THE POST-KANTIAN CONVERGENCE

I take it that rationality, in the most general sense, is a matter of justification, evidence, argument, and truth (see, for example, Audi Reference Audi2001 and Brown Reference Brown1988). Specifically, a rational belief is a belief that purports both to describe some truth about how one or more things in the world really are—hence to offer a metaphysical claim in the broadest sense, rather than merely a claim about appearances—and to be justified in offering such a description where justification itself is understood as deriving entirely from some kind of evidence-based argument (i.e., a demonstration or proof) that obeys, above all, the principle of noncontradiction. A person is rational to the degree that he or she (1) has rational beliefs and (2) acts on the basis of those beliefs.

Rationalism. With such a formulation in mind, we may say that “rationalism,” broadly construed, has both a descriptive and an aspirational aspect. The descriptive aspect itself involves (at least) two distinct claims. The first is that humans have—uniquely among earthly creatures, according to many versions—the capacity to be rational. The essence of humanness lies in our ability to formulate and hold rational beliefs and to act on the basis of those beliefs. The second is that normal human activity really does reflect, in some nontrivial sense, the exercise of our rational abilities. We are not only capable of having and acting on rational beliefs—beliefs that are justified in virtue of being logically entailed by the available evidence—but, in the usual course of events, actually do have such beliefs and actually do act in that way. If philosophy, at its best, is the rigorous and systematically self-conscious pursuit of rational belief, and if the philosopher-king is the archetypical and mythical apotheosis of the rational person, we are all nonetheless participants, whether rigorous or otherwise, in a larger enterprise that may indeed be called, without irony, philosophical. The aspirational aspect of rationalism involves two distinct claims as well. On the one hand, the pursuit of rationality is typically confronted by a kind of intellectual challenge. Specifically, it is often difficult to determine exactly which beliefs are best justified by the available evidence. But the pursuit of rationality is also threatened by a kind of psychological challenge. We are appetitive as well as cognitive creatures—and are also subject, at the margins, to distortions associated with such things as the framing effect—and our desires and cognitive hiccups sometimes confound and interfere with either rational belief formation, the cultivation of suitably rational dispositions, or both. In effect, then, rationalism of whatever variety holds that we are indeed essentially and profoundly rational beings, but always imperfectly so, and that the quest to achieve a more rational life—to have and act ever more faithfully on beliefs that are ever more strongly justified, hence are ever closer to the truth—is a permanent and extremely important part of the human agenda.

In this context, I suggest that various and distinct concepts of rationality—hence, various and distinct forms of rationalism—differ from one another largely insofar as they embrace different notions either of what counts as evidence or of what counts as an adequate justificatory argument from evidence to conclusion; from this it follows that they embrace different notions of what counts as truth (or warrantability, validity, etc.). Here, I believe, is where contemporary criticisms of rationalism in politics get into serious trouble. Political rationalism just is the idea that political activity, like any other kind of human activity, is undertaken by individuals who have both the general capacity and the broad disposition to act in politics on the basis of rational belief and who actually do act in that way (though, again, they do so more or less successfully in the light of relevant intellectual and psychological challenges). As such, theories of political rationalism are apt to be as varied as theories of rationalism sans phrase. This fact seems to have been largely ignored by critics of political rationality, who often paint with a very broad brush indeed. In particular, and most important for present purposes, anti- or non-rationalist strains in recent political theory have evinced little interest in, and make virtually no reference to, a variety of enormously influential, indeed canonical arguments of twentieth-century philosophical and metaphysical thought, analytic and continental alike, that have produced, collectively, a new and deeply compelling notion of what it means to be rational and that provide thereby a potentially powerful but badly underexplored foundation for thinking about politics as a fundamentally rational enterprise.

The priority of conceptual schemes. Quine presents a central problem of modern philosophy—perhaps the central problem of modern philosophy—with characteristic concision: “How much of our science is merely contributed by language and how much is a genuine reflection of reality?” (Quine Reference Quine1950, 632). If we generalize this question in what I believe to be an entirely friendly way by replacing “science” with “knowledge of the world” and “language” with “thought,” we have arguably the principal agenda for latter-day metaphysical speculation, an agenda according to which metaphysics and epistemology are ultimately inseparable. Specifically, how can we persuasively assess the accuracy of our thoughts about the world if any such assessment cannot but be an instance of thought itself? Quine also offers, again with wonderful economy, the kind of response to which philosophers since Kant have gravitated. Specifically, “to answer the question we must talk about the world as well as about language, and to talk about the world we must already impose upon the world some conceptual scheme peculiar to our own special language.” Quine does not deny that we can “improve our conceptual scheme,” but he also insists that “we cannot detach ourselves from it and compare it objectively with an unconceptualized reality” (Reference Quine1950, 632).

Such an approach rejects as deeply misguided the traditional, dogmatic/metaphysical project of seeking rationally to describe the essential, universal, immutable, and well-ordered features of reality—the truth about things in themselves—as they subsist entirely separate from and independent of our engagement with them—a project that ignores or elides the very basic problem that any effort to evaluate and justify critically the arguments of human thought can be undertaken only from within the perspective of human thought itself because there is, for us, no other available perspective. The critique is indeed essentially Kantian, and I suggest that the first great and in some ways most characteristic elaboration of it is to be found in the opening pages of Hegel's Phenomenology of Mind. There, Hegel seeks to show that our engagement with the world presupposes a complex repertoire of basic notions and that each such notion entails—logically—some number of further and progressively more elaborate determinations. Thus, when I encounter a thing, I must think of it as “this” thing. I must treat it ostensively, and I must therefore already have, in advance, the category “this.” To think of it as “this” thing, however, is necessarily to distinguish it from a different thing, namely, the “this” that is me. Looking at or otherwise experiencing a thing empirically is perforce already to operate in terms of two “thises.” Implicit, then, in the sheer being (reinen Sein) of immediate sense perception—the most primitive and basic engagement with the world—are both the particular thing that I am and the different particular thing that is the entity (ein Dieser als Ich, und ein Dieses als Gegenstand) (Hegel Reference Hegel1952 [Reference Hegel1807], 80). Importantly, the imperative here is a matter of reason, of rationality. It would be unintelligible—it would make no sense, it would be incoherent and self-contradictory—to consider the entity without also considering, and positing the fact, that it is a different thing from me; for Hegel, this means that in some sense the distinction—the cognitive material—precedes the experience. Of course, such a distinction entails, in turn, any number of further discriminations. The thing is “there” while I am “here,” the thing is the “object” of thought while I am the thinking “subject,” and so on, ad infinitum. In this sense, the project of the Phenomenology as a whole is nothing less than the patient, step-by-step, rational reconstruction of an enormous and complex conceptual apparatus on the basis of which, and exclusively on the basis of which, we are able to have intelligent experiences of the world—a shared system of typically implicit metaphysical presuppositions without which, moreover, we could not even begin to communicate with one another and that comprises thereby the very foundations of social life.

We have here the germ—and, with Hegel, an elaborate and influential though also particular and contentious development—of a massively important and deeply compelling idea. This is the idea that our engagement with the world and the claims that we make about it are, in one way or another, products of an underlying intellectual scheme that is prior to any intelligible experience we might have. Schemes are prior in the sense of being presupposed in the very possibility of having an experience, and all thought about the world is undertaken internal to some such apparatus. This includes, of course, thought about the scheme itself. It is certainly true that particular claims about the world might be right or might be wrong. In employing our intellectual scheme we may make mistakes. But such mistakes could be discovered only through an internal critique (i.e., a process of rational reconstruction that emerges from the perspective of the scheme itself, according to which a proposition is ruled out if it fails to comport with the larger structure of propositions and conceptual claims on which experience is based). And because there are no intelligible thoughts that are not internal to some such preexisting structure of cognition, there is no possibility of an external critique. All critique is immanent.

