Truth and Objectivity

views expressed dated: 2025-09

This Peircean view has it [...] that there is an unseverable connection between asserting a statement and claiming that it is true. But it also urges us to look to the practice of assertion and to the commitments incurred in it, so that we can say something further?something about what truth is.

—Cheryl Misak, "Deflating Truth: Pragmatism vs. Minimalism"

Discourse Transcendental Objectivity

This will be a short note, locating my views on truth and objectivity at least at a 30,000ft view. Consider what it takes to make any claim intelligible as an assertion of truth. To assert that "the cat is on the mat" is not merely to emit noises or to point at things in the world. Nor is it to merely opine. It is to undertake a commitment that makes one accountable to reasons, evidence, and challenge. It is to open oneself to a mutually recognitive act with other reasoners. To count as doing so, we presuppose certain high-level features of rationality that cannot be coherently rejected without undermining the very possibility of a discourse of claiming at all:

• First, correct concept use goes beyond any finite community's attitudes because it normatively binds us to rules that go beyond any finite community's attitudes.

  • Claiming, and therefore truth as well, presuppose that there are determinately correct and incorrect ways of applying concepts—that concept-use itself involves norms of correctness that transcend any individual's or finite community's mere taking something to be so. Without this presupposition, we could have no shared practice of reason-giving at all. Anyone could mean anything by any concept, and discursive practice would be impossible.
  • Both the modal nature of things in the world and the rule-based nature of commitments we make hold us to rules that go beyond any finite community's claimings.
    • When about things in the world, claimings presuppose rules that stem from the modal nature of ordinary empirical things. I cannot properly regard something as table salt and deny that it dissolves in water. In even engaging in a truth-apt process of claiming things about the world, I commit myself to rules that extend beyond any finite community's attitudes. In the Kant-Sellars vein, this modal structure of things is not something we discover empirically, but something that makes empirical discovery possible.
    • When about deontic statuses, claimings presuppose rules that stem from concepts of rational duties we have. I cannot hold two contradictory premises without being open to rational challenge. The same goes for failing to follow through on a promise.
  • In this sense, the rules and concepts come from us, but their proper application entails getting their hooks into something external to us that we are responsible TO for the correct application of those concepts (as we table salt and water). This is a correctness by responsiblity to social norms not an unexplained correspondence relation with the world.

• Second, To engage rationally (with reasons) with another is to mutually attribute rationality. It is to engage in an act of mutual recognition of one another as being able to see the better reason. I cannot see only my reasons as truth-apt without already committing myself to another's also being so.

  • On my account, what constitutes our rational normative duties in the first place is that we play a game where the buy-in to be seen as rational is to see the other as rational.
  • This accounts for why "Why be rational?" does not make sense as a question, because to even pose it is to attempt not to play a game one is already playing by asking for the reason at all.
  • In being rational at all, one already necessarily and implicitly takes a stance of seeing oneself as part of a collective of all rational beings (what some philosophers of group-rationality/intentionality call a "strong we-mode").

If follows from these that for any given claim, if it is truth-apt, it must be acknowledged that there is some determinate and unconditional truth about it that goes beyond any finite community's attitudes. Claiming-true at all presupposes that there is some determinate truth beyond any attitude or finite set of attitudes about a given matter.

• Without this presupposition, claiming anything at all would be unintelligible because claiming would collapse into mere attitude expression—we could never distinguish between "I/We believe P" and "P is true," and thus could never make sense of the possibility of being wrong, which is constitutive of what it means to make a claim at all.
• The same goes for the ultimate truth of some claim being conditional (unless of course it is a conditional-claim). In the claim "The number of times Cleopatra Sneezed was either odd or even", I cannot sensibly claim that the true number depends on any other thing.

Through the first discourse-transcendental presupposition, we secure that there is some determinate correctness of evidential standards (receptivity to the world) of concept application and rule-based standards of concept application, domain-dependent and complex though they both may be. There are certain concepts the world simply will not let me have, like table salt $NaCl$ that does not dissolve in water. Through the second presupposition, we secure that the rules of the game are constitutive, not optional. Following from these, we secure that relativism could not possibly be the case without a radically mistaken account of claiming and therefore of truth. I am not arguing for a metaphysical account of truth, but rather a procedural concept of truth that emerges from our practices, yet necessarily reaches beyond them.

