Habermas's theory of communicative action rests on the idea that social order ultimately depends on the capacity of actors to recognize the intersubjective validity of the different claims on which social cooperation depends. In conceiving cooperation in relation to validity claims, Habermas highlights its rational and cognitive character: to recognize the validity of such claims is to presume that good reasons could be given to justify them in the face of criticism. TCA thus points to and depends on an account of such justification—that is, on a theory of argumentation or discourse, which Habermas calls the “reflective form” of communicative action.
As mentioned above, Habermas proposes a multi-dimensional conception of reason that expresses itself in different forms of cognitive validity: not only in truth claims about the empirical world, but also in rightness claims about the kind of treatment we owe each other as persons, authenticity claims about the good life, technical-pragmatic claims about the means suitable to different goals, and so on. As he acknowledges, the surface grammar of speech acts does not suffice to establish this range of validity types. Rather, to ground the multi-dimensional system of validity claims, one must supplement semantic analysis with a pragmatic analysis of the different sorts of argumentative discourse—the different “logics of argumentation”—through which each type can be intersubjectively justified (TCA 1: 8–42). Thus, a type of validity claim counts as distinct from other types only if one can establish that its discursive justification involves features that distinguish it from other types of justification. Whether or not his pragmatic theory of meaning succeeds, the discursive analysis of validity illuminates important differences in the argumentative demands that come with different types of justifiable claims. To see how Habermas identifies these different features, it is first necessary to understand the general structures of argumentation.
The pragmatic analysis of argumentation in general. Habermas's discourse theory assumes that the specific type of validity claim one aims to justify—the cognitive goal or topic of argumentation—determines the specific argumentative practices appropriate for such justification. Discourse theory thus calls for a pragmatic analysis of argumentation as a social practice. Such analysis aims to reconstruct the normative presuppositions that structure the discourse of competent arguers. To get at these presuppositions, one cannot simply describe argumentation as it empirically occurs; as we already saw in TCA, one must adopt the performative attitude of a participant and articulate the shared, though often tacit, ideals and rules that provide the basis for regarding some arguments as better than others. Following contemporary argumentation theorists, Habermas assumes one cannot fully articulate these normative presuppositions solely in terms of the logical properties of arguments. Rather, he distinguishes three aspects of argument-making practices: argument as product, as procedure, and as process, which he loosely aligns with the traditional perspectives on argument evaluation of logic, dialectic, and rhetoric. Pragmatically, each of these perspectives functions as a “level of presupposition” involved in the assessment of the cogency—the goodness or strength—of arguments. Habermas seems to regard these perspectives, taken together, as constituting the pragmatic idea of cogency: “at no single one of these analytic levels can the very idea intrinsic to argumentative speech be adequately developed” (TCA 1: 26).
At the logical level, participants are concerned with arguments as products, that is, sets of reasons that support conclusions. From this perspective, arguers aim to construct “cogent arguments that are convincing in virtue of their intrinsic properties and with which validity claims can be redeemed or rejected” (ibid., 25). Following work by Stephen Toulmin and other informal logicians, Habermas regards most if not all argumentation as ultimately resting on ampliative arguments whose conclusions do not follow with deductive certainty but only as more or less plausible or probable. The logical strength of such arguments depends on how well one has taken into account all the relevant information and possible objections. Thus the term “logical” has a broad sense that includes not only formal but also informal logics, in which strength depends on the interrelated meanings of terms and background information that resists complete formalization: induction, analogy, narrative, and so on.
Given the ampliative character of most arguments, logical assessment presupposes the dialectical adequacy of argumentative procedures. That is, we may regard the products of our argument-making practices as logically strong only if we presume, at the dialectical level, that we have submitted arguments and counterarguments to sufficiently severe procedures of critical discussion—as Habermas (TCA 1: 26) puts it, a “ritualized competition for the better arguments.” Dialectical treatments of argumentation typically spell out the “dialectical obligations” of discussants: that one should address the issue at hand, should respond to relevant challenges, meet the specified burden of proof, and so on.
However, robust critical testing of competing arguments depends in turn on the rhetorical quality of the persuasive process. Habermas conceives the rhetorical level in terms of highly idealized properties of communication, which he initially presented as the conditions of an “ideal speech situation” (1973a; also 1971/2001). That way of speaking now strikes him as overly reified, suggesting an ideal condition that real discourses must measure up to, or at least approximately satisfy—motifs that Habermas himself employed until rather recently (cf. 1993, 54–55; 1996b, 322–23). He now understands the idea of rhetorically adequate process as a set of unavoidable yet counterfactual “pragmatic presuppositions” that participants must make if they are to regard the actual execution of dialectical procedures as a sufficiently severe critical test. Habermas (2005b, 89) identifies four such presuppositions as the most important: (i) no one capable of making a relevant contribution has been excluded, (ii) participants have equal voice, (iii) they are internally free to speak their honest opinion without deception or self-deception, and (iv) there are no sources of coercion built into the process and procedures of discourse. Such conditions, in effect, articulate what it would mean to assess all the relevant information and arguments (for a given level of knowledge and inquiry) as reasonably as possible, weighing arguments purely on the merits in a disinterested pursuit of truth. These conditions are counterfactual in the sense that actual discourses can rarely realize—and can never empirically certify—full inclusion, non-coercion, and equality. At the same time, these idealizing presuppositions have an operative effect on actual discourse: we may regard outcomes (both consensual and non-consensual) as reasonable only if our scrutiny of the process does not uncover obvious exclusions, suppression of arguments, manipulation, self-deception, and the like (2003a, 108). In this sense, these pragmatic idealizations function as “standards for a self-correcting learning process” (2005b, 91).
