In 1982 Alvin Plantinga delivered a presidential address titled “How to Be an Anti-realist” at the annual meeting of the Western Division (now the Central Division) of the American Philosophical Association (APA).2 He proposed to mediate a dispute about realism encapsulated in addresses given by three previous APA divisional presidents, with Hilary Putnam and Richard Rorty on the anti-realist side and William Alston on the realist side. Plantinga argued that contemporary anti-realism is “wholly unacceptable” and Platonist “unbridled realism” is “unlovely.”3 The right way to be an anti-realist, he said, is to be a theist in the manner of Thomas Aquinas.
Like Plantinga, I think the Anglo-American dispute between realists and anti-realists is significant for how we think about truth. Questions about propositional truth are at the heart of this debate. Plantinga characterizes Putnam and Rorty as “creative anti-realists” concerning propositional truth. He says that, following a line of thought that supposedly began with Kant, Putnam and Rorty think if human minds “can’t settle the question whether a proposition is true, then there’s no truth there to be known.”4 In other words, the truth of a proposition depends on whether the proposition can be humanly justified. Plantinga raises two objections to this apparent reduction of truth to justification. First, Putnam and Rorty tend toward “self-referential incoherence”: if truth is as they say it is, then their construals are not true.5 Second, their anti-realist views, which “begin by taking fundamental disagreement seriously,” end up “denying its possibility.”6
Nevertheless, Plantinga does acknowledge that truths and propositions can hardly be “totally independent of minds or persons.” He is not attracted to “Platonist” realism, by which he means the position that the existence of truths and propositions is completely independent “of minds and their noetic activity.”7 Accordingly, he splits the difference and proposes theistic anti-realism in the style of Aquinas as an alternative to both Platonism and creative anti-realism. On a theistic anti-realist construal, truth must be independent of human noetic activity but cannot be independent of God’s noetic activity. Propositions exist because God thinks them, and they are true if and only if God believes them.8
Undoubtedly this is a bold set of claims to make, especially in a presidential address to the largest professional association of philosophers in North America. Yet whether Plantinga avoids the two problems he attributes to Putnam and Rorty is an open question. Although I agree that what he calls “creative anti-realism” is deeply problematic, I find Plantinga’s epistemological reversion to medieval theism no more acceptable than Plantinga considered the positions of Putnam and Rorty to be.9
To find a better resolution to this dispute, I plan to discuss the theory of truth put forward by German social philosopher Jürgen Habermas. Habermas has moved from an anti-realist to a quasi-realist conception of propositional truth, and this trajectory points in a promising direction. Moreover, he has taken this path in response to Putnam and Rorty. First I describe the development of Habermas’s theory of truth and examine a recent version in greater detail. Then I propose an alternative account of propositional truth, one that critically appropriates Habermas’s insights and aims to resolve the realism/anti-realism dispute.
Before I discuss Habermas, let me explain two sets of terms: alethic realism and anti-realism, and nonepistemic and epistemic conceptions of propositional truth. These two sets overlap to a large extent, and I use them somewhat interchangeably: alethic realism is usually nonepistemic, and alethic anti-realism is usually epistemic. Taking a cue from Michael Lynch, by alethic realism I mean the position that propositional truth “hinges not on us but on the world.” Whether a proposition is true depends on whether “things in the world are as that proposition says they are.”10 Correspondence theories of truth, which define truth as correspondence between propositions and facts or states of affairs, are good examples of alethic realism. Alethic anti-realism is the position that propositional truth does not hinge on whether things in the world are as propositions say they are. Instead, it hinges on other factors, such as how well the proposition coheres with other propositions (coherence theories) or what consequences follow from employing the proposition in inquiry or action (pragmatic theories). Plantinga’s position straddles these alternatives, since he argues that it is not the world’s existence but God’s noetic activity that makes propositions true.
The distinction between nonepistemic and epistemic conceptions of truth pertains primarily to the relation between the truth of propositions and justifications of their truth. A nonepistemic conception claims “whether a proposition is true does not depend on whether anyone is justified in believing the proposition.”11 In that sense, propositional truth is “mind-independent,” as Plantinga holds with respect to human minds but not with respect to God’s mind. An epistemic conception claims that whether a proposition is true does depend to some significant degree on whether someone is justified in believing it.12 In principle, one could subscribe to alethic anti-realism without having an epistemic conception of truth, but usually the two go together,13 as they do in Habermas’s earliest formulations of a truth theory.
Habermas’s conception of truth emerges from his lifelong attempt to spell out a critical theory of contemporary society that accords with deeply democratic politics. His development of this conception has gone through three stages, which I label consensus theory, formal pragmatics, and pragmatic realism. For more than two decades, Habermas developed an epistemic and anti-realist conception of truth. In the 1990s, however, he began to change his mind, and his current conception incorporates some elements of alethic realism.
Habermas’s consensus theory of truth emerged in the early 1970s.14 Like Rorty and Putnam during that decade, Habermas proposed an epistemic conception of truth. In it, he tries to explicate the concept of truth by saying how truth claims can be justified. His central claim is that truth primarily has to do not with a supposed correspondence between propositions and facts but with the discursive redemption or vindication (Einlösung) of the validity claims that we unavoidably raise when we use language to reach understanding with one another. There are three such validity claims – claims to truth, to rightness, and to truthfulness or sincerity (Wahrhaftigkeit) – and our various speech acts raise all three either explicitly or implicitly.15 The claim to truth arises most explicitly in constative speech acts – in assertions, explanations, descriptions, and the like.