Truth on This Account:

According to this account, if rational inquiry (all rational inquirers included), following the transcendentally necessary norms that constitute discourse itself, would render X indefeasible in the infinite limit, then X is true. Truth in this sense transcends any actual justification, but not all possible (subjunctively) justification. This makes sense of the unconditionality of truth without divorcing it from our practices via some unexplained metaphysical relations of "reference" or correspondence.

Following Cheryl Misak's interpretation of C.S. Peirce, this is not a biconditional "definition" of truth, but only a making explicit of the necessary presuppositions of truth-practices and the intertwined concepts of inquiry, assertion, and objectivity. It is not a analytic definition.

Let $I$ stand for: What rational inquiry, following the transcendentally necessary norms that constitute discourse itself (what would be indefeasible in the limit).

• Asserted: $I$ $\rightarrow$ Truth
• Not asserted: Truth $\rightarrow$ $I$

The latter is only a hope or expectation, not an assertion. This makes the present account an explication of the necessary features of truth, not a definition of truth. This explication (not definition) of truth and truth-norms is inherently subjunctive and so it leans heavily on modality. This account is bolstered by the Kant-Sellars thesis of modality which claims that ordinary empirical content is not intelligible without a necessary modal component. Say, of an iron bar, that what makes it what it is is that it would rust if left in water.

It is only because the expressions in terms of which we describe objects, even such basic expressions as words for perceptible characteristics of molar objects, locate these objects in a space of implications, that they describe at all, rather than merely label. The descriptive and explanatory resources of language advance hand in hand.

-- Sellars, "Counterfactuals, Dispositions, and the Causal Modalities"

Since this understanding of truth entails an infinite horizon of inquiry, it serves as a regulative idealization of necessary presupposition of discourse, not as an actually achievable state. Since the view of truth here is a subjunctive conditional one, it is no worse off for this.

No Mysterious Reference/Correspondence Relation

I have briefly mentioned already that the view of truth here completely eschews any appeal to a correspondence or representation relation between claims and the world. Such a relation never has and likely never will receive an intelligible status, except perhaps in a naive sense as in "The cat is on the mat if and only if the cat is—in fact—on the mat." The kind of relation required here between truth and truth makers is like that of a picture and a pictured thing. For such a relation to have a useful and intelligible account of it requires us to be able to step outside of our practices and somehow compare the picture with the pictured thing in an unmediated way. This cannot be done. Further, even framing the notion of correspondence or reference this way leaves the gate swinging wide open for skepticism to come roaring in because it opens up a chasm between what could be true and how we could possibly know it.

Truth is Indefinable

The present account is not at all attempting to define truth. It only seeks to make explicit the necessary features of a concept of truth. Definitionally speaking, I think truth is as primitive of a notion as we have and cannot be given a full biconditional definition. This is not, however, a theoretical shortcoming, but is itself a feature of the transcendental boundary that truth in part constitutes. Truth is inextricably tied up in unrejectable notions of correctness and mutual rational recognition. It is therefore not possible to engineer some concept of truth which is entirely divorced from these normative notions of correctness that serve as a hinge, only in the presence of which everything else can turn.

Bonus: Accounting for Naïve Correspondence

Almost all substantive views of truth want to capture the intuition around a naïve correspondence view. The sentence "the cat is on the mat" is true, seemingly, if and only if the cat is on the mat. The present view is able to easily explain this naïve correspondence without invoking a lot of extra baggage. Such a sentence is true under naive correspondence because our mutually recognitive practices of making claims about "cats" and "mats" hold us rationally responsible to things in the world (cats and mats) for their correctness. To use these concepts correctly (establishing the basis for communication at all) just is to hold myself responsible in that way to cats and mats. Though such a claim may turn out wrong in some wild scenario, by the very nature of understanding cats and mats the claim is as clearly true as any obvious example.