Analytic perspectives. The range and variety of post-Kantian and, perhaps especially, twentieth-century thinkers who have adopted one or another version of this general approach is large and impressive. We have already seen an indication of Quine's position, and it is sufficient to note here that his account proceeds in an avowedly pragmatist direction. In contrast, and from a more specifically Kantian perspective, Strawson proposes what he calls a “descriptive metaphysics” and invites us to consider “the model of an elaborate network, a system of connected items, concepts, such that the function of each item, each concept, could, from the philosophical point of view, be properly understood only by grasping its connections with the others, its place in the system” (Reference Strawson1992, 19). Strawson suggests an analogy with grammar. A conceptual apparatus is a more or less orderly and coherent structure of interrelated parts much like the rules of a language; just as ordinary and reasonably experienced speakers of a language are able to follow those rules with astonishing skill and accuracy even as they do so unself-consciously, so too are the members of a culture able to employ with great efficacy a shared network of typically implicit conceptual materials that comprise, collectively, an account of how things in the world really are (i.e., a metaphysics) and that make social intercourse itself possible. According to this model, the task of the philosopher, like that of the grammarian, is to uncover—rationally to reconstruct—such materials, not simply to provide understanding and insight but also to help users navigate difficult cases; this task, moreover, is shared by ordinary and more or less thoughtful users themselves—again by analogy, one does not have to be a professional grammarian to look up a word in the dictionary—such that philosophy comes to be seen as merely a particularly rigorous and systematic version of thinking itself.

Strawson acknowledges that we might be tempted “to step outside the entire structure of the conceptual scheme we actually have and then to justify [our claims] from some extraneous point of vantage” (Reference Strawson1992, 64). We must resist this temptation, however, for the simple reason that “there is nowhere to step. . . no such extraneous point of vantage” (Reference Strawson1992, 64).Footnote ² This very same point is absolutely central, moreover, to Putnam's so-called internal realism, according to which “[t]ruth. . . is some sort of (idealized) rational acceptability—some sort of ideal coherence of our beliefs with each other and with our experiences as those experiences are represented in our belief system” (Putnam Reference Putnam1981, 49–50; emphasis in the original). Like Strawson, Putnam insists on the impossibility of finding an external perspective from which to judge the claims we make about the world: “[T]here is no God's-eye point of view that we can know or usefully imagine” (Reference Putnam1981, 50; see again Quine Reference Quine1950, 632). Even more strongly, “objects do not exist independently of conceptual schemes. . . . We cut up the world into objects when we introduce one or another scheme of description” (Putnam Reference Putnam1981, 52; also see Strawson Reference Strawson1992, 33–35, 66). Putnam does not doubt that there are experiential inputs, but he does explicitly “deny that there are any inputs which are not themselves to some extent shaped by our concepts” (Reference Putnam1981, 54).

Focusing on problems in the philosophy of mind, Searle operates very much within this broad perspective. Thus, an intentional state, by which Searle largely means the state of having a belief, “only determines its conditions of satisfaction—and thus only is the state that it is—given its position in a Network of other Intentional states and against a Background of practices and preintentional assumptions” (Reference Searle1983, 19; emphasis in the original). Particular beliefs, in other words, necessarily function within broad and socially shared structures of logically interconnected presuppositions. Brandom's so-called inferentialism is, if anything, even clearer in describing the notion of a conceptual apparatus as an interrelated, internally coherent system: “[I]n order to grasp any concept, one must master many concepts. . . . For grasp of one concept consists in mastery of at least some of its inferential relations to other concepts” (Reference Brandom2000, 29). Brandom is especially emphatic in describing what is at stake here, namely, the nature of what it means to be a thinking human being as opposed to, say, an animal or a machine: “At the very center of this account is its rationalism; the pride of place it gives to specifically inferential articulation, to playing a role in practices of giving and asking for reasons” (Reference Brandom2000, 22–23; emphasis in the original).

POST-METAPHYSICAL POLITICAL THEORY AND THE CRITIQUE OF RATIONALISM

Such formulations—and I ignore here, for reasons of space, a number of more or less equally important, closely related, and characteristically post-Kantian contributions by Davidson, Dummett, McDowell, and Sellars, among others—represent today something like a philosophical consensus concerning the larger parameters of rationality and rational judgment. In this context, to contemplate the idea of rationalism in politics would require nothing less than a systematic engagement with materials of this kind. But quite to the contrary, contemporary critics of political rationality have almost entirely ignored such materials and have, as a result, presupposed conceptions of rationality that I believe to be both outdated and seriously distorted. At the same time, they have produced conceptions of politics—rhetorical, agonistic, realist—that are themselves deeply dependent on important, if only implicit and unrecognized, elements of rational argument and metaphysical truth. This may seem paradoxical, but in fact it is no such thing, for much of their work unavoidably confirms, even as it seeks explicitly to deny, the fundamentally rational character of political thought and action.

Rhetoric and politics. In a stimulating, influential, and, I believe, highly representative contribution to the Arendtian understanding of politics, Zerilli demonstrates with considerable clarity and power the different directions that post-metaphysical political thought has been taken by, in particular, John Rawls and Arendt herself. On the one hand, both of them worry about “the dogmatic character of [truth] claims and their potentially corrosive effects on the public realm” (Zerilli Reference Zerilli2012, 8; for the broader contexts of my criticisms, see Steinberger Reference Steinberger2015). Indeed, just as Rawls sharply distinguishes the metaphysical from the political—a “political conception does without the concept of truth” (Rawls Reference Rawls1993, 94; but see, unsatisfyingly, pp. 150–51)—so does Arendt argue that “truth carries within itself an element of coercion,” involving the pursuit of evidence-based and rationally demonstrated propositions that would be, as such, “beyond agreement, dispute, opinion, or consent” and that would thereby preclude debate, which “constitutes the very essence of political life” (Reference Arendt1968, 240–41; see Estlund Reference Estlund2008, 21–22). On the other hand, Rawls invokes a “method of avoidance” (Zerilli Reference Zerilli2012, 8–9) that has as its goal setting rational limits on what can be discussed politically and that, as such, elides real disagreements among individuals holding a diversity of comprehensive doctrines; Arendt's conception seeks, quite to the contrary, to liberate political debate, “to animate or enhance the capacity to judge,” and to foster thereby the opportunity of citizens to engage as broadly as possible in public discourse (Zerilli Reference Zerilli2012, 8–9). In this sense, the Arendtian critique of rationalism in politics applies not only to traditional, metaphysically oriented forms of political rationalism but also to approaches, such as Rawls's, that may renounce the pursuit of metaphysical truth but nonetheless insist on connecting the legitimacy of political outcomes to the possibility of achieving some kind of rationally justified consensus: “[T]he attempt to restrict the public articulation of competing perspectives, comprehensive doctrines, or worldviews results not, as Rawls would have it, in a world that is shared but in a loss of what we have in common” (Zerilli Reference Zerilli2012, 22). Arendt herself celebrates the engagement of opinion with opinion wherein the goal is not demonstration in the logical or rational sense but instead a kind of “validity” that emerges from vigorous, passionate, and even courageous debate and that is “communicated by means of persuasion and dissuasion” (Reference Arendt1968, 247). In place of this, Rawls offers little more than “the charade of deliberative discourse” (Zerilli Reference Zerilli2012, 16).Footnote ³