To illustrate, say you and I are talking and I tell you, “Alumni Hall is in the Victoria College Building at the University of Toronto.” When I make this assertion, I unavoidably and simultaneously claim that the proposition I assert – that Alumni Hall is in the Victoria College Building – is true. If you are puzzled by my assertion or question its veracity, then I owe it to you to back up my truth claim. Such questioning and backing of claims typically occurs through argumentation, according to Habermas. Argumentation and potential agreement concerning validity claims are what he has in view when he writes about their discursive vindication. On a consensus theory, he says, “The truth condition of propositions is the potential assent of all others . . . The universal-pragmatic meaning of truth, therefore, is determined in terms of the demand of reaching a rational consensus.”16 In this way Habermas indexes the truth of propositions to the quality of their justification.
Habermas modifies and deepens his account of discursive justification in his two-volume Theory of Communicative Action, published in German in 1981.17 There he begins to distinguish more clearly between truth and justification by linking justification with meaning rather than with truth per se. In place of a consensus theory of truth, he proposes a formal pragmatic theory of meaning.18
Habermas’s most important move in this context is to embed linguistic meaning – the meaning of sentences – within communicative meaning – the meaning of human interactions that employ sentences to reach mutual understanding. To understand linguistic meaning, he says, we need to understand how people use language to communicate. At bottom, linguistically mediated interaction serves three pragmatic functions: to represent states of affairs, to establish and renew interpersonal relations, and to present one’s own experiences. The first function is cognitive; the second, regulative; and the third, expressive. Just as truth-conditional semantics that stem from Gottlob Frege argue that we understand the meaning of a sentence when we know under what conditions the sentence is (or would be) true, so Habermas argues that we understand the meaning of a linguistically mediated interaction when we know the conditions under which it is (or would be) acceptable. Given the differentiation of language functions, however, and the correlated differentiation of validity claims, Habermas says one must expand these acceptability conditions “beyond the truth of propositions” and no longer identify them “on the semantic level of sentences but on the pragmatic level of utterances” or speech acts.19
The acceptability conditions for assertoric speech acts are assertibility conditions. According to Habermas, to understand the meaning of my asserting “Alumni Hall is in the Victoria College Building,” you need to know more than the truth conditions for the proposition asserted. You also need to know when I, as the speaker making the assertion, have “good grounds to undertake a warrant that the conditions for the truth of the asserted sentence are satisfied.”20 In other words, you need to know what would authorize me to claim the asserted proposition is true and what would correlatively motivate you to accept my claim. So, a hearer understands the meaning of a speaker’s act of assertion if the hearer knows two sets of conditions: first, the conditions that would make the asserted proposition true (e.g., what must be the case in order for the proposition “Alumni Hall is in the Victoria College Building” to be true); second, the conditions under which the asserter would have convincing reasons for claiming this asserted proposition is true (e.g., what sorts of additional information I would have to provide to support my claim that the asserted proposition about Alumni Hall is true).
Despite the distinction between truth conditions and assertibility conditions, however, Habermas has not really provided a theory of propositional truth. Instead, he has offered a theory of discursive justification. Essentially, truth means “justified” or “warranted” assertibility. As Barbara Fultner pointed out in 1996, however, there is “a patent tension between truth and warranted assertibility.”21 For, as Putnam indicated in 1981, truth is supposed to be a feature “that cannot be lost, whereas justification can be lost.”22
Habermas responds to such concerns in a 1996 essay titled “Richard Rorty’s Pragmatic Turn.”23 Emphasizing both the distinction between truth and justification and their internal relation, he argues that “truth cannot be reduced to . . . justified assertibility.”24 This argument relies heavily on two claims. The first claim concerns human behaviour: “as interacting and intervening subjects, we are always already in contact with things about which we can make statements” – not despite language but precisely in virtue of the linguistically mediated character of human action. The second claim concerns the presuppositions of communication: participants in communicative action cannot reach understanding with each other unless we “refer to a single objective world” and suppose this world to be “independent of our descriptions.” Whereas Rorty tends to deny both claims, Habermas thinks that linguistically mediated behavioural contact with things and the supposition of an independent objective world are necessary if everyday practices are not to “come apart at the seams.”25
Accordingly, in Truth and Justification, Habermas presents what he calls a “Janus-faced” concept of truth.26 The current version of his theory is (roughly) realist with respect to truth and anti-realist with respect to justification. Habermas’s own description suggests the label “linguistic-pragmatic epistemological realism” or, more simply, “pragmatic realism.”
Habermas formulates his Janus-faced concept of truth in response to two questions. First, if all forms of human life emerge contingently from natural evolution, how can we account for our unavoidably raising truth claims that transcend the contexts in which we raise them? Second, how can we reconcile the claim that any human access to so-called reality is linguistically mediated with the intuitively plausible realist assumption that “there is a world existing independently of our descriptions of it and that is the same for all observers”?27 If Habermas cannot give satisfactory answers, then he will need to adopt Richard Rorty’s epistemic, anti-realist, and radical contextualism with respect to propositional truth. Habermas responds by introducing several pairs of distinctions: between communication and representation, between lifeworld and objective world, between meaning and reference, and, most significantly for our purposes, between justification and truth. Let me summarize these distinctions.