Clarifications

Not Convergence

This is not a "convergence" theory at all as it does not depend in the slightest on what beliefs an actual or possible finite communities of inquirers arrives at. Nor is it an "ideal conditions" or "ideal assertability" view, in a sense that must be held accountable for what those conditions are (because it does not claim at all to be a definition, but only an explication).

Not Ideal Conditions/Assertability

This account differs fundamentally from "ideal conditions/assertibility" theories in several key respects. Unlike such theories, it does not attempt to specify epistemic conditions under which beliefs would be guaranteed to be true. The transcendental norms $(1 \& 2)$ are not "ideal" in the sense of optimized epistemic states, but rather constitutive features of what makes rational discourse possible at all. Instead of such a problematic specification, the present theory only appeals to beliefs/claims that would, in Misak's words:

fit with experience and argument and which would satisfy all of the aims of inquiry, no matter how much the issue was subject to experiment, evaluation, and debate.

The "infinite limit" functions as a regulative horizon rather than an achievable endpoint—there is no state we could reach and recognize as "ideal conditions." Following Misak's interpretation of Peirce, this approach avoids commitment to any "end of inquiry" where all questions would be decided or to "epistemically ideal conditions" that inquirers could acknowledge as such.

Again, this is an explication rather than a definition of truth. Therefore, it carries no burden to specify what ideal conditions would look like. "Ideality" is not invoked here, only the trasncendentally necessary features something must have to count as asserting/claiming at all. Nothing in the present account:

 IF rational inquiry, following the transcendentally necessary norms that constitute discourse itself, would render X indefeasible in the infinite limit, THEN X is true.

commits us to ideal conditions/assertability of any kind any more than any other claim would. More on this issue shortly.

A Justificationist View and a Note on Anti-Justificationist Views

The present view, though it focuses on indefeasibility, is not an anti-justificationist view, though it does share some fallibilist affinities with such views. Calling that which is indefeasible in the limit truth does not imply that there is no role for justification. Something will be infinitely indefeasible necessarily because there are reasons for it. It is not intelligible that something be true with no reasons for it and only reasons against other views. This is true for two reasons. The first is that giving reasons against is not a game one can play without being able, in principle, to play the game of reasons for. Second, there are an infinite number of theories (all with the possibility of being undetermined with respect to the evidence). For theories to only be in epistemically good standing by virtue of reasons against other theories, those reasons against the other theories would have to be infinite and arbitrary. Infinite because one can easily infinitely generate theories, and arbitrary because of the selection we must make against equally underdetermined theories.¹

An Epistemic Account?

This is in an important sense an epistemic account of truth and in another important sense not. It is an epistemic account insofar as it is a substantial account of truth, not a deflationist one, and also not a metaphysical account. However, this account of truth hinges only on what must be presupposed to render assertoric/claiming practices intelligible without radical consequences.

Circularity Charge?

An important objection that truth-accounts like the present one face goes as follows. It is claimed that if

 rational inquiry (as such) would yield X as indefeasible in the limit 

then X is true. It is then objected that this is circular on the grounds that we have snuck in a notion of truth already. If either "rational inquiry (as such, correctly done according to the above presuppositions)" or "indefeasible in the limit" have "truth-generating" sneakily baked into them, then the account is circular. Let us see if this is the case for the present view.

Does "rational Inquiry (as such)" Presuppose the Concept of Truth?

Examining the above detailing of these necessary presuppositions ($1,2$), it must be shown that none presupposes a concept of truth. $1$ and $2$ have to do with correctness of concept application and the mutual attribution of rationality. Neither presupposes a concept of truth. Correctness of concept application and we-mode mutually of meaning are certainly related to truth, but do not amount to any sort of hidden definition of truth. Further, merely $1 \& 2$ are not sufficient for truth. If inquiry stopped this instant, despite us all following the correct norms of concept application and mutual attribution of rationality, we would not have secured truth.

Does "indefeasible in the limit" