasible are not in fact feasible.

Does social justice lose all application in the (imaginary, of course) conditionin which people are morally flawless? The answer, I will argue, is that it doesnot — justice might still have application. This is one lesson of mybroader thesis in this paper, that there is a variety of conditions we would allregard as highly idealistic and unrealistic which are, nevertheless, not beyondjustice. The idea of “circumstances of justice” developedespecially by Hume and Rawls may seem to point in a more realistic direction,but we can see that this is not so once we distinguish between conditions ofneed for norms of justice, conditions of their emergence, and conditions ofapplicability of the standard of justice. Justice, I argue, can have applicationeven in conditions where no mechanism of justice is present or needed, such asthe case of internalized motives of justice.

In this essay, I argue against the bright-line distinction between ideal andnonideal normative political theory, a distinction used to distinguish“stages” of theorizing such that ideal politicalprinciples can be deduced and examined before compromises with the flawedpolitical world are made. The distinction took on its familiar form in Rawls andhas enjoyed a resurgence of interest in the past few years. I argue that theidea of a categorical distinction — the kind that could allow for asequencing of stages of theorizing — is misconceived, because wholly“ideal” normative political theory is a conceptualmistake, the equivalent of taking the simplifying models of introductory physics(“frictionless movement in a vacuum”) and trying todevelop an ideal theory of aerodynamics. Political organization and justice areabout moral friction in the first instance. I examine both logical andepistemological arguments for the position that we need the uniquely idealizingassumptions of ideal theory in order to arrive at, or to know, a genuine theoryof justice or political morality; and I find them wanting. Such assumptions asfull compliance, consensus, and the publicity principle of universal knowledgeabout consensus can sometimes be useful, if used carefully and withjustification; but they are not categorically different from other idealizingand abstracting assumptions in generating normative theory. What is referred toas “nonideal” theory is all that there is, and it is manykinds of theory, not one — the many ways in which we learn aboutjustice and injustice, and seek to answer questions of practical reason aboutwhat ought to be done in our political world.

In The Law of Peoples, John Rawls introduces a framework forrealistic utopia, within which the limits of practicable political possibilityare probed through the further development of his international theory. Thisessay addresses the apparent paradox of realistic utopianism within the contextof, and in relation to, ideal theory, in an attempt to explore the scope andlimits of Rawls’s theory. The ideas behind Rawls’srealistic utopia are discussed in detail, the concept is contrasted with idealtheory in order to assess to what extent Rawls’s framework forrealistic utopia introduced in The Law of Peoples differs fromother forms of ideal theory, and the limits of realistic utopianism areidentified.

I argue first, that, in an attempt to address the potential feasibilityconstraint, Rawls tries to distinguish his framework of realistic utopia fromthat of more traditional ideal theory. I then proceed to examine the differencesbetween realistic utopianism in The Law of Peoples and idealtheory in A Theory of Justice. I then conclude that Rawls onlypartially meets the challenge of establishing practicable political possibility.In actuality, Rawls’s focus on ideal agents in ideal as well asnonideal theory, together with his emphasis on societies as closed andself-sufficient, ignores the potential for noncompliance by liberal and decentsocieties, as well as interdependencies between societies that can cause or leadto injustice, conflict, and instability. I argue that despite these flaws,Rawls’s approach nevertheless provokes new insights into the functionof the principles of the ideal theory framework as guidelines for real-worldpolicies striving toward peace, stability, and justice.

The ideal theory debate rests on two conflicting claims: that justice is“the first virtue of social systems” (justicefirst), and that a just society is one in which“everyone accepts and knows that the others accept the sameprinciples of justice” (universal consent).Justice first holds that questions about the meaning ofjustice — and thus about what an ideally just society would look like— must be settled before we can effectively pursue justice. However,universal consent entails a project of justification thatcan only take place over time. I propose that we avoid this impasse by treatingfreedom rather than justice as the “first virtue” of aliberal society. Liberal freedom has two distinct and complementary dimensions,which give rise to two distinct and complementary moral aims: on the one hand,to create the social conditions that make responsible agency possible(republican freedom), and on the other hand to carve out a social space withinwhich the demands of responsible agency are relaxed or absent (market freedom).Striking the appropriate balance between these two dimensions of liberal freedomis irreducibly a matter of judgment. A freedom-centered liberalism thereforerequires that we treat justice as the endpoint rather than the starting point ofpolitical action, thus severing the link between legitimacy and consent.