To begin with, Habermas insists that representation and communication are equiprimordial functions of language usage. In our speech acts, we communicate with others by referring to “the world and its objects,” and vice versa: “As representation and as communicative act, a linguistic utterance points in both directions at once: toward the world and toward the addressee.”28 Both representation and communication are made possible by “deep-seated structures of . . . the life-world,”29 and they pragmatically presuppose a shared objective world. The lifeworld encompasses all types of rule-governed action, both linguistic and nonlinguistic and both communicative and strategic or instrumental. The notion of an objective world, by contrast, pertains to the subset of rule-governed actions that aim “instrumentally to intervene in or strategically to influence the course of events.”30 This “performatively established relation to objects,” in turn, has links to “the semantic relation to objects that interlocutors establish in asserting facts about them.”31 Agents and language users pragmatically presuppose “a shared objective world as the totality of objects to be dealt with and judged” and “as existing independently and as the same for everyone.”32 This pragmatic presupposition underlies our truth claims.
Habermas links successful reference and propositional truth to “the normativity of successful coping.”33 To explain this link, he introduces a firm distinction between meaning and reference.34 The distinction relies on two claims. First, linguistic practices enable us to “refer to language-independent objects about which we assert something.” Second, any language user, regardless of his or her context, can “refer to a common system of possible referents” and can “identify independently existing objects.” Accordingly, “reference to the same object must remain constant even under different descriptions.”35 This is possible because the purposive contact we have with objects funds the “semantic relations” we “explicitly establish” when we make assertions.36 So, for example, our asserting “this is water, not oil” is rooted in our nonlinguistic actions of drinking water, swimming in it, using it to cultivate crops, and the like. If we were not able, on the basis of ordinary practices, to keep the reference of “water” constant, then we could not use the term successfully in multiple situations.
Yet the constancy of reference does not explain the truth of statements or propositions, which “can be justified only by means of other statements” and not by a direct appeal to practical experience.37 To that extent, Rorty’s contextualism is not off base. Nevertheless, Habermas points to three factors that cannot be reconciled with radical contextualism, namely, the assumption of “epistemological realism” in ordinary practices, the “power of learning processes” to revise the contexts in which they occur, and the tendency of truth claims to have “universalist import” that transcends the context in which we raise them.38 In everyday life, when we assert “this is water, not oil,” we assume the statement to be true with respect to the liquid we experience; we can use such a statement to learn something new about our environment; and we consider the statement to be true not just here and now and for ourselves, but also at any relevant time and place and for anyone who could respond to our assertion.
Habermas’s pragmatically realist conception of truth aims to account for these factors. He does not give up his emphasis on rational discourse as the privileged mode for sorting out truth claims.39 Yet, he says the fact that epistemologically we must connect truth and justification should not mislead us into thinking that truth is simply an idealization of justification. According to Habermas, a nonepistemic concept of truth shows up in everyday action. Ordinarily, in the course of action, we do not doubt the truth of our beliefs. We rely on “certainties of action” in our “practical dealings with an objective world” and naively take the beliefs that guide our actions “to be true absolutely.” As Habermas puts it, “We don’t walk onto any bridge whose stability we doubt.” Implicit in such “realism of everyday practice” is “a concept of unconditional truth.” As soon as our practices fail and contradictions arise, however, we begin to see the truth of the relevant beliefs “as merely ‘presumed truths,’ that is, as fundamentally problematic truth claims.” Moreover, if someone else challenges the truth of what we assert, then our only recourse is to try to justify “the now thematized truth of controversial propositions.”40
In other words, the concept of truth plays a different role in the domain of action than it does in justificatory discourse. In action, we implicitly hold our beliefs to be undoubtedly true and implicitly make truth claims about them when we communicate. At this level we assert facts that concern “objects themselves.”41 In discourse, by contrast, our claims to truth are always and unavoidably provisional. The only way to vindicate such claims is via “the convincing power of good reasons,” and even the best reasons remain “under the proviso of fallibility.”42 A discursive agreement authorizes us to accept only the justified claim that a proposition is true, not the truth of the proposition as such.
Habermas thinks the nonepistemic concept of truth presupposed in action provides a “justification-transcendent standard for orienting ourselves by context-independent truth claims.” This in turn sheds a pragmatic light on the concepts of objectivity, facts, and states of affairs that figure prominently in more robustly realist conceptions of truth. According to Habermas, the concept of objectivity arises from our failures to cope with the world, when we “experience in practice that the world revokes its readiness to cooperate.” Such failures force us to recognize that “the world is not up to us” and exists independently of our language usage and actions. Moreover, the concept of an objective world enables us to “refer to things that can be identified as the same under different descriptions” and so to regard the world as “the same for all of us.”43 Such pragmatically recognized objectivity sustains the robustly realist claim that true assertions refer to facts about the world or to states of affairs that obtain, a claim that calls attention to “a connection between . . . the truth of statements and the ‘objectivity’ of that about which something is stated.”44 Accordingly, truth cannot be a merely epistemic notion.
Compared with his earlier consensus theory of truth and his formal pragmatics of meaning, Habermas’s Janus-faced pragmatic realism goes some distance to answer objections like the ones Plantinga poses for Putnam and Rorty. Yet it also raises two new questions that need to be addressed. One concerns the relation between propositions and what we experience as an “objective world.” In what way, if any, does this relation make propositions true? The other concerns the relation between propositional truth and discursive justification. Does argumentation aim only to establish the acceptability of truth claims? Or does it also seek to affirm the truth of propositions as such? I do not believe Habermas has a satisfactory answer to either question.45 Instead, I propose an alternative conception that, reworking Habermas’s insights, promises a post-Plantingian resolution to the debate between realists and anti-realists in Anglo-American philosophy.