Zerilli's formulation merits particular attention because it seeks to flesh out Arendt's view in concrete terms, specifically by working carefully through what she takes to be an exemplary case of Arendtian political speech, namely, Frederick Douglass's famous address of 1852 titled “The Meaning of July Fourth for the Negro.” This address is a stinging indictment of American slavery and a systematic account of how the principles and promises of the U.S. Constitution ring utterly hollow for mid-nineteenth-century African Americans. It has, on Zerilli's analysis, two principal features. First, it is explicit and resolute in insisting that, with respect to slavery, there is no need to engage in anything like systematic, proof-oriented, rationalistic discourse: “I submit, where all is plain there is nothing to be argued. . . . Must I undertake to prove that the slave is a man. . . . Must I argue the wrongness of slavery?” Any attempt rationally to demonstrate the iniquity of slavery would be, at once, superfluous, ineffective, and perverse. It is superfluous to spend time demonstrating a truth that is already obvious and known; it is ineffective to argue in front of an audience for whom argument has clearly and systematically failed; it is perverse to attempt to treat as a matter for inquiry and analysis something that is entirely and completely self-evident. Second, in place of argument and proof Douglass opts for a practice of rhetoric. According to Zerilli, his speech is a rich tableau of “scorching irony, biting ridicule, blasting reproach, withering sarcasm, and a barrage of powerful tropes and figures” (Reference Zerilli2012, 17), and anyone who reads the address will know Zerilli's characterization to be, in this sense, entirely accurate. Douglass makes no secret of the anger, moral outrage, and deep disdain that he feels for an American civilization that sustains slavery, and his presentation is nothing less than a brilliantly articulate and deeply moving expression of those feelings.

It is important to note that in embracing Douglass's address as representing the kind of politics of which Arendt approves, Zerilli argues not only that it differs from but, indeed, is also profoundly and explicitly critical of the kind of orientation that Rawls would come to adopt. Specifically, “Douglass's speech. . . aims to expose the hypocrisy of the canons of political rationality and deliberation” (Zerilli Reference Zerilli2012, 16). Certainly Douglass seems to be attacking just the kind of truth-oriented, metaphysically driven approach to political argument characteristic of traditional political rationalism. This would be evident, for example, in his explicit rejection of Garrisonian abolitionism on the grounds that it seeks to prove through rational argument the moral correctness of a viewpoint that does not need, and indeed is cheapened by the effort to provide, such a proof. But Zerilli wants to suggest, further, that the attack is germane as well to the Rawlsian notion of public reason, understood as an approach that eschews any and all explicit claims to metaphysical or moral truth but that endorses, nonetheless, the primacy of deliberative, argumentative rationality.

At first blush, such an account certainly seems plausible, and it is, without doubt, faithful to its Arendtian roots. The problem, however, is that the Fourth of July speech is, I think, actually quite different from how Zerilli characterizes it. Indeed, it seems to me that Douglass offers nothing less than a systematic, eminently rational argument against slavery that is rooted precisely in an explicit, evidence-based account of how things in the world really are. At the core of his speech, in other words, is a set of very fundamental truth claims about African Americans that demonstrate—rationally prove—the intellectual and moral bankruptcy of slavery.

In what I believe to be the material heart of his presentation, Douglass says the following:

Here we have, I would suggest, a virtually paradigmatic case of rational demonstration, of a kind that post-Kantian writers would very much recognize and endorse. Douglass is making explicit certain reasoned—indeed, metaphysical—commitments that all of his listeners cannot but share, commitments that they may not have articulated to themselves, but that are nonetheless firmly inscribed thousands of times over in their own lives and in all manner of ordinary, everyday action and experience. Those commitments entail a simple deductive argument: Implicitly or otherwise, we believe it to be true that (1) African Americans are human beings like any other, (2) human beings should not be enslaved, and (3) as a logical consequence, African Americans should not be enslaved. The argument is virtually identical to, for example, an argument about German anti-Semitism in the era of Nazism made by Barbara Herman, a contemporary analytic philosopher entirely committed to rational demonstration: “It is not as if individual Nazis were in no position to see (because of impoverishment of culture or upbringing, say) who was and who wasn't a person, or didn't know (because they were moral primitives, perhaps) what kinds of things it was morally permissible to do to persons” (Reference Herman1993, 91; in this connection, see also Section 156 of Nietzsche's Beyond Good and Evil). Such arguments reflect and represent, I suggest, very much the kind of metaphysical discourse that post-Kantians describe when they contemplate rationality as essentially a matter of adducing the logical consequences of concepts and truth claims to which we are already committed, in advance, so to speak, of explicit, self-conscious reflection.

The fact that Douglass feels no need to trace out in detail each step of his deduction does not deny in any way its status as a proof. Slavery is incoherent. Its incoherence is a rationally demonstrable consequence of the structure of metaphysical and moral presupposition to which the audience—and, indeed, the society itself—is committed, commitments that are themselves regularly and routinely embedded in the infinite patterns and practices of daily life. The development and persistence of slavery are thus evidence of a culture badly out of touch with its own deep-seated understanding of how things in the world really are. None of this is to doubt that Douglass's speech is also a deeply rhetorical document. But among the many rhetorical devices that it wields, perhaps the most effective is precisely some version of paralipsis in the roughly Ciceronian sense, according to which, in this case, the orator insists on the irrelevance of rational argumentation even as he himself makes rational arguments.

To be sure, Zerilli does not deny that Douglass makes arguments. Indeed, she says explicitly that he uses rhetoric to make his arguments “come alive” (Zerilli Reference Zerilli2012, 17). In saying this, she may appear to be endorsing a view that no proponent of rationality and truth seeking in politics, at least since Aristotle, would deny, namely, that the skillful use of rhetorical tools can be of great practical value in getting truth claims accepted by the body politic. Along these lines, several authors have recently described some of the ways in which certain kinds of rhetorical practice might be useful in supplementing or strengthening argumentation that appeals to reason (see, variously, Abijadeh Reference Abizadeh2007, 453–66; Beiner Reference Beiner1983; Bohman, Reference Bohman1996, 205; Chambers Reference Chambers2009, 335; Dryzek Reference Dryzek2010, 328–32; Garsten 2007; Miller Reference Miller1995, 57–58; Schnapper Reference Schnapper1994, 80; Tarnopolsky Reference Tarnopolsky2007; Viroli Reference Viroli1995, 174–75; Yack Reference Yack2006; Young Reference Young, Benhabib and Cornell1987). Zerilli, however, seems resolutely to distance herself from any such notion. She explicitly indicates that rhetoric “should be understood here as far more than mere persuasion that makes it easier to absorb rational truth, as the philosophical tradition would have it.” To the contrary, it is “the very basis of rational thought. . . [providing] the framework. . . within which the proof can come into existence at all” (Zerilli Reference Zerilli2012, 17). The specific force of this latter remark is, admittedly, not immediately clear; it seems to be connected to what I regard as a tendentious reading of Wittgenstein (see the discussion below). But the overall standpoint is certainly intended to be deeply incompatible with any notion of using rhetoric merely or even mainly to reinforce—to make more attractive—the claims of rationality. Indeed, attempts to carve out a role for rhetoric in argumentative, truth-oriented discourse, of the kind I have just described, generally fail to raise serious doubts about the primary—one might say the executive—role of rationality in political endeavor. Against this, Zerilli's view is that the practice of rhetoric, understood broadly as an exercise in the free, artful, and impassioned exchange of opinion, should largely replace, rather than merely serve the discourse of proof and demonstration.

