Habermas begins to clarify the relationship between truth and justification relative to the “semantic” (deflationary) and an epistemic approach to truth. I’m going to examine his epistemic sense of truth later.
Tarski’s Convention T is the matter at hand for JH (though deflationary theory is not just about Tarski), so that’s all we have to care about, as to what the “semantic” theory is about, relative to JH’s interest in using a semantic level of concern to make a differentiation between that and the pragmatic level.
T: “P” is true iff p.
Standardly, truth theory seeks the simplest practicable examples (instantiations of P). JH uses an infelicitously complex example (instantiation) for P, so let me use a simpler example:
‘The rain outside continues’ is true iff the rain outside continues.
Habermas uses the formulation T above, but I think that this is misleading. The deflationist has in mind:
T1: “P(s)” is true iff s is the case, where s is the state of affairs denoted by P.
I may know that s is the case by looking out the window, having remembered beforehand that the rain earlier continued, such that I’m now feeling anxiety about not having my umbrella in hand. But alas, the rain has stopped; I don’t need my umbrella, and P(s) is false. Knowing whether or not the rain continues involved nothing linguistic.
But JH claims that Convention T (my T1) is a 2-part linguistic condition.
...the truth-predicate when used in this way establishes a relation of equivalence between two linguistic expressions---the whole point of the Tarskian strategy of explanation depends on this. (OPC 361)
Sure, Convention T depends on such a relation; but it’s about a linguistic and non-linguistic state of affairs. JH recognizes this:
...what is envisaged in every case are pictures of relations that extend beyond language. (ibid.)
Our relations to states of affairs can be (and often are) independent of the linguistic representation of states of affairs. Perception, broadly conceived (intelligence generally), is richly non-linguistic. ‘True’ pertains to the lingustic relation to non-linguistic states of affairs, such that T1. If I’m blind, I would have to hear the rain or stick my hand out the window to know that s is the case. Non-linguistic animals, too, may know that the rain continues and may choose to stay dry (which is why the cat is still on the mat, and the birds are not flying). (Animals have knowledge and make cognitive determinations without language.)
So, I could agree that:
...the disquotational function...already presupposes the representational function. (ibid.)
Though, it seems redundant to say “already pre-”. But it’s not the case for linguistic reasons that:
...the disquotational function is not sufficiently informative because it already presupposes the representational function.
Recognizing that the rain continues supposes the perceptual function, which is nothing more than to perceive rain continuing. T1 supposes decidability, including perception of states of affairs, but that’s not a linguistic supposition; it’s a cognitive supposition (represented as a cognitive supposition proven epistemically valid by experience, which we usually convey to others linguistically, but could do through gesture, e.g., pointing to the window, in order to not wake the baby).
An infant may understand what rain is, even though s/he has no linguistic representation of this. He perceives the rain continuing and infers that he will get wet if he sticks his arm beyond the cover of the stroller cover. An instrument in a room may indicate precipitation, such that a computer that is connected to the instrument flashes to a deaf person the message “The rain continues.”
So, there is a distinction between understanding and meaning in at least this sense: Cognitive truth—deflated trueness—is non-linguistic.
JH seems to conceal this, for his purposes, in writing that:
One understands the meaning of Convention T when one knows what is meant [sic] (gemeint) [sic] with the right-hand side of the biconditional. (362)
Is this different from saying that one understands Convention T, etc.? That is: What’s the difference between understanding X and understanding the meaning of X? JH is making the point that understanding the meaning of a compound T requires understanding the meaning of its components. Understanding anything may require understanding its components, but that doesn’t make components the same logical type as the gestalt. (Besides, the gestalt nature of perception and understanding is such that the differentiation of components is usually a separate act; JH is stipulating a situation that is abstracted from basic experience, as if there’s a linguistic atomism prevailing).
Of course, one has to be disposed to find out if the rain continues, in order to evaluate whether of not the assertion of this holds, pertains, or is cognitively good, i.e., valid. So, yes:
...“the meaning of the truth-predicate in the sentence...is parasitic on the assertoric mode....
But whether or not the indicated state of affairs exists (i.e., is the case) is not parasitic on linguistic cognizance of it. What’s “deflationary” about Convention T is that it expresses no dependence on a “thick” notion of truth as a property; rather it displaces the question of what is meant by ‘true’ to the variable means of satisfying interest in whether or not s is the case (i.e., engaging in practices, enactivity of finding out ). “P(s) is true” need mean nothing more than that: Indeed: s. If you want to know whether “The rain continues” or not, get real: Find out. The deflationist point seems to be that the “truth-predicate” has no ontic merit.
...with respect to the pretheoretical orientation toward truth inherent in everyday practices, a semantic conception of truth does not help us at all. (363)
Not true, unless one simply presumes that all evaluative action is linguistic, as if valuation is intrinsically linguistic. But that’s a view undermined by developmental cognitive science (including the cognitive linguistics of George Lakoff, Ray Jackendoff, and others, as well as conceptions of embodied cognition within cognitive science of intentional action: Why do we act?). “[A] weak realist supposition of a world independent of our descriptions” may keep a semantic theory pertinent to science, JH notes (OPC 362), and such a supposition seems quite ordinary, from an “everyday” perspective (especially for the semiotic dailiness of valuational lives). We represent our valuative and evaluative interest and confess our realism about things not always being the case by expressing desire to know whether or not “s” holds. Reliabilism in epistemology seems to give credence to a semantical/semiotic theory of valuation. Besides, “methods of inquiry and theory selection” (362) are an extension of everyday action policy formation competences. Daily evaluation of what holds good gets formalized as processes of methodic inquiry.