My overarching conception of truth is much broader than a concept of propositional truth, and my conception of the authentication of truth is considerably more expansive than a concept of discursive justification.46 I assume, for example, that propositional truth is neither the only nor the privileged mode of truth. I also assume that discursive justification is only one manner of authentication and that it regularly supports and receives direction from other ways of authentication. With those qualifications in mind, I will discuss the concept of propositional truth in relation first to empirical facts and then to discursive justification.
In the analytic literature on truth, disagreements about the relation between propositional truth and empirical facts have to do with “truth bearers,” “truth makers,” and the connection between the two. Although Habermas rejects alethic realism as a general position, he wishes to retain aspects of the connection realism posits between truth bearers and truth makers. Specifically, he wishes to retain the realist intuition that the truth of propositions is indexed to the existence of facts or states of affairs. Yet Habermas does not follow the empiricist path of positing facts as experienced objects. Facts are related to what we experience in our practical engagements with the world, he says, but we do not experience them as such. Rather, they are formulated when we assert propositions or, more broadly, when our speech acts employ sentences that have propositional content.
If, contrary to standard versions of correspondence theory, the facts that make assertions and propositions true do not “exist” in the “real world,” then Habermas needs to account for how facts relate to the things with which we are “in contact” as “interacting and intervening subjects”47 – what I will call practical objects. Habermas’s view of this relationship is not very clear. However, I posit a two-way process of decontextualization and recontextualization to make sense of how he views the fact/object relation.
In the ordinary course of action, I experience a liquid that looks, smells, tastes, and feels a certain way and I say to you, “This is water.” I have no reason to doubt the liquid is water, and so I take the belief I have formulated to be straightforwardly true. In this context, true seems to mean that the belief in question is reliable in practice and that my assertion is a reliable articulation of my belief. If you challenge my assertion, you question the reliability of my belief and of what I have said. Such questioning decontextualizes the practical belief and the content of what I have said: “Is it really water?” Only under such conditions of decontextualization does the fact that “this is water” emerge, when the content of what I have said becomes a topic for disagreement and potential agreement. What is reliably believed in practice is decontextualized into what is questionable in fact.
Recontextualizing occurs when sufficient agreement is reached for both interlocutors to rely on the belief whose content was questioned. For example, we could end our discussion by both deciding not to imbibe the mysterious liquid. Alternatively, I could offer you a cup of cold water. At this point, whether the liquid is water no longer functions as a fact to be disputed but as the reliable content of what both of us believe in practice.
I find much to commend this proposed reworking of Habermas’s approach. But there is something missing. What is it about practical objects that allows us to acquire true beliefs about them – i.e., practically reliable beliefs – in the first place? What is it about practical objects that allows us to make truth-claiming assertions about them? Habermas has no answer to these questions. Moreover, I suspect that the linguistic-pragmatic underpinnings to his truth theory would not support his venturing an answer. This is one bridge he is not ready to cross. I, by contrast, think an answer is necessary if we want to give full credence to the correct intuitions of correspondence theories.
It is not enough, in my view, to presuppose an independent objective world that is the same for everyone. We also need to presuppose, in both action and discourse, that practical objects lend themselves to our actions and interactions, and do so in ways that are not solely of our own making or choosing. In other words, we need to presuppose an interdependent world, one where the “readiness to hand” or “handiness” (what Heidegger calls Zuhandenheit) of practical objects sustains our projects in ways that supervene on the particular uses to which we put such objects in specific contexts. We can have reliable beliefs about water, for example, because water really does quench our thirst and we can rely on it to do so.
One of the ways in which practical objects lend themselves to our actions and interactions is by letting us talk about them. In our talking we refer to them and make predications. I call something “water” and tell you it is cold. If we speak the same language, you probably understand what I mean. To explain how this is possible requires more than theories of reference and predication and more than accounts of speech acts and their propositional content. It also requires acknowledging how practical objects let us talk about them.
If, in addition, we want to explain how our talking about practical objects can be reliable and how our assertions about them can be true, then we need to account for the specific ways practical objects support reliable talk and true assertions. Elsewhere I identify this feature of practical objects as predicative availability.48 The predicative availability of practical objects is the way they allow us to assert something about them. To do so, they must let us refer to them when we use language. Practical objects let us establish their identity under referring expressions and via referential uses of language without allowing this referential identity to be the entirety of what they are. Predicative availability is the face that practical objects turn toward us when we call their names – i.e., when we successfully refer to them using language.
There is more to predicative availability than linguistic referability, however. To lend itself to reliable talk and true assertions about it, a practical object must also allow its linguistically established identity to be linguistically specified. “The water is cold,” we say, or “Alumni Hall is in the Victoria College Building,” and the water lets itself be specified with respect to temperature, while Alumni Hall lets itself be specified with respect to location. Predication is the linguistically embedded specification of linguistically identified objects, and it is a normal part of everyday language usage. Predicative availability is how practical objects let us refer to them with specificity in language.