Clearly, the claim to have found in Douglass's oratory a “mode of public speaking that. . . is not based upon reasons and evidence after discussion and due reflection” (Zerilli Reference Zerilli2012, 18) seems to me unsustainable. The problem, however, is not merely exegetical. To the degree that the doubts I have expressed are generalizable, they raise serious questions about the viability of any project that seeks, whether empirically or normatively, to separate politics from the rational pursuit of truth.

Agonistic democracy. The putative rejection of rationalism in politics is, if anything, even more emphatic among certain authors associated with the so-called agonistic approach to democracy, as exemplified perhaps most notably by Mouffe. In her view, contemporary democratic thought is profoundly ill equipped to understand and deal with problems of modern democratic society, and the principal reason for this failure is nothing other than “the rationalistic framework which informs the main currents of political theory” (Mouffe Reference Mouffe2005, 60). Indeed, the limitations of rationalism are such that present-day liberalism is utterly and entirely “unable to grasp the nature of political life” (Reference Mouffe2005, 60; see also Honig Reference Honig1993, 3–5). As with Zerilli, important targets here are Rawls in particular and deliberative democracy in general, both of which are understood, in the first instance, as critiques of the aggregative view of democracy associated canonically with Schumpeter (see, for example, Green Reference Green2010 and Shapiro Reference Shapiro2003). For Mouffe, Rawls presents “a form of rationality” that privileges abstract “rational consensus” over concrete, practical agreement (Reference Mouffe2005, 85–87). In so doing, he largely ignores the fact that “the domain of politics—even when fundamental issues like justice or basic principles are concerned—is not a neutral terrain that could be insulated from the pluralism of values and where rational, universal solutions could be formulated” (Reference Mouffe2005, 45). But Mouffe's argument is aimed equally at Habermas. Indeed, she claims that, with regard to the question of political legitimacy, there are between Rawls and Habermas “no fundamental differences” (Reference Mouffe2005, 89). The problem of the relationship between Rawls and Habermas is, to be sure, difficult and complex, and one may well have doubts about Mouffe's characterization of it. Her larger point, however, is clearly to reject rationalism in politics conceived in the broadest possible terms, which would presumably include notions of deliberation in both their Rawlsian (as with Cohen Reference Cohen and Elster1998) and Habermasian (as with Benhabib Reference Benhabib and Benhabib1996) iterations. Most importantly for our purposes, Mouffe claims that “the solution to our predicament does not reside in replacing the dominant ‘means-ends rationality’ by another form of rationality. . . . [S]imply replacing one type of rationality by another is not going to help us address the real problem” (Reference Mouffe2005, 95). The very idea of connecting political endeavor with rational belief and rational action, however formulated, is for her a massive mistake.

Mouffe thus follows the work of, among others, Schmitt in conceiving political activity as something sui generis. An emphasis on reason, truth, and rational argumentation “misses the specificity of the political” (Mouffe Reference Mouffe2005, 46; also Honig Reference Honig1993, 13–15) by formulating the problems of politics in decidedly nonpolitical terms. Post-metaphysical theories of politics generally seek to carve out a uniquely political area of human endeavor. But Rawls's version of this sphere—a realm of “public reason” governed by the norm of reasonableness—is criticized for being hostage to, variously, economic or ethical modes of thought that are understood to be inherently rationalistic. Thus, for Arendt and Arendtians such as Zerilli, politics is a distinctive realm of spontaneity and speech, plurality and persuasion, unbound, as we have seen, by the coercive claims of rational, evidence-based, truth-oriented argument; in contrast, Honig (Reference Honig1993, 3–5, 149–56), in roughly agonistic terms, highlights the “remainders” and “disruptions” of politics that are ignored by “virtue” theorists such as Rawls, and Connolly Reference Connolly(1983) focuses on, among other things, the inherently irresolvable nature of certain “essentially contested concepts” of political consequence. Mouffe's own formulation is different from these in emphasizing, instead, something akin to Schmitt's famous account of friend and enemy.Footnote ⁴ Schmitt, of course, sharply distinguishes the political friend-enemy motif from the kinds of conflicts that one finds in, for example, the economic marketplace or the military battlefield (Schmitt Reference Schmitt1996 [Reference Schmitt1932], 28–29, 34–37). Again, the political is sui generis, and in Mouffe, this manifests as a distinction between a kind of non-political “antagonism” in which the goal is to destroy the enemy and a kind of productive “agonism” mobilized in part by “collective passions” or emotions, in which one's opponent is not an enemy but an “adversary” (Mouffe Reference Mouffe2005, 102–3) with whom one can negotiate, however provisionally, something like a modus vivendi (see McCabe Reference McCabe2010). Here, Mouffe echoes Connolly, for whom the agonistic connection, properly understood, is “a social relation of respect for the opponent against whom you define yourself even while you resist its imperatives and strive to delimit its spaces of hegemony” (Connolly Reference Connolly1997, 155; see, more generally, Connolly Reference Connolly1991, 72–73, 178–79).

I would suggest that Mouffe's attack on rationalism, like Zerilli's, invokes a set of caricatures that are systematically belied by many or most important forms of contemporary post-Kantian philosophy. The full extent of these caricatures is explored in the last section of this essay. But it is also important immediately to see that her account, again like Zerilli's, actually fails at several key junctures to sustain its own nonrationalist ambitions. Mouffe insists, for example, that agonism “does not mean accepting a total pluralism” and acknowledges that “some limits need to be put to the kind of confrontation which is going to be seen as legitimate in the public sphere” (Reference Mouffe2005, 93). She elaborates by emphasizing that any such limit should be “political,” rather than moral or rational, in nature, but it is hard to see how this formulation could be persuasive. If the political just is a realm of opposition among a plurality of elements, then any constraints on who can participate and how—limits on “kinds of confrontation”—cannot emerge from the political itself. If, in contrast, the political realm is a realm of limited opposition among a specific plurality of elements, then politics cannot simply be agonistic, but must be underwritten by non-agonistic principles of one kind or another, and it is unclear how such principles could be formulated without reference to some type of reasoned process of justification. Elsewhere she indicates that her position “does not necessarily entail accepting a relativism that would justify any political system” (Reference Mouffe2005, 62, emphasis in the original). Indeed, she insists that “political judgment would not be made irrelevant, since it would still be possible to discriminate between just and unjust regimes” (Reference Mouffe2005, 62–63). It seems unlikely, though, that agonism in and of itself could maintain such a distinction; differentiating just from unjust would seem to require a systematic moral argument or demonstration of some kind. Mouffe explains that agonism, unlike rationalism, rejects the idea that there is or could be any one answer to the question of what is a just political order; there are, rather, a plurality of answers (Reference Mouffe2005, 62). But it would be difficult to find an exponent of rationalism in politics who would deny this—even Plato wrote the Statesman and the Laws—and the very notion of a plurality of just regimes presupposes defensible and demonstrable criteria of justice external to the regimes themselves.

