Kant, Rawls, and the Moral
Foundations of the Political
Abstract: This paper recasts the problematic of the moral foundations of the political in light of John Rawls's critical appropriation of Immanuel Kant's practical philosophy (GMS, KpV, MdS, and political writings). It is shown to what extent one may succeed nowadays in preserving the normative principle of universalizability without falling back into moral foundationalism and in accounting for the stability of a "well-ordered society" in the very terms of an unsociable sociability combining morality and legality (articulation of the universal principle of justice and the categorical imperative). By recasting a theory of justice as fairness, starting from the "fact of pluralism" and from a conception of a "public political culture", Rawlsian contractarianism not only corrects its own inconsistencies but reiterates its Kantian-inspired proceduralism. Despite its shift from a comprehensive doctrine of justice in his 1971 masterpiece (A Theory of Justice), Rawls’s later political conception of justice (esp. Political Liberalism) recasts Kant’s procedural device of self-determination and autonomy insofar as social agency is inevitably caught in reflective equilibrium. In lieu of celebrating a shift from the moral foundations of the political towards a specifically political theory of justice, Rawls’s constructivism recasts some of the very problems inherent in a Kantian critique of moral realism and intuitionism in ethics and political philosophy.
This paper reflects the partial results of a broader research on the so-called Rawls-Habermas debate.As indicated by the title, I am dealing here with the problem of justifying (in the sense of begründen, grounding, giving reasons for) a theory of justice in light of John Rawls’s political appropriation of Kantian constructivism. What is ultimately at stake hence is the task of reappropriating Immanuel Kant's political theory of justice as a constructivist alternative to what has nowadays been identified with moral foundationalism in realist, intuitionist attempts to ground a political theory. No other work has received as much attention and raised as many criticisms in political philosophy today as Rawls’s A Theory of Justice and the later writings that led to the volume on Political Liberalism. It is my contention here that Rawls’s monumental work not only emerges as a useful referential in today’s ethical-political framework, but it also rescues the Kantian thrust of our contemporary debate opposing universalists and communitarians, constructivists and intuitionists, realists and anti-realists, deontologist and utilitarian models in ethics and political theory. From the very outset, I should like to outline the main theses and sets of problems that guide this study:
I placed this inquiry into the foundations of a theory of justice at the heart of the current debate among universalists insofar as the nature of morality and political philosophy is concerned. According to this view, political philosophy cannot account for the nature and justification of social, political institutions without presupposing a universalizable, normative conception of morality. Such is the Kantian thrust of this view, as opposed to the communitarian grounding of ethics and political philosophy upon the tradition and context out of which discursivity itself takes place. Grosso modo, both universalists and communitarians can be called "cognitivist," insofar as they agree on the possibility of knowing the foundations of moral principles and the necessity of coming up with some moral theory. In short, there must be objectivity in moral reasoning, as one seeks to avoid the pitfalls of both foundationalism and relativism. In this sense, both teleological (i.e., virtue ethics and utilitarianism) and deontological ethics (i.e., Kantian-inspired and others) are to be opposed to noncognitivist approaches to moral philosophy. As Kenneth Baynes has shown in his seminal work on Kant, Rawls, and Habermas, the constructivist account of practical philosophy advocated by these thinkers aims at specifying "a procedure for critically assessing the legitimacy of social norms and institutions by reference to a normative conception of practical reason". Moreover, by elaborating on the main arguments of these versions of constructivism, the latter is shown to constitute a highly defensible "clarification of the normative grounds" of social criticism, whose justification is "ultimately reflexive or recursive in the sense that there can be no higher appeal to something beyond the idea of that to which free and equal persons can rationally agree."