Reliable talk and true assertions with respect to practical objects require more than reference, predication, and predicative availability. The way a practical object shows itself to us must line up with some other relevant way in which it is available. If I say to you, “The water is too cold,” for example, that which allows me to refer to it as “water” and to speak of it as “too cold” must also allow me to relate this predicate to the water’s nonpredicative availability for, say, swimming. This capacity on the part of practical objects is what I call predicative self-disclosure. Predicative self-disclosure is what the object under reference allows us to specify in relation to at least one other way in which the object is available to us.
Here the concept of a fact becomes significant. A fact is simply the predicative self-disclosure of an object or of a range of objects on the occasion of language usage in which reference and predication occur with respect to these objects. A fact is not either real or linguistic – it is both. It is a way in which objects relate to human practices, insofar as these practices include linguistic ones of reference and predication.
To this extent, facts are relevant for all reliable talk, and not only for constative speech acts in general or assertoric speech acts in particular. Habermas casts facts too narrowly when he portrays them as what we assert and what makes statements or propositions true. Although he is right to warn us against equating facts with practical objects as such, he fails to see that facts are a mode in which practical objects disclose themselves.
But how should we understand the relation between assertions and propositions on the one hand, and facts on the other? I agree with Habermas that assertions are speech acts that raise a truth claim and that facts make propositions true. I also agree that propositions are the content of our assertions and that, via analysis, we can assign a propositional content to any speech act, whether assertoric or not. Yet I find implausible his early claim, which he seems not to have given up, that truth “is not a property of assertions” and is rather only something we claim concerning the propositions or statements (Aussagen) we assert.49 For in everyday communication we are just as concerned about the truth of what is asserted as we are concerned that someone’s asserting something is true. If you are not convinced of the water’s temperature when I assert, “The water is too cold,” you could exclaim, “That’s not so!” But you could also ask, “Are you sure about that?” A question or exclamation of the first sort pertains to the truth claimed for the proposition – that it is true that the water is too cold. A question of the second sort pertains to the truth of my asserting this.
If my description is correct, then the concept of truth appears to have a double role when we assert propositions. In one role it pertains to the correctness of the practice of making assertions. In the other role it pertains to the accuracy of the result of engaging in this practice, namely, the accuracy of the content of what one asserts – the accuracy of the proposition.50 How, then, do facts relate to both the truth of asserting (i.e., correctness) and the truth of propositions (i.e., accuracy)? From the preceding analysis of facts, it appears that asserting is true (i.e., correct) just in case the practical object about which we make an assertion self-discloses in the manner asserted. My asserting “the water is too cold” is a correct speech act on a specific occasion iff, for example, the water shows itself to be insufficiently warm for us to swim in.
The truth of propositions, by contrast, involves a further decontextualization from this underlying assertoric relation. The propositional content of an assertion takes the form “that something is the case,” for example, “that the water is too cold.” Typically, we use this form to take up assertions in discourse. We argue about the content of what someone asserts and seek agreement about whether this or that is the case. Facts also undergo decontextualization in relationship to propositions. They are no longer simply a way in which practical objects self-disclose for linguistic practices. Now they are “state of affairs” that must “obtain” in order for the proposition to be accurate. Due to double decontextualization, the proposition is available for anyone to discuss, and the state of affairs is projected beyond a particular object to encompass any practical objects that would be similar in a relevant way.
According to the proposed analysis, the relation that sustains propositional truth is not a correspondence between propositions and facts or states of affairs. Rather, it is one of decontextualized disclosure. In this relation the proposition helps us identify the asserted object as an abiding topic for potential discussion and debate, and the fact as a state of affairs presents the asserted object as nothing other than how it has been asserted to be. A proposition about a practical object is true (i.e., accurate) just in case the asserted object presents itself as nothing other than how it has been asserted to be. Propositions are decontextualized statements of decontextualized predicative identity.
Insofar as contemporary truth theories restrict truth to propositional truth, and insofar as propositional truth involves decontextualization, we can understand why theorists such as Putnam claim that propositional truth “cannot be lost.” Apart from its role in assertoric practices and its underpinnings in action, interaction, and their practical objects, however, propositional truth would never be gained. Although Habermas’s Janus-faced concept of truth in both action and discourse points in this direction, I hope to have offered a better explanation for the relation between propositional truth and empirical facts. I suggest that we must distinguish three facets of truth – practical reliability, assertoric correctness, and propositional accuracy – and not subsume all three under the notion of propositional truth.
Habermas’s pragmatic realism also raises questions about the relation between propositional truth and discursive justification. Although he has revised his conception of truth, the main lines of his account of justification remain the same: discursive justification is a process of argumentation that aims to establish the validity of the claims that are raised with speech acts. The claims it aims to establish are ones whose validity would transcend the context in which a claim arises.
What do we attempt to justify discursively when propositional truth is at issue? According to Habermas, we do not try to justify the truth of the proposition as such. Rather, we try to justify our claim to propositional truth – the claim to truth that is raised with our speech act. We do not argue for or against the truth of the proposition as such, but we exchange reasons for accepting or rejecting the claim that an asserted proposition is true. Our readiness to back up our claims with convincing reasons sustains our assertoric interactions.
Here, however, an issue arises that Habermas has not successfully addressed. Does the truth or falsity of a proposition make any difference to the outcome of discourse about truth claims? Intuitively, one wants to say yes, it does make a difference. It appears that even the most convincing reasons I can give would not be good reasons to claim my asserted proposition is true if it is untrue. Habermas, however, is hesitant to say this. In his account, what makes for better and worse justifications is not so much the content of the reasons we give as the manner in which we give them. What matters most for him is the degree to which justification approximates a genuinely democratic process.