Indeed, at virtually every point Mouffe's agonism struggles to deny the claims of reason, argument, and truth even as it invokes them, however tacitly. This, of course, is exactly what the post-Kantian convergence would expect, namely, an unavoidable reliance on a set of rational beliefs, often only implicit, about how things in the world really are. Mouffe, for example, says that if a power has been able to impose itself on a society, this can only occur “because it has been recognized as legitimate in some quarters”; she adds that legitimacy itself is always “based on some form of successful power” (Reference Mouffe2005, 100). But at the same time, she insists on the “obvious” truth that not all power is “automatically legitimate” (Reference Mouffe2005, 100). Certainly she cannot have it both ways. Either agonism in practice is viciously circular—legitimacy is a function of power and power a function of legitimacy—or else it relies on a tacit, implicit criterion of legitimacy that could not but reflect some kind of reasoned and putatively justified moral standpoint. Mouffe's distinction between enemy and adversary presupposes that we recognize an adversary's, but not an enemy's, “right” to defend his or her views (Reference Mouffe2005, 102). But what could it mean to invoke a right—rather than, say, a useful opportunity—that lacks any kind of inherent rational justification? She says that agonism involves both “a shared adhesion to the ethico-political principles of liberal democracy” and, at the very same time, disagreement “concerning the meaning and implementation of those principles” (Reference Mouffe2005, 102). But if we do not agree about the meaning of “X,” what sense could it make to say that in endorsing “X” we are endorsing the same thing? At the very least, a shared commitment to principles of liberal democracy must mean that practical agonism would rest on an argumentative structure of some kind—a system of presumably rational claims about right and wrong. If those claims do not decide all questions of public consequence, if they only set parameters within which particular decisions about institutions and policies must fall, this seems to me no different from what virtually all rationalists have argued in their different ways, from Plato and Aristotle to Habermas and Rawls.

Neo-realism. Yet a third perspective to reckon with reflects an arguably broader focus on the basic and constitutive facts of everyday political life. Such a view regards the activities of the state as, above all, a series of mechanisms for identifying and managing social and political conflict. As with Arendt, Rawls, and Habermas, the presupposition is that any political society of any consequence will comprise a plurality of groups and individuals having a diversity of competing ideologies, interests, and goals. Here, however, the political life of the state neither is nor could be driven primarily by deep considerations of justice, authentic self-disclosure, or any other abstract principle, just as it neither is nor could be driven primarily by considerations of rational coherence or truth. Rather, politics is the activity of dealing with “the struggles which result from the collisions between human purposes” and involves, as such, “the immense array of expedients and practices which human beings have invented to co-operate, as much as to compete, with one another in pursuing their purposes” (Dunn Reference Dunn2000, 133). Political interaction is understood to be essentially a matter of who does what to whom and for whose benefit; to think politically is, above all, “to think about agency, power, and interests” (Geuss Reference Geuss2008, 25; see also, Williams Reference Williams2007). The political theorist who would intelligently engage these kinds of materials is thus not much different from the traditional political scientist for whom politics is a matter of “who gets what, when, how” (Lasswell Reference Lasswell1936).

Such a perspective may well be Hobbesian in inspiration (Dunn Reference Dunn2000, 19–46; Geuss Reference Geuss2008, 22; Runciman Reference Runciman2010, 42–44; Williams Reference Williams2007, 4). As with Arendtian and agonistic perspectives, it “wants to give greater autonomy to politics (from morality, economics, etc.) as a discrete sphere of human activity” (Sleat Reference Sleat2014, 315; also Rossi and Sleat Reference Rossi and Sleat2014, 689; Williams Reference Williams2007, 3). In so doing, however, it seeks accurately to reflect and be responsive to the on-the-ground facts of political life. It pursues a form of practical realism. Laws are to be understood as useful stipulations resulting from processes of conflict, negotiation, and compromise. They are devices of convenience that generally lack, and certainly need not have, any rational/argumentative ambitions whatsoever. To the contrary, policies merely reflect efforts to manage disagreement; hence, they represent essentially pragmatic considerations—considerations of what works in a particular circumstance. None of this would commit anyone to any sort of higher proposition about what is or is not true or even rational. In these terms, moreover, Rawls is, once again, a particular target, because Rawlsian thought is, among other things, “naïve” in failing to accommodate the most basic facts of political life, including and especially the fact of power (Geuss Reference Geuss2008).Footnote ⁵

It is true that the recent explosion of literature on realism in fact comprises a range of different positions. For some, the separation of politics from the larger realm of (“pre-political”) moral and philosophical discourse is virtually absolute, whereas for others there “is a place for morality in politics” (Rossi and Sleat Reference Rossi and Sleat2014, 690). Certain authors, following Williams, focus heavily on the uniquely political value of legitimacy (Rossi Reference Rossi2012, 156–58), whereas others would go beyond legitimacy to include such things as stability, freedom, equality, toleration, and even justice (e.g., Sangiovanni Reference Sangiovanni2008). One perspective concedes that theorizing about politics (as opposed to actually doing it) might indeed be best understood as a philosophical, nonpolitical enterprise, whereas another sees the practice of political theory as inherently political, such that, for example, it may be necessary for the political philosopher to hide the truth or even lie about it (Jubb and Kurtulmus Reference Jubb and Kurtulmus2012). Despite such differences, however, realist critics broadly agree in rejecting the rationalist understanding of politics as governed by the aims and outcomes of truth-oriented philosophical analysis and in claiming, to the contrary, that political endeavor does and should operate according to “practice-dependent” standards and exigencies that are peculiar to it.

There is no doubt that such an approach can, and routinely does, provide powerful and compelling accounts of many important features of the political process. As a critique of rationalism in politics, however, its relevance is difficult to discern. Among other things, it seems not to acknowledge or accept a crucial distinction between motivations and intentions. Writing about speech-acts and the use of language, Skinner—whose view of the history of political thought has connections to realism—argues that “[t]o know about intentions is to know about such facts as whether the writer was joking or serious or ironic or in general what speech-act he was performing. To know about motives is to know what prompted those particular speech acts” (Reference Bourdieu1972, 393–408). The point is broadly applicable to all acts, not just speech-acts. To describe a motive is to identify part of what “efficiently” produced the act. It is to engage in a causal analysis. A motive is a “contingent antecedent condition”—economic, sociological, psychological, and the like—that helps explain an action in the sense of identifying at least one of the factors that brought it into being. An intention, in contrast, describes in part the particular goal of the action, its purpose or function, that which it is designed to do. As such, an intention is “a feature of the [action] itself” (Skinner Reference Skinner1972, 401). It helps explain the action not by identifying the factors that brought it about but by describing an important part of its character, its inherent logic. When I punch you in the nose, the action must be understood as something designed—intended—to punish or inflict pain or otherwise do damage. But the question of why I did this, what prompted me to punch you in the nose, is an entirely separate matter. The former describes (in part) what the action itself actually, essentially is. It provides, so to speak, an interpretation. The latter describes (in part) not what the action is, but what caused it to occur.

Rationalist approaches argue that politics is essentially a matter of establishing laws and policies whose authority is underwritten by some kind of reasoned argument. Such an argument might reflect a set of shared metaphysical claims about how things in the world really are or, to the contrary, might represent merely the fruits of an impartial, fair, and avowedly non-metaphysical process of deliberation. But in either case, the goal is not mere agreement but rational consensus. Efforts of this kind are thought by the rationalist to describe the basic intention of all political activity; hence, to describe its fundamental, constitutive character. In the same way that a punch in the nose just is the attempt to inflict (in a certain way) pain and damage, so is political activity the attempt to arrive at coherent and argumentatively justified answers to difficult questions of politics. In this sense, the rationalist does not at all deny what no one could deny, namely, that the activities of lawmaking and policy making are often, perhaps typically, motivated (see Geuss Reference Geuss2008, 9 on “real motivation”) by the need to manage interest-driven conflict. Yet an account of motives in itself says nothing about the characteristic form and structure of the particular instrumentalities in virtue of which such a goal is to be achieved. The desire to discover conflict-managing arrangements is one thing; the underlying logic of such arrangements quite another.