"Justice as fairness," according to John Rawls, "is a theory of human justice and among its premises are the elementary facts about persons and their place in nature."(TJ 257) In effect, Kant seems to shift away from the non-demonstrable Faktum der Vernunft assumed in the second Critique towards a human practical reason in his later writings (notably MdS and political writings) so as to account for the tension between autonomy and heteronomy in the very "unsociable sociability" that characterizes human nature. Hence the modern problem of articulating ethics and political philosophy through a critical conception of human nature lies at the heart of both Kant’s and Rawls’s critique of metaphysical foundations. For Kant, the place of human persons in nature constitutes the counterpart to the Copernican revolution in theoretical philosophy (KrV). Rawls carefully contrasts a plausible interpretation of a Kantian intuitionism in the theoretical use of pure reason with the constructivism of his practical philosophy. According to Rawls, such is indeed the systematic, philosophical hallmark of Kant’s transcendental idealism, namely, that the unity of reason stands and falls with the two-world thesis. In order to avoid a foundationalist predicament, Rawls views the original position "as a procedural interpretation of Kant’s conception of autonomy and the categorical imperative"(TJ § 40) with the proviso that "the person’s choice as a noumenal self" is taken as "a collective one."(TJ 257) O’Neill has convincingly shown that Rawls’s shift from his use of the metaphor of "contract" in TJ to that of "construction" in PL deserves further investigation, particularly in its tension between an idealized, instrumentalized conception of practical rationality and a relativist, historicized view of practical reason anchored in a liberal, political culture. O’Neill points thus to a problematic "coherentism" supposedly grounding Kant’s practical philosophy, according to Rawls’s interpretation, especially as outlined in his Stanford paper on "Themes in Kant’s Moral Philosophy." It is within this problematic that Rawls resorts to the distinction between the conceptions of "reasonable" and "rational" so as to maintain the Kantian articulations of theoretical and practical reason, on the one hand, and of autonomy and heteronomy, on the other, accounting for human action in non-utilitarian, non-teleological terms. Later on, Rawls will resort to the same line of reasoning in order to reject Kant's moral constructivism insofar as the latter's view of autonomy fails to justify in political terms the overlapping consensus underlying the fact of reasonable pluralism in our liberal democracies. That would also amount to an attempt to respond to both Sandel's criticisms of a supposedly metaphysical self and the communitarian critique of Rawls's political shift in the later writings.
In effect, the criticisms raised by utilitarians and communitarians against Rawls’s theory of justice seem to converge on the correlated problems of maintaining, on the one hand, the Kantian priority of right (Recht) over the good --or the universalizable principle of justice over the principle of utility—and a model-conception of the person in a given political culture, on the other hand. Although the former problem appears to be dealt with in TJ while the latter is only formulated in PL, I argue that they complement each other and are essential to a correct understanding of Rawls’s constructivist theory of justice, as over against moral intuitionism and utilitarianism in both texts. It is thus my contention here that Rawls’s political constructivism can be better understood in the very terms of its critical account of the foundations of a theory of justice, and more specifically, in light of his critical appropriation of Kantian moral constructivism.
One of the greatest pretensions (and, to my mind, one of the greatest merits) of Rawls’s theory of justice is to provide us with an ethical-political conception of the normative foundations of social life. The theory of justice may be thus viewed as a universalizable procedure of construction capable of accounting for human sociability in constitutional, democratic societies, where claims to basic liberties and fair participation in social life allow for the pluralist coexistence of different, incompatible religious, philosophical, and moral beliefs. Therefore, at the very level of its foundation, the concept of justice is to practical philosophy what truth is to the theory of knowledge(TJ § 3).According to Rawls, "a conception of justice characterizes our moral sensibility when the everyday judgments we do make are in accordance with its principles."(TJ 46) The two fundamental principles (the Equal Liberty Principle and the Difference/Equality Principle) formulated by the Rawlsian theory of justice, as well as the original devices (dispositifs) of the original position and reflective equilibrium, are to be understood in this precise context of foundation--if not in the sense of a Kantian Grundlegung, at least as a Begründung--, insofar as they must be understood as formal-procedural rules capable of establishing normative criteria and determining results to be judged fair (beurteilen). Just as the sense of grammaticalness is presupposed in everyday practices of speaking one’s mother-tongue (or at least "functioning" in a certain language) and a rational faculty is presupposed in the conception of judgments and thoughts, the sense of justice and capacity for a conception of the good are inherent to the conception of moral persons, free and equal, living in a democratic society. "Justice as fairness," according to Rawls, seeks