Hence Habermas’s account of justification poses a challenge for someone who shares his democratic, procedural emphasis but sees a more substantial link between propositional truth and discursive justification and does not endorse a correspondence theory of truth. To explain this link, I view the argumentative justification of assertoric truth claims as a discursive mode of authenticating propositional truth. As I explain elsewhere, authentication in general encompasses all the ways we bear witness to truth, and truth in general requires authentication in order to unfold.51 Accordingly, the argumentative justification of assertoric truth claims is how we bear witness to propositional truth and how propositional truth unfolds. This implies that, although propositional truth “cannot be lost,” it also cannot be static – propositional truth is not eternal or immutable. It also implies that discursive justification is normatively indexed to propositional truth and not simply to procedural norms of argumentation.
Let me elaborate. The proposition one asserts, if it is true, discloses a fact – itself a mode of self-disclosure – in a decontextualized way. When I assert a proposition, I simultaneously claim that the proposition is true in the sense that it is accurate. The accuracy of a proposition is its ability simultaneously to decontextualize and to disclose a practical object’s self-disclosure. I raise a truth claim not simply to elicit your agreement but also to affirm the insight that the proposition affords. I am not simply saying, in effect, “This is true, don’t you agree?” I am also saying, “This is a reliable insight, correctly asserted, that anyone should be able to share.” Although it is so that discourse unavoidably “fallibilizes” both my claim and the purported insight, such fallibilism can only make sense if we simultaneously expect propositions to be genuinely insightful, albeit in a decontextualized way. Good arguments connect a proposed insight with other insights that are relevant for understanding the matter about which the original assertion was made.
Moreover, such connecting of insights occurs in a manner that is distinctive to argumentation. It occurs with reference to the societal principle of logical validity. Although this principle is only one among many that guide human action and interaction, and although in nondiscursive contexts it is rarely the primary principle offering guidance, the pursuit of logical validity plays a leading role in discourse. Argumentation is the best way to justify problematized truth claims because it allows us to disclose propositions within the horizon of logical validity, and thereby to discover just how far their purported insight reaches – i.e., whether they have the universality and necessity we claim for them.
On this account, the aim of argumentation is not simply to vindicate the truth claims raised for an asserted proposition, but also to bear witness to the truth – i.e., accuracy – of the proposition itself. We raise truth claims and argue about them because we presuppose, when we assert a proposition, that it offers genuine insight and because we want to discover whether this is so. We assume that the insight offered is one anyone, not merely the person or community or organization that asserts the proposition, can understand and accept. We also assume that argumentation itself, when carried out well, either discloses the insight in question or shows it not to be a genuine insight – by linking it with other insights, by opening it to consideration by other people, and by testing its implications. In other words, discursive justification serves to authenticate propositional truth.
By enriching the concepts of discourse and propositional truth, I suggest that the concept of truth in everyday practice is not as “unconditional” or “absolute” as Habermas claims it to be, nor is the discursive vindication of truth claims as fallibilistic as he suggests. For in action and discourse our assertoric speech acts orient us to the disclosure of practical objects and the propositional disclosure of such disclosure. We ordinarily understand that not every purported insight is genuine – otherwise we might not bother to make assertions. But we also understand that the claims we raise with our assertions and about which we argue would hardly be worth raising if none of our asserted propositions were accurate. Nor do we need to presuppose ideal justificatory conditions in order to engage in discourse. The principle of logical validity, which is always already in effect in a society such as ours, is sufficient to orient our attempts to give good reasons and better arguments.
If my critical appropriation of Habermas is on the right track, then the framework of the Anglo-American debate about realism is misconceived. The fundamental question about propositional truth is not whether it exists independently of our knowing it. Rather, it concerns the interdependence between linguistically competent, intersubjectively connected human agents and the events and entities these agents act and interact with and make predicative reference to. Plantinga is right to suggest that Kant’s Copernican revolution disturbs our understanding of this relationship. Yet he is wrong, in my judgment, to propose that locating true propositions in the mind of God will undo the revolution, for Plantinga’s proposal presupposes the problem it is meant to solve. We need to begin instead with the interdependence of “mind” and “object” and with the corporeal multidimensionality of both human knowers and that about which they acquire knowledge.
Habermas’s linguistic-pragmatic conception of truth is a postmeta-physical attempt to retain truth’s context-transcending quality in the face of contextualism such as Rorty’s that would deny this quality. Plantinga also wishes to preserve this context-transcending quality, but by reverting to a metaphysical paradigm. And, like Habermas, he ends up with a compromise between realism and anti-realism. To my way of thinking, such compromises indicate problems rather than solutions. I believe the only way to solve these problems is to go beyond the contemporary framework of realism versus anti-realism. What I have proposed could be called post-anti/realism with respect to propositional truth. The key to my proposal is to replace questions of independence with questions of interdependence. Questions of interdependence with respect to propositional truth assume continuity between human agents and whatever they take up in their actions and interactions. For a post-anti/realist, it does not make sense to locate propositions on one side of a mind/object gap and then argue about whether what makes them true is independent of the assertions that raise a claim to truth. Rather, it makes sense to recognize how propositions and truth claims arise within an interrelation between human practices and practical objects that is always already in effect. We also need to acknowledge that the discursive justification of problematized truth claims is unavoidably indexed both to this interrelation and to a societal principle of logical validity that is also always already in effect.
How, then, should one not be an anti-realist? I have suggested three ways. One should not be a radical contextualist à la Rorty, for this forces one to give up a substantial concept of propositional truth. One should not be a radically theist anti-realist à la Plantinga, for this compels one to remove the concept of propositional truth from contexts of discursive justification. And one should not be a Janus-faced pragmatic realist à la Habermas, for this reinscribes the gap between propositional truth and discursive justification that his consensus theory and formal pragmatics sought to remove.
Instead, one should subsume the insights of contemporary anti-realism into a post-anti/realist conception of propositional truth. If we follow this path, then we can also recontextualize the concepts of propositional truth and discursive justification within a broader conception of truth and authentication. By sorting out the many matters of truth that go beyond propositions and argumentation, such a conception will demonstrate why truth matters.
1 This chapter was originally published under the same title as an article in Philosophia Reformata 77 (2012): 1–18.
2 Alvin Plantinga, “How to Be an Anti-realist,” Proceedings and Addresses of the American Philosophical Association 56 (1): 47–70.
3 Ibid., 47.
4 Ibid., 62.
5 Ibid., 64.
6 Ibid., 66.
7 Ibid., 67–8.
8 Ibid., 68–70.
9 Although this is not the occasion to discuss Plantinga’s theistic anti-realism in depth, the positions he took in his 1982 presidential address have remained central to his work. Kevin Diller rightly observes that from the 1980s onward Plantinga has consistently identified “creative anti-realism” and “naturalism” as “the primary contemporary rivals to Christian thought, though sometimes giving more emphasis to the offspring of creative anti-realism: ‘relativism and anti-commitment.’ ” In a lecture given at a 1994 symposium dedicated to the memory of Herman Dooyeweerd, Plantinga forthrightly claimed that naturalism and creative anti-realism not only are inconsistent with Christianity but also “have no place for the notion of truth” – unlike theism, which allows Christian philosophers to make “attractive and useful” contributions to philosophical conversations on truth and knowledge. Plantinga’s positions in “How to Be an Anti-realist” remain important in more recent discussions of truth theory as well. For example, the late William Alston, in a paper prepared for a major analytic anthology on truth, cites Plantinga’s paper as a key backup to Alston’s own “extensional” argument against Putnam’s “ideal justifiability conception” of truth. Similarly, although without citing Plantinga, David Efird’s 17 August 2009 online review of the book Truth and Truth-Making, ed. E.J. Lowe and A. Rami in Notre Dame Philosophical Reviews, uses Plantinga’s move to explain the (human-) mind-independence of propositional truth: “I propose that a proposition’s being true is metaphysically grounded in, or metaphysically explained by, God’s knowing that proposition.” Hence the 1982 presidential address is a good place to start if one wants to understand Plantinga’s own contributions as well as some central issues in contemporary truth theory. Kevin S. Diller, “The Theology of Revelation and the Epistemology of Christian Belief: The Compatibility and Complementarity of the Theological Epistemologies of Karl Barth and Alvin Plantinga,” PHD dissertation (University of St Andrews, Scotland, 2008), 57–8; accessed 9 January 2012, http://hdl.handle.net/10023/497. Alvin Plantinga, “Christian Philosophy at the End of the 20th Century,” in Christian Philosophy at the Close of the Twentieth Century: Assessment and Perspective, ed. Sander Griffioen and Bert M. Balk (Kampen: Kok, 1995), 47–8. William Alston, “A Realist Conception of Truth,” in The Nature of Truth: Classic and Contemporary Perspectives, ed. Michael Lynch (Cambridge, MA: MIT Press, 2001), 59.
10 Michael Lynch, “Realism and the Correspondence Theory: Introduction,” in The Nature of Truth, 9.
11 Ibid., 11.
12 Plantinga holds something like this with regard to divinely known truth, although he speaks only of God’s thinking and believing a proposition, not of God’s offering a justification for thinking and believing as God does.
13 That is one reason why Plantinga frames his criticism of Putnam and Rorty’s epistemic conceptions in terms of their purported “creative anti-realism.”
14 The consensus theory receives its most expansive articulation in the Christian Gauss Lectures Habermas gave at Princeton University in 1971 and in the seminal article “Wahrheitstheorien” (Truth Theories) published in 1973. See “Reflections on the Linguistic Foundation of Sociology: The Christian Gauss Lectures (Princeton University, February–March 1971),” in Habermas’s On the Pragmatics of Social Interaction: Preliminary Studies in the Theory of Communicative Action, trans. Barbara Fultner (Cambridge, MA: MIT Press, 2001), 1–103, and “Wahrheitstheorien,” in Wirklichkeit und Reflexion: Walter Schulz zum 60. Geburtstag, ed. Helmut Fahrenbach (Pfullingen: Günther Neske, 1973), 211–65. Habermas has never allowed “Wahrheitstheorien” to be published in an English translation. He indicates his subsequent reservations about his “consensus theory” in the 1983 additions he inserted into the footnotes to both the Gauss Lectures and “Wahrheitstheorien” in his Vorstudien und Ergänzungen zur Theorie des kommunikativen Handelns (Frankfurt: Suhrkamp, 1984).
15 In the writings under consideration, Habermas identifies “intelligibility” as a fourth unavoidable validity claim, but in subsequent writings he drops it from the list and treats it instead as a general condition for communicative action as such.
16 Habermas, On the Pragmatics of Social Interaction, 89.
17 Jürgen Habermas, The Theory of Communicative Action, trans. Thomas McCarthy, 2 vols. (Boston: Beacon Press, 1984, 1987).
18 The shift to a formal pragmatics of meaning is apparent in Habermas’s 1976 essay “What Is Universal Pragmatics?” in Jürgen Habermas, Communication and the Evolution of Society, trans. Thomas McCarthy (Cambridge: Polity Press, 1984), 1–68, 208–19. Although it does not thematize truth, this essay marks a transition from stage one to stage two in his theory of truth.
19 Habermas, The Theory of Communicative Action 1, 277.
20 Ibid., 318.
21 Barbara Fultner, “The Redemption of Truth: Idealization, Acceptability and Fallibilism in Habermas’ Theory of Meaning,” International Journal of Philosophical Studies 4 (2, 1996): 238.
22 Hilary Putnam, Reason, Truth and History (Cambridge: Cambridge University Press, 1981), 55.
23 Jürgen Habermas, “Richard Rorty’s Pragmatic Turn,” in Richard Rorty and His Critics, ed. Robert B. Brandom (Oxford: Blackwell, 2000), 31–55. The same essay can be found in Jürgen Habermas, On the Pragmatics of Communication, ed. Maeve Cooke (Cambridge: Cambridge University Press, 1998), 343–82. Worth noting is that Brandom chose Rorty’s “Universality and Truth” (1–30), Habermas’s essay, and Rorty’s “Response to Jürgen Habermas” (56–64) to open Rorty and His Critics. Despite their disagreements about truth, Rorty was highly appreciative of Habermas’s democratic politics and his emphasis on the communicative character of rationality.
24 Habermas, “Richard Rorty’s Pragmatic Turn,” 40.
25 Ibid., 41.
26 Jürgen Habermas, Truth and Justification, ed. and trans. Barbara Fultner (Cambridge, MA: MIT Press, 2003). With the exception of two essays, this book is a translation of Habermas’s Wahrheit und Rechtfertigung: Philosophische Aufsätze (Frankfurt: Suhrkamp, 1999). My commentary focuses on the book’s “Introduction: Realism after the Linguistic Turn” (1–49). The German title of this essay speaks not of the “linguistic” turn but of the “language-pragmatic” (sprachpragmatisch) turn.
27 Ibid., 2.
28 Ibid., 3.
29 Ibid., 11.
30 Ibid., 15.
31 Ibid.
32 Ibid., 16.
33 Ibid., 15.
34 In the passage under consideration (33–6), Habermas does not explicitly mention his formal pragmatic conception of meaning. But it is implied throughout. Habermas appears to have been persuaded by Albrecht Wellmer and especially by Cristina Lafont that one must give up the tendency in the German hermeneutical tradition to think that meaning determines reference. See Cristina Lafont, The Linguistic Turn in Hermeneutic Philosophy, trans. José Medina (Cambridge, MA: MIT Press, 1999).
35 Truth and Justification, 33. Habermas employs Hilary Putnam’s theory of direct reference in order to account for such constancy of reference.
36 Ibid., 35.
37 Ibid.
38 Ibid., 36.
39 “Argumentation remains the only available medium of ascertaining truth since truth claims that have been problematized cannot be tested in any other way. There is no unmediated, discursively unfiltered access to the truth conditions of empirical beliefs.” Ibid., 38.
40 Ibid., 39.
41 Ibid.
42 Ibid., 40.
43 Ibid., 254.
44 Ibid., 255.
45 These are not the only questions needing better answers, of course. In addition to the writings by Barbara Fultner and Cristina Lafont cited earlier, see James Swindal, Reflection Revisited: Jürgen Habermas’s Discursive Theory of Truth (New York: Fordham University Press, 1999) and Maeve Cooke, “Meaning and Truth in Habermas’s Pragmatics,” European Journal of Philosophy 9 (April 2001): 1–23.
46 I have presented these conceptions on other occasions and do not repeat the details here. The relevant publications, in chronological order, include Artistic Truth: Aesthetics, Discourse, and Imaginative Disclosure (Cambridge: Cambridge University Press, 2004), especially 77–100; Social Philosophy after Adorno (Cambridge: Cambridge University Press, 2007), especially 77–106; “Truth Matters: Heidegger and Horkheimer in Dialectical Disclosure,” Telos 145 (Winter 2008): 131–60; “Unfinished Business: Toward a Reformational Conception of Truth,” Philosophia Reformata 74 (2009): 1–20; and “Religion in Public: Passages from Hegel’s Philosophy of Right,” University of Toronto Journal for Jewish Thought 1 (April 2010), http://cjs.utoronto.ca/tjjt.
47 Habermas, “Richard Rorty’s Pragmatic Turn,” 41.
48 Zuidervaart, Artistic Truth, 88–94; “Unfinished Business,” 10–15. In the vocabulary of older reformational philosophy, the term “predicative availability” points to the “logical object function” of any entity or event about which we can talk and make assertions. In the vocabulary of classical philosophy, it refers to “properties.”
49 Habermas, On the Pragmatics of Social Interaction, 86; cf. “Wahrheitstheorien,” 211–12.
50 Although the content of what one asserts could be called an “assertion,” it is easy to confuse this usage with our using the term assertion to indicate a practice or speech act. So, like Habermas, I adopt the well-established convention of calling the content asserted a proposition.
51 Zuidervaart, Social Philosophy after Adorno, 101–6; “Unfinished Business,” 15–18.