Now it is true that any comprehensive account of nose punching would require us to pay attention not only to intentions (and motives) but to techniques as well. How—from what angle, with what force, etc.—are punches actually thrown and to what effect, and how might they be thrown better in the future (vis-à-vis, e.g., possible defensive tactics)? But note that every such analysis would be hostage to—would be literally unintelligible absent—the underlying intention to break, for whatever reason, your nose. And so too for politics. A comprehensive account of lawmaking and policy making would obviously look at strategies and techniques of power and influence, which means examining causes and effects—whether actual or prospective—in the empirical world. If, in studying power, motive, and strategy, the orientation is theoretical and general, we call this political science; if practical and particular, the art of the possible. But in either case, my claim is that any such account, theoretical or practical, necessarily presupposes an underlying conceptual logic of intention—hence of rational argument—that underwrites, informs, directs, and shapes, however tacitly, the exercise of power.

Theorists of political realism seem to suggest that the status of laws and policies as articles of convenience actually does constitute their essence. If an arrangement works, then that is all we need to know about it; that is all it really is. The problem, however, is that not just anything can function as a useful and effective stipulated agreement. The substance of agreement never arises simply out of thin air. It always reflects, in post-Kantian terms, judgments and presuppositions that are inherent in a shared universe of discourse. This is certainly not to say that laws and policies perfectly reflect claims of reason to which everyone is committed. It is obvious that we do not always agree about what is true or rational, or perhaps more accurately, we do not always recognize our agreements—and putting agreements into practice is itself a highly imperfect enterprise. Nor is it to deny that some outcomes do indeed defy any kind of logical accounting. When an entirely unrelated rider is attached to a budget bill in the U.S. Congress, we might well say that pure expediency is at work. But the fact that such tactics are often criticized—often quite successfully—precisely for being incoherent is testament to a general and broad-based belief that public policy ought to be underwritten, in the end, by good reasons. Again, such a notion speaks to the very core of political rationalism according to which the goal of political endeavor—the intention, though not necessarily the motive—is to generate outcomes that are justified on the basis of reasoned argument.

In this respect, we can see that both the virtues and limitations of rhetorical and agonistic perspectives are quite similar to those of realism. Attention to the importance of rhetoric in politics may, and often does, permit a deeper understanding of motive and strategy. Thus, Frederick Douglass's Fourth of July speech is indeed driven by the most profound sense of outrage, and it pursues a strategy of persuasion that is hardly reducible to the protocols of systematic, truth-oriented analysis. But the underlying intention is, again, to utilize such materials in the service of propositions that purport to describe how things in the world really are and that reflect, however tacitly, a shared if only implicit structure of rational, evidence-based argument, very much as contemplated by post-Kantian philosophy. Similarly, the great contribution of agonism is to remind us of the centrality of conflict for politics, to uncover the distinctive (e.g., non-antagonistic) “logic” of such conflict, and to show how structures of power, desire, and strategy may distort or frustrate the pursuit of rational truth. But none of this has any force against the larger claim that politics is, in the first instance, constituted by the intention to seek and adopt policies or decisions that faithfully reflect the conclusions of systematic, reasoned discourse, whether actual or prospective.

It is important to note along these lines that the large majority of laws or policies, proposed and enacted alike, are accompanied more or less explicitly by some structure of argumentative justification that goes well beyond mere conflict management. It is hard to imagine the political proposal or even the political compromise that does not express or imply one or another set of substantive (i.e., not merely strategic) reasons for why it should be adopted. Of course, if an argument is to be at all persuasive—if a reason is to function as a reason—it needs to make sense; but making sense means, minimally, comporting with and being underwritten by some shared set of broader understandings about what is and is not a good justification. If nothing else, a more or less common system of rational consensus always establishes parameters that determine the kinds of things that could function effectively in managing our disagreements; this means that most or all such mechanisms, as reflections of social action, cannot but be embedded in and reflective of an underlying universe of discourse composed of claims about what is and is not coherent.

TOWARD A RATIONALISM IN POLITICS

From such a perspective, we can see that a rationalist understanding of politics informed by post-Kantian metaphysics would have, like rationalism in general, both a descriptive and an aspirational aspect. Descriptively, it might well accept, indeed embrace, many of the most important insights of rhetorical, agonistic, and realist theories regarding the nuts and bolts of political struggle while nonetheless insisting that such struggle is invariably underwritten by the idea of discovering and implementing, in any particular case, the demonstrably best course of action. This is only to propose, as a matter of actual fact, that political endeavor involves, at its core, the simple but virtually ubiquitous practice of asking whether one proposed policy is indeed better than another and seeking to answer that question by adducing good reasons based on a structure of sound, evidence-based argumentation—a practice to be found, variously, on the campaign trail, in the halls of the legislative assembly, among the deliberations of senior councilors, within the offices of bureaucratic administration, at the bar of the appellate court, and so on. With respect to aspiration, then, the political rationalist would understand that posing and answering any such question—and implementing the results—will inevitably be challenged by those elements of motive and strategy, conflict and power, rhetoric and partisanship that theorists such as Zerilli, Mouffe, Geuss, and many others have analyzed to telling effect. The difference is that the rationalist would neither celebrate nor prescribe nor give pride of place to those factors, but rather would understand them as important and persistent features of political life that help explain why the formative, definitive, and ever-present aspirations of political reason are never perfectly realized.

To reject all of this in the name of a hard-headed, putatively non-metaphysical realism is, one might suggest, to be guilty of what Oakeshott (Reference Oakeshott and Oakeshott1962) calls “empiricism.” According to Oakeshott, the empiricist thinks of political activity as a matter of “politics without a policy” entirely driven, indeed, by the desire to arrive at useful mechanisms of mutual accommodation among self-interested parties in which theories, presuppositions, and judgments—metaphysical commitments—are thought to play no important role. In Oakeshott's view, such an approach simply fails to understand politics as it really is, namely, as a “concrete manner of activity” (Reference Oakeshott and Oakeshott1962, 114–15) involving, above all, “the amendment of existing arrangements by exploring and pursuing what is intimated in them” (Reference Oakeshott and Oakeshott1962, 124)—a pursuit that is governed, in the end, by considerations of coherence. If Oakeshott is deeply hostile to the “rationalism” of traditional political truth seeking, I would suggest that his approach is nonetheless profoundly consistent with a kind of post-Kantian—in Oakeshott's case, neo-Hegelian—conception of human reason understood as a socially located process of rational reconstruction. The pursuit of intimations is precisely the activity of making (more or less) explicit some set of shared, “pre-political” presuppositions about how things in the world really are, presuppositions to which we are already implicitly committed and that constitute, as such, a coherent structure of thought. Oakeshott makes it quite clear, moreover, that his view is intended to be descriptive and not simply normative: “[F]rom a theoretical point of view, purely empirical politics are not something difficult to achieve or properly to be avoided, they are merely impossible; the product of a misunderstanding” (Reference Oakeshott and Oakeshott1962, 115).

Avoiding such a misunderstanding would require, I am proposing, a close examination of the relevance for political thought and action of post-Kantian philosophy, broadly construed.

Five theses. The writings of authors such as Quine, Strawson, Putnam, Searle, and Brandom are canonical texts of late twentieth-century analytic philosophy that have given rise to, and have established the parameters for, large complex literatures involving all manner of discussion and debate. Collectively they have set the tone for much philosophical inquiry of the last half-century and have given us notions of rationality and of the role that rationality plays in the life of the mind, far different from what one finds in the various dogmatic traditions of pre-Kantian thought. As such, their work provides, I suggest, important foundations for any attempt at thinking seriously and systematically about the nature of political thought and action.