As an understanding of the rhetorical perspective, Habermas's highly idealized and formal model hardly does justice to the substantive richness of the rhetorical tradition. One can, however, supplement his model with a more substantive rhetoric that draws on Aristotle's account of ethos and pathos (Rehg 1997). In that case, the rhetorical perspective is concerned with designing arguments for their ability to place the particular audience in the proper social-psychological space for making a responsible collective judgment. But the “space of responsible judgment” still remains an idealization that may not be reduced to any observable actual behavior, but can at most be defeasibly presumed. The same probably holds for dialectical procedures. Although the dialectical perspective draws on the tradition of public debate, dialectical norms, when understood as pragmatic presuppositions, are not identical with institutionalized rules of debate (1990a, 91). A neutral observer can judge whether interlocutors have externally complied with institutional procedures, whereas engaged participants must judge how well they have satisfied the dialectical presupposition of severe critical testing.
The differentiation of argumentative discourses. If the different validity claims require different types of argumentation, then the relevant differences must emerge through a closer analysis of the ways the above aspects of argumentative practice adjust to different sorts of content, that is, the different validity claims at issue (cf. 2005b, chap. 3). To be sure, Habermas does not regard every validity claim as open to discourse proper. Sincerity claims (or “truthfulness claims,” as it is sometimes translated) are the prime example. These are claims an actor makes about his or her interior subjectivity: feelings, moods, desires, beliefs, and the like. Such claims are open to rational assessment, not in discourse but by comparison with the actor's behavior: for example, if a son claims to care deeply about his parents but never pays them any attention, we would have grounds for doubting the sincerity of his claim. Note that such insincerity might involve self-deception rather than deliberative lying.
Truth and rightness claims, by contrast, are susceptible to argumentative justification in the proper sense, through what Habermas calls “strict discourses.” As he first analyzed the discourses connected with these two types of validity (1973a), they had much in common. Although the types of reasons differed—moral discourse rested primarily on need interpretations, empirical-theoretical discourse on empirical inductions—in both cases, the relevant reasons should, in principle, be acceptable to any reasonable agent. In the case of empirical truth claims, this process-level presupposition of consensus rests on the idea that the objective world is the same for all; in the case of moral rightness, it rests on the idea that valid moral rules and principles hold for all persons. In both cases, the appropriate audience for the testing of claims is universal, and in making a truth or rightness claim one counterfactually presupposes that a universal consensus would result, were the participants able to pursue a sufficiently inclusive and reasonable discourse for a sufficient length of time. Although his early statements are somewhat unclear, on one reading Habermas defined not only moral rightness but also empirical truth in terms of such ideal consensus (similar to C. S. Peirce). He now further distinguishes truth from moral rightness by defining the latter, but not the former, in terms of idealized consensus. More on that below.
Authenticity claims, unlike truth and rightness claims, do not come with such a strong consensual expectation. Habermas associates this type of claim with “ethical” discourse. Unlike moral discourse, in which participants strive to justify norms and courses of action that accord due concern and respect for persons in general, ethical discourses focus on questions of the good life, either for a given individual (“ethical-existential” discourse) or for a particular group or polity (“ethical-political” discourse). Consequently, the kind of reasons that constitute cogent arguments in ethical discourse depend on the life histories, traditions, and particular values of those whose good is at issue. This reference to individual- and group-related particularities means that one should not expect those reasons to win universal consensus (1993, 1–18; 1996b, 162–68). However, Habermas (2003b) seems to recognize one class of ethical questions that do admit of universal consensus. Choices of technologies that bear on the future of human nature, such as genetic enhancement engineering, pose species-wide ethical issues. Such issues concern not merely our self-understanding as members of this or that particular culture or tradition, but how we should understand our basic human dignity. The core of human dignity, and thus the basis for a human-species ethics, on his view, lies in the capacity of human beings for autonomous self-determination.
In sum, Habermas's discourse theory aligns different types of validity claim with different types of justificatory discourse. At the logical level, cogent arguments must employ somewhat different sorts of reasons to justify different types of claims. Although some sorts of reasons might enter into each type of discourse (e.g., empirical claims), the set of relevant considerations that are individually necessary and jointly sufficient for making logically strong arguments will differ. Thus, claims about what human beings need are relevant reasons in moral arguments about welfare obligations, but not for supporting the truth claim that quarks exist. At the dialectical level, one must meet different burdens of proof by answering different types of challenges. For example, in defending the ethical authenticity of Tom's pursuit of a career in medicine, one need not show that medicine is a career everyone must follow, but only that such a career makes sense, given Tom's personal background, talents, and desires. One can also examine Tom's career choice from a moral perspective, but in that case one need only show that anyone in his circumstances is morally permitted to pursue medicine. At the rhetorical level, finally, the scope and depth of agreement differs according to the type of claim. Moral rightness claims and empirical truth claims are justified by reasons that should be acceptable to a universal audience, whereas ethical claims are addressed to those who share a particular history and tradition of values.
Having differentiated types of discourse, Habermas must say something about how they interrelate. Clearly, some discourses depend on other types: most obviously, moral and ethical discourses partly depend on empirical claims, and thus depend on the outcome of empirical discourses about the circumstances and consequences of behavioral rules and the collective pursuit of the good life. The question of interrelationship becomes especially urgent in the political sphere, where different discourses intertwine and lead to competing conclusions, or when issues arise in which discourse types cannot be cleanly separated, so that the standards of cogency become obscure or deeply contested (McCarthy 1991, chap. 7; 1998). Because Habermas (1996c, 1534f) rejects the idea of a metadiscourse that sorts out these boundary issues, he must answer this challenge in his democratic theory. Before taking up that topic, Habermas's theory of truth deserves a closer look