to unveil the fundamental ideas (latent in common sense) of liberty and equality, of an ideal social cooperation and the person. Rawls continuously reviews his theory of justice so as to better elucidate its foundations. In particular, Rawls addresses many of the questions raised by his critics as for his interpretation of Kant’s moral philosophy, on the one hand, and as for the arguments he uses against utilitarian conceptions, on the other. A "theory of justice" was already understood as a philosophical analysis of what justice is, avoiding both metaethical and substantive exclusive approaches to ethics. Thus, Kant’s moral philosophy, a refutation of eudaimonism, intuitionism, and utilitarianism, the rehabilitation of the concept of justice inherent to constitutional contractarianism--classical-liberal (Locke) and radical-democratic (Rousseau)--, the problem of constructivism, the question of the foundation of moral principles, the question of the just and the good (Aristotle)-- these and other related problems are all thematized by the TJ. At the end of the book (TJ § 87), Rawls reminds us that his conception of the foundations or justification of morals is to be distinguished from the two models that prevailed in the history of ethics, namely, the Cartesian deductive model (inferring a body of standards and precepts out of self-evident, moral first principles) and the naturalist model (definitions of moral concepts can be compared/reduced to nonmoral concepts). Rawls clings to the Socratic principle (TJ § 9) insofar as moral theory always brings us back to review our principles and judgments, and stresses that "justification rests upon the entire conception [of justice] and how it fits in with and organizes our considered judgments in reflective equilibrium."(TJ 579) Only then can we proceed to a "substantive theory of justice". By his implicit reformulation of a theory of society and of a theory of moral person, a theory of justice as fairness is supposed to strike us as being more defensible and more effective than any other version of contractarianism.(TJ 584) This is of course understood to be extended to any other theory of society. If human beings want to live in society and keep all their cultural, religious, and moral differences they should subscribe to such a theory of justice. My main contention here is that the question of its foundations (Rawls’s constructivism) underlies the entire development of concepts that structure the Rawlsian theory as a whole, being extended and more explicitly articulated with questions of political-practical order in later writings, notably in his lectures on "Kantian Constructivism in Moral Theory" (Columbia University, 1980) and in the volume on Political Liberalism. Despite all the criticisms and methodological problems that separate these works, I am stressing their continuity so as to highlight the question of foundations. Even when he seems to concede to some form of theoretical retraction, Rawls ends up reformulating an original concept or its first version in TJ so as to deepen the central theses of his theory of justice. For instance, in PL Rawls indicates straightforwardly that the major problem with TJ lied in the inconsistencies between the account of stability (part III, in part. § 76, the problem of relative stability) and the rest of his magnum opus. In other words, the question of articulating a well-ordered society, conceived as a regulative ideal of a society that seeks to promote its well-being through the public conception of justice, with a basic notion of an "association of social cooperation." In spite of his insightful remarks on the idea of social union in TJ § 79, the question of sociability remains problematic in Rawls’s conception of a hypothetical society strategically idealized in a form accepted by each citizen who also knows that all the others accept the same principles of justice, satisfied by the basic social institutions (TJ § 69). According to a theoretical-conceptual construction of the original position in TJ, the two principles are the only ones to be effectively chosen by the parties for the realization of society tout court, i.e., to account for the state of civil society. The problem, as reformulated by Rawls in PL, is to sustain this theory as an alternative to utilitarian and intuitionist conceptions of morality underlying our liberal, democratic societies. In effect, the alliance of these apparently opposed traditions (the liberal and democratic contractarian models, respectively upheld by John Locke and Jean-Jacques Rousseau) constitutes the political-philosophical platform common to Kant, Rawls, and Habermas. Rawls’s "Kantian interpretation" is precisely what marks him off from Habermas’s appropriation of the same principle of universalizibility. The democratic pluralism of liberal societies, as opposed to the ideal speech situation of discourse ethics, is regarded as a problematic starting-point by Rawls, not so much for the diversity of religious, philosophical, and moral doctrines, but for the pluralism of comprehensive doctrines that prove themselves incompatible for not being accepted by all involved parties --Rawls resorts thus to a distinction between the public and the nonpublic viewpoints (non-private). To be sure, Rawls does not distinguish between the moral and the political in TJ (cf. PL xv), the social contract being understood within moral philosophy: whatever is just always excels that which is better for society. A just well-ordered society must be founded in such a way that people will put up with all their religious, ethnic, and cultural differences, as free and equal persons who seek to live well.
TJ is later presented as a procedural construction, more precisely as a procedural device of construction capable of theoretically representing the two fundamental principles of political justice (PL III 93). The guiding idea of TJ (§ 3/p. 11) is after all that "the principles of justice for the basic structure of society are the object of the original agreement." As principles that regulate all subsequent agreements, pacts, and contracts, they specify the types of social cooperation and forms of government to be established. What is at stake, therefore, is to found the procedure that best accounts for the construction of a free and fair/ reasonable society. Just as the overall aim of TJ was, according to Rawls, to generalize and carry "to a higher order of abstraction the traditional theory of the social contract as represented by Locke, Rousseau, and Kant"(TJ 11), he will maintain later that his conception of the contract is not utilitarian but Kantian, at the very level of its foundations. "The principles of justice are also categorical imperatives in Kant’s sense"(TJ § 40), insofar as the CI procedure claims no validity on the basis of the agent’s desires, aims, inclinations, motivations, or intentions, but solely in virtue of the person’s nature as a free and rational being. According to Kant, "nothing in the world --indeed nothing even beyond the world-- can possibly be conceived which could be called good without qualification except a good will." The categorical imperative provides us with the principle of universalizability according to which the form, matter, and end of our moral actions are self-determined by the modeling of maxims upon the moral law itself, in a procedure that accounts for the equation of the will with pure practical reason. Nevertheless, critics of Rawls have questioned his Kantianism, on the basis of his indebtedness to contractarian and supposedly utilitarian arguments, esp. when dealing with issues of rational choice. The book on Political Liberalism recasts the main theses of TJ, so as to corroborate the public conception of justice as fairness in a society where political power is exercised in accordance with a constitution --whose essentials must be endorsed by all citizens in light of principles and ideals accepted to them as reasonable and rational. According to Rawls, "this is the liberal principle of legitimacy."(PL 136f., 216f.) The Kantian conception of practical reasonableness (constructivism) is thus combined with the contractualism of liberal-democratic tradition. The liberalism at stake is of course political and not economic, just as Rawls’s constructivism is said to be "political," and not metaphysical or moral. For Rawls, Kantian constructivism turns out to be the most appropriate normative model for the foundation of a theory of justice today. The foundations of a theory of justice must be understood in procedural terms, following the Kantian premises outlined in the Foundations of the Metaphysics of Morals, in the second Critique, in the Metaphysics of Morals, and in the political writings (esp. "On Perpetual Peace" and "Idea for a Universal History"). Rawls succeeds in recovering the political-constitutional dimension that ensures the juridical fulfillment (de jure) of moral conscience qua "fact of reason" (de facto), fundamental to Kant’s practical philosophy.
In order to elaborate on his theory of justice, Rawls recasts the Kantian foundations of ethics. Besides its refusal to ground the supreme principle of morality in a conception of the good or in a principle of utility, the Kantian procedure refuses the intuitionist thesis, according to which pure or sensible intuition or the experience of the senses, instincts, and emotions could found morals. In its broad definition, Rawls conceives of intuitionism as "the doctrine that there is an irreducible family of first principles, which have to be weighed against one another by asking ourselves which balance, in our considered judgment, is the most just."(TJ 34) On the one hand, Rawls maintains the moral-political correlation established by Kant, as well as the distinction between legality and morality. On the other hand, he seeks to recover the binding force of the principle of justice inherent to Kant’s appropriation of the liberal, democratic conception of contract, as a regulative idea of practical reason. According to Rawls, universalizability and the primacy of right over the good are precisely what allow for a nonmetaphysical, detranscendentalized formulation of the principles of justice in reflective equilibrium. It is in this sense that Rawls emphasizes the strictly political-philosophical character of the foundations of a theory of justice. By recasting Kant’s constructivism, Rawls aims at the normativity of practical reason in a contractarian context, ordered by a constitution and formed by free, morally equal persons, historically and socially conditioned --and not in a supposedly neutral standpoint. Rawls’s contractarianism combines thus the Lockean principle of tolerance with Rousseau’s general will, already appropriated by Kant. In the conception of Recht (justice/right) lies an articulation between the moral-rational Sollen and the political-constitutional Wollen capable of carrying out the "volonté générale" of the social contract. Hence the role of political philosophy lies in the construction (an idea-like abstraction) of a fair society ordered according to the rational principles of freedom that make us act out of duty. For Rawls, the watershed consists in Kant's contention that the moral foundation of the political cannot be done away with.
It must be recalled that for Kant, the object of anthropology is defined in terms of a knowledge of human rationality, where humans are conceived, at once, as beings endowed with reason that they are (animal rationabile) and as rational beings they ought to be reasonable(animal rationale)(Anth B IV, 673). It is very instructive that Kant’s conception of a human Sollen is found in the articulation of human nature in terms of rationality and historicity in such a manner that political philosophy is made dependent on moral philosophy. According to Kant, history is concerned with narrating the appearances of freedom of the will, or human actions, which like other natural events are determined by universal laws. Kant does not seek what gives the noumenal-phenomenical relation its foundation and justification on the same level of representations, but on a transcendental a priori that makes representations themselves possible. A new metaphysics had to emerge, as "a system of pure rational concepts independent of any conditions of intuition" (MdS 375), so that philosophical knowledge could deal with representations. Transcendental philosophy is precisely what accounts for the positivity of the social, juridical norms. For the transcendental field reveals a nonempirical, finite subject that determines in its relation to an object = x all the formal conditions of experience in general, making possible the synthesis between representations. While theoretical philosophy deals with the problem of representations and their determinations in the subject-object relations of knowledge, practical philosophy seeks to account for the self-determination of the human will and its representations. Such is therefore the legacy of the critical revolution that decentered metaphysical truth towards a region defined within the limits of human reason. By the methodic, systematic attacks of the Kritik on the pretensions to suprasensible knowledge, Kant set out to establish the true principles that constitute metaphysics as a science that makes possible legitimate knowledge of both nature and freedom. Hence, from the outset, Kant was led to draw the fundamental distinction between the theoretical and practical uses of pure reason, whether constituting or regulating the representations of its objects, respectively directed by the understanding (Verstand) applied to the cognition of nature or by reason (Vernunft) applied to the realm of freedom. Rawls uses thus the procedural representation of the categorical imperative (required by pure practical reason in the formulation of reasonable, universalizable maxims) so as to construct the content of a political conception of justice (in TJ, the two principles of justice are chosen by the parties in the original position) in order to represent their societal interests, although the original position is not itself constructed. However abstract as it sounds and actually is, Rawls's "Kantian interpretation" does justice to the latter's conception of justice at least insofar as the veil of ignorance is "thought up" (upbuilt) in the original position. There is no possibility of thinking justice without having not already presupposed such a device of representation. And yet Rawls's political constructivism fails to fulfill its anti-realist promises as the political thrust of his liberalism still refers us back to the moral conceptions of personality and correlates (sense of justice, concept of good, human freedom). Even the political view of liberal toleration seems to betray a definite commitment to moral principles --if not to a certain "morals by agreement". By way of conclusion, as Carlos Thiebaut put it, "Rawls’s contribution proves itself to be contractarian insofar as the contract theory is the Kantian theory."
1. Funded by the Brazilian National Research Council (CNPq), which also supported my participation at the IXth International Kant Congress at Humboldt-Universität, Berlin, where this paper was read in March, 2000. A printed version of this paper is found in Kant und die Berliner Aufklärung: Akten des IX Internationalen Kant-Kongresses, ed. Volker Gerhardt, Rolf-Peter Horstmann und Ralph Schumacher, Berlin: W. de Gruyter, 2001, p. 286-295. For an elaborated account of the Kantian arbitration between Rawls and Habermas, cf. my Tractatus ethico-politicus (Porto Alegre: Edipucrs, 1999), and the essay "Critique of Public Reason Revisited: The Rawls-Habermas Debate," Veritas 44/4 (2000): 583-606.
2. John Rawls, A Theory of Justice, Cambridge, Mass.: Harvard University Press, 1971; Political Liberalism, New York: Columbia University Press, 1993. Abbr. TJ and PL. For an overview of Rawls’s theory of justice, cf. my Rawls (Rio de Janeiro: Zahar, 2002).
3. Thomas W. Pogge, Rawls, München: C.H. Beck, 1994, p. 189-198.
4. Cf. Onora O'Neill, "Vindicating Reason", in P. Guyer (ed.), Cambridge Companion to Kant. Cambridge University Press, 1992, p. 280-308.
5. Kenneth Baynes, The Normative Grounds of Social Criticism: Kant, Rawls, Habermas, Albany: SUNY Press, 1992.
6. I am using the following abbreviations for Kant’s works: KrV, KpV, GMS, and MdS.
7. Onora O’Neill, Constructions of Reason: Explorations of Kant’s Practical Philosophy, Cambridge: Cambridge University Press, 1989, p. 206-218.
8. In Eckart Förster, ed., Kant’s Transcendental Deductions: The Three Critiques and the "Opus Postumum", Stanford: Stanford University Press, 1989, p. 81-113.
9. Cf. TJ viii: "The theory that results is highly Kantian in nature."
10. GMS, Trans. Lewis White Beck, Indianapolis: Bobbs-Merril, 1959, p. 9.
11. Contrato Social: Ontem e Hoje, ed. P. Krischke, São Paulo: Cortez, 1994, p. 284.