At the risk of oversimplification, I identify five major themes that should be of particular relevance to politicsFootnote ⁶ :

First, in our engagement with the world, the mind is best understood—I borrow this metaphor from Brandom (Reference Brandom2000, 8)—not as a mirror but as a lamp. “Reality,” including the most basic reality of Hegelian sense-certainty, is always an interpretation. It is constructed by humans wielding a cognitive apparatus—a structure of thought—that we already have, independent of the world itself (for example, Strawson Reference Strawson1992, 21). In this respect, the analytic themes pursued here resonate with any number of motifs emerging from a presumably very different set of traditions. Consider, to pick just one example, Bourdieu's signature notion of the habitus, presented as an acquired system of “generative schemes” (Reference Bourdieu1972, 199). Such schemes are understood precisely as classificatory or practical taxonomies that “confer on our [everyday] works and practices. . . the regularity and at the same time objectivity that defines their specific ‘rationality’” (Reference Bourdieu1972, 180, emphasis in the original). The parallel here with, say, Strawson, is palpable. Along these lines, moreover, we should also make note of Heidegger himself, who constantly reminds us that our engagement with the world—with its “equipment” or “stuff” (Zeug)—is never simply naïve. With respect to spatiality, for example, he says, “I necessarily orient myself in and from already being alongside a ‘familiar’ world” [je schon sein bei einer ‘bekannten’ Welt] (Heidegger Reference Heidegger1967 [1927], 109). Indeed, the fact that “I am already in a world is no less constitutive for the possibility of orientation than is the feeling for right and left” (Reference Heidegger1967 [1927], 109); Heidegger adds that “the ontologically well understood ‘subject,’ Dasein, is spatial [räumlich], and because Dasein is spatial in the way described, space shows itself as a priori” (Reference Heidegger1967 [Reference Heidegger1927], 111). In saying this, he is actually providing, I suggest, a version—mutatis mutandis, to be sure—of Brandom's claim that “one must already have concepts” in order to experience the world intelligently (Reference Brandom2000, 26; see Strawson Reference Strawson1992, 84). For Heidegger, as for the analytic philosophers to whom I have referred, the mind, far from being a tabula rasa, is the very opposite—a rich and complex repository of pre-understandings. Some such view is apparent in authors as diverse as Oakeshott (Reference Oakeshott and Oakeshott1962, 119) and Wittgenstein (Reference Wittgenstein1958, 15–16) himself; indeed, the motif according to which “one already [schon, déja] knows”—as in one always has the relevant knowledge of something in advance of experiencing it intelligently—is a commonplace across analytic and continental traditions of twentieth-century philosophy. It is true, of course, that theorists often differ sharply about how the idea of an already existing apparatus should be understood—as a structure of concepts, a set of beliefs, a hard-wired network of mental dispositions, a horizon, a system of traditional prejudice, a background of intellectual capacities, an inferential web of premises and conclusions, a universe of discourse, or Geist. But there is nonetheless broad agreement that some such structure, whatever it is, necessarily imposes itself in some way on raw data, shapes those data into something intelligible, and provides thereby the basis for experience in the full sense of the word.Footnote ⁷

Second, cognitive activity (i.e., the process of actually wielding conceptual materials with a view toward achieving something that we can call truth—or, variously, warrantability, correctness, validity, or knowledge) is always, at base, a matter of discovering connections among conceptually laden claims. We are dealing here with a species of holism (Brandom Reference Brandom2000, 15; Searle Reference Searle1983, 21; Strawson Reference Strawson1992, 21, 24, 84–85; see Heidegger's notion of a “totality of mutual engagements” (Bewandtnisganzheit) (Reference Heidegger1967 [Reference Heidegger1927], 85; also at, for example, p. 80 regarding the situatedness of signs; and consider, again, Oakeshott's notion of experience that is “concrete”). Thoughts about the world are necessarily bound up with other thoughts about the world, thereby composing a system of some kind. From these considerations, moreover, political thoughts are not at all exempt; indeed, there is no clear reason why they should be. And so, as we have seen, Fredrick Douglass's claims about slavery, despite their rhetorical richness, point to and reflect—and are justified, however tacitly, on the basis of—the larger network of widely shared beliefs and meaningful experiences that make those claims intelligible in the first place. Similarly, claims that reflect the putative value and limits of political agon, as well as practical proposals regarding real-world forms of accommodation, negotiation, and policy, are always hostage to and embedded in one or another system of metaphysical presupposition. The task of philosophy in particular and critical thought in general is primarily to discover—to uncover—as much of the system as possible.

Third, this very task presupposes that a large part of any cognitive apparatus will be implicit. It will be composed of claims, propositions, assertions, and presuppositions about how things are in the world that constitute, in principle, genuine knowledge but that are, much of the time, only tacit, latent, unexpressed. In this sense, critical reflection is always, at base, an exercise in what I have been calling rational reconstruction. As with Strawson's grammarian or Brandom's inferential “score-keeper,” the goal is to unwrap and demonstrate the underlying structure of a shared conceptual apparatus with a view to identifying both its constitutive logic and, at the same, those gaps, infelicities, stresses, and self-contradictions—either in itself or in its application—that call for repair. On such an account, the life of the mind is a matter of thought thinking itself, a conceptual structure engaged in a process of self-examination; in many versions, the individual human being him- or herself is best characterized as essentially a vehicle of this larger process. Indeed, holism, in one form or another, is a massive rejection of atomistic individualism. We are, qua individuals, deeply situated in one or another shared universe of discourse that, at the very least, sets firm parameters as to what is possible for us coherently to think. To the degree that we are well-functioning vehicles, moreover, the result will be a set of propositions about how things in the world really are—metaphysical claims—that are not necessarily true from a God's-eye point of view but that are, from a human's-eye point of view, rationally required and perhaps even irrefutable; and if irrefutability is the best we can hope for, it is also all that we would ever need by way of objective knowledge.

Fourth, any conceptual apparatus—any universe of discourse—will be, at least prospectively, a structure of rationality; this is why the exploration of such a structure, whether undertaken by a philosopher or by an ordinary person on the street, will be a matter of rational reconstruction. Thus, “what makes a statement or a whole system of statements—a theory or conceptual scheme—rationally acceptable is, in large part, its coherence and fit” (Putnam Reference Putnam1981, 55; also Strawson Reference Strawson1992, 84). Thinking thinking itself is and can only be governed by the laws of thought; and this means, for example, that contradiction – however productive it may be in inciting thought to think itself—is never, in the end, acceptable. We cannot make sense of the world unless we make sense, and making sense is, at the very least, a matter of self-consistency.

Finally, our engagement with the world, understood along these lines, cannot but have a history. Any conceptual apparatus necessarily inhabits a “realm of culture” (Brandom Reference Brandom2000, 33; see also Strawson Reference Strawson1992, 27) and will, as such, evolve over time. Rational reconstruction of whatever kind is always rooted in and derived from previous, shared efforts at rational reconstruction. There is no way out of this. Neurath's famous image, cited in this context by Quine (Reference Quine1950, 632), of a boat permanently at sea that occasionally springs a leak and that must be repaired piecemeal, one timber at a time, presents, among other things, an idea of historical development. In principle, the boat in question might eventually become a completely new boat, in the sense that all of the original timbers might come to be replaced by new ones. But the repair process itself would hardly be random. To the contrary, it would reflect a logic—a rational account of what it is for a boat to be a boat and how best to achieve the desired outcome. Thus, the so-called new boat will be deeply connected to the old boat via practices of, in effect, reasoned argument and rational judgment. To speak metaphorically, the new boat will be the often tacit but nonetheless rationally reconstructable inference that has been derived, as a matter of logical entailment, from its predecessor.