So, a semantic/semiotic conception may be quite helpful for understanding the lifeworld character of valuing. It seems to me quite useful to include a semantic/semiotic sense of truth-predication in a general approach to truth that is appreciative of everyday valuation.
I take the stance toward Convention T that I do because I’m not largely a deflationist about truth; I find deflationism useful for raising epistemic presumptions about our valuational (or axiological) interest in what’s the case. I’ve accepted long-forgotten arguments from others that deflationism is untenable as a general theory of truth (or as a theory about the lack of content in truth-predicates), such that I’ve not given much attention to deflationism strictly speaking. I have given close attention to what Habermas says about deflationism in “Rorty’s Pragmatic Turn” (section of that chapter titled “Truth and Justification”), I believe, relative to his interest in a pragmatic view, relative to my interest in a pluralist view, and relative to my background concerns about JH’s “reading” of the lifeworld. I would so welcome thoughtful disagreement, inasmuch as I seem to be misreading Habermas.
digression on linguistic analysis
Deflationary critique of truth theory seems to be part of the 20th century trend in philosophy that finds ordinary life to be a fruitful resource for philosophical investigations. A deflationary interest seems to be kindred with the phenomenological interest in getting over (or away from) traditional baggage (or onto-epistemically thick presumptions), “back to the things themselves,” and to appreciation of the experiential basis of understanding, if only as a therapeutic against ontologism that retrieves a sense of tenability toward the world as it is. Habermas seems to
tend to dismiss the basic action-theoretical and phenomenological motive of English philosophy of language, even as he grounds formal pragmatics is speech action. I would argue that English philosophy of language wasn’t linguistic-relativist in motive (though there was quite enough linguistic relativism among disciples!). It was action-theoretically phenomenological in motive, finding new ways of phenomenal inquiry through linguistic analysis. (Responding to the question "Does God exist?" with the question "What do you mean by capped-'God'?" is not implying any linguistic relativism, rather: putting the cognitivity of the question into reflective question.)
I would not be surprised to learn that Austin and Wittgenstein read Husserl. (In fact, I saw kindredness between the latter two as an undergraduate. Surely, they saw each other as kindreds in post-positivist inquiry into the nature of our conceptuality.)
If you call language-centric philosophy of the 20th century a pragmatic turn in interest (which seems accurate to me), it’s a turn that isn’t as such relative to linguisticality, rather relative to an ordinary-actional turn through focus on the prevailing conceptuality between us: linguistic—our conceptuality through language, not necessarily an essentially linguistic conceptuality. To me, Ordinary Language Philosophy (J.L. Austin, the neo-Wittgensteinians, and that legacy) is more associable with an action-analytic turn in conceptual inquiry, of which the overtly linguistic turn is a part. Comprehending the grammar of language is a prelude to capturing the “grammar” of cognition, which became cognitive science (and the trope of grammar is antedated by models of cognition, exemplified by cognitive linguistics). An action-theoretical interest invites greater cognitivism; a linguistic-theoretical turn may not as easily find its way into cognitive-scientific modeling.
Not all language is linguistic! Intelligent action is not reducible to enacting linguistic representations. Individuating intelligence is not axially about linguistic socialization. Yet, analysis of linguistic means—the ordinary language romance of mid-20th century philosophy—provided phenomenological resources that deflate ontic pretenses, without implying linguistic relativism (though strongly bolstering dispositions of understanding that are linguistically relative, because one may prefer this, due to research interest).
I don’t read Habermas’s formal pragmatics as linguistcially relativist, rather cognitivist. But I find him not seeing the cognitivist/semiotic potential in his own appropriations (i.e., formal pragmatics as such), which can be more congruent with cognitive science than he may anticipate. He is communicative-theoretical because his interest and audience is social-theoretical or sociocentric (whereby publically critical issues would tend to be eco-social or sociocentric). But he tends to assimilate within his theory of action equiprimordial action types into being relative to overriding communicative action—though he says that he doesn’t want to be read to be doing so. He can indeed be read to not be communication-theoretically overbearing; but does he appreciate as fully as possible how that reading can be sustained without compromising his prevailing interests?
Individuated cognitivity is not fundamentally linguistic, because intelligence is not primordially linguistic. But his Meadian reading of individuation is unduly sociocentric, which I’ve mentioned, in terms of others' work: search 'Mead' here. Not even G. H. Mead was sociocentric!
I find no difficulty accomodating Habermas’s communicative-action theory to a cognitivist view, just as theory of intelligence may easily accomodats cognitive linguistics without becoming a linguistic-relativist conception of intelligence.
Next: “‘pragmatic’ truth”
Also: This discussion is associated with the “conceptual inquiry” area of gedavis.com.