Rationality and politics. Contemporary critics of rationalism in politics often argue that rationalistic theories presuppose a kind of individualism according to which individuals are “prior to society, bearers of natural rights, and either utility maximizing agents or rational subjects” (Mouffe Reference Mouffe2005, 95–96; see also Connolly Reference Connolly1991, 73–76 and Geuss Reference Geuss2008, 7, 64–68). Such a criticism may have some force against, inter alia, Rawlsian liberals, utilitarians, and contract theorists such as David Gauthier, but it is, as we have seen, wholly irrelevant with respect to the main currents of recent metaphysical inquiry. Indeed, to the exact contrary: The widespread holism of contemporary philosophical rationalism involves nothing less than a systematic exploration of the ways in which we are all perforce both products and servants of larger, shared cognitive systems.

Critics complain that rationalists are “universalists” who ignore “context” (e.g., Abijadeh Reference Abizadeh2007, 269–70; Geuss Reference Geuss2008, 7, 10, 24; Mouffe Reference Mouffe2005, 16, 62–67; Zerilli Reference Zerilli2012, 10–12, 14). Now it is true that philosophical analysis presupposes the need to make sense by avoiding contradiction and seeking coherence, which typically involves, in turn, conceptions of logic understood to be broadly applicable to human thought per se. Of course, one would be hard pressed to find post-metaphysical or nonrationalist writings that proudly pursue and celebrate their own illogic. The more important point, though, is that the kinds of rationalist philosophers we have been considering are, at the least, agnostic regarding and, more often, entirely comfortable with many or most forms of cultural and historical relativism.Footnote ⁸ They view conceptual schemes precisely as social artifacts, and this means that context is absolutely crucial. One universe of discourse may be different from another in all kinds of ways and for all kinds of reasons; and in light of this, questions about relative merits and possible interrelationships—for example, the idea of a “fusion of horizons” that produces the “one great horizon” (Gadamer Reference Gadamer1972, 288–89)—may well be enormously interesting and important. But what gives any such universe its identity and force is, at least prospectively, a structure of interconnectedness that reflects, however aspirationally, an internal logic of rational inference and entailment; and the acquisition, application, and critical appropriation of that logic are always matters of rational reconstruction, whether explicit or otherwise, designed to reveal the argumentative underpinnings of any and all relevant claims, moral and metaphysical alike.

What this indicates, moreover, is that post-Kantian rationality, as a kind of holism, is most emphatically not instrumental rationality; it is not the kind of disembodied, disconnected, ungrounded, “unaided” manner of thinking that Burke or, later, Oakeshott criticized, nor is it the airy and unworldly structure of bloodless speculation out of which one can spin any kind of ideology or justify any kind of policy. Indeed, it is the exact and precise opposite. If rationalism properly understood is, as I have shown, the effort logically and systematically to discover and explicate the implications and entailments of a shared, though often only tacit or even hidden structure of presupposition about how things in the world really are—a structure that composes, in every case, the very foundations of a way of life—then political rationalism is neither more nor less than the effort to explore and realize through argument and analysis what those shared commitments mean for the pursuit of a coherent, intelligible, deeply situated, and organically contextualized kind of public action.

To be sure, any serious relativist doctrine will, as such, acknowledge the influence of beliefs and commitments not reducible to the larger canons of sound reasoning. Indeed, virtually all interesting forms of rationalism—even the most austere varieties of mathematical or scientific practice—recognize that there is no such thing as presupposition-less discourse. Yet from the fact that premises qua premises resist rational demonstration one can hardly conclude that rational demonstration is not central to intelligent experience of the world. In this connection, Wittgenstein, often cited by post-metaphysical political theorists as a critic of rationalism (Mouffe Reference Mouffe2005, 60–79; Zerilli Reference Zerilli2012, 8, 17), is a particularly complex and revealing case; with respect to that case, I note only that Wittgenstein, despite his deep skepticism, insists on logic as the “essence” of thought (Reference Wittgenstein1958, 44), that his famous claim to the effect that “explanations come to an end somewhere” (Reference Wittgenstein1958, 3) itself presupposes the near ubiquity of explanations, and that his own account of, for example, family resemblances among various games (Reference Wittgenstein1958, 31–32) is nothing less than a paradigmatic case of rational reconstruction.

Post-metaphysical critics accuse rationalism of pursuing absolute truths or final answers where no such truths or answers exist (Geuss Reference Geuss2008, 10; Mouffe Reference Mouffe2005, 93; Zerilli Reference Zerilli2012, 21, 24). But exactly to the contrary, the tradition I have been looking at understands truth claims to be inherently provisional. They are the upshots of conceptual schemes, and insofar as such schemes evolve over time, as they are wont to do, our understandings of metaphysical and moral reality will change as well. But those understandings, however formulated, will always be thought to reflect, explicitly or otherwise, arguments of reason embedded in the particular discursive networks out of which they have emerged.

By systematically ignoring many of the most important currents of latter-day philosophy, political thought has embraced criticisms of rationality that are, as an intellectual matter, doubtful at best. But the resulting problems are not simply intellectual in the narrow sense. Having rejected rationalism, post-metaphysical political theorists opt, as we have seen, for something else: rhetoric (Mouffe Reference Mouffe2005, 67, 70; Zerilli Reference Zerilli2012, 17), emotion and affect (Geuss Reference Geuss2008, 38; Mouffe Reference Mouffe2005, 95), interest (Geuss Reference Geuss2008, 9, 11; Mouffe Reference Mouffe2005, 46), and power (Geuss Reference Geuss2008, 27, 51–53; Mouffe Reference Mouffe2005, 21, 49, 99). No one could doubt the role that such things play in politics. At the same time, no one could doubt the dangers they pose. The practice of rhetoric, in and of itself, cannot reliably distinguish uplifting, humanizing oratory from debasing, mendacious propaganda; destructive, hateful emotions are neither more nor less real or authentic than healthy, caring ones. To say that history testifies to the horrors of unreason—including the perverse and intentional misuse of the language of rationality to disguise the very opposite—is only to belabor the obvious.

The post-Kantian convergence of late twentieth-century philosophy, if applied to politics, would have little relevance for the kinds of criticisms that post-metaphysical political thought has leveled against rationalism in general, even as it would allow us to uncover and contemplate robust considerations of truth and justifiability regarding matters of public consequence—considerations to which we are already implicitly committed and that underwrite, however tacitly, what we actually do. Of course, models for such an approach already exist; let me mention, to pick just one notable example, the work of Charles Taylor, wherein objective, reasoned evaluation and culturally dynamic holism are understood not merely to coexist but to be mutually reinforcing, indeed mutually dependent. In this sense, moreover, I suggest that the distinctiveness of politics as a mode or area of human activity is little more than a question of subject matter. All human endeavor is embedded in the rational appropriation and application of one or another conceptual scheme, and political endeavor—however unusual it may be in its focus on issues of power, justice, legitimacy, and comparative advantage—is no different. Indeed, the principal lesson of the post-Kantian convergence is that post-metaphysical political theory contemplates an impossibility. The ubiquity of metaphysical commitment and of the demands of reason is inescapable. A failure to acknowledge this is no argument against it; however, it increases the likelihood that bad arguments will stand unchallenged, their flaws unrecognized even as they influence the direction of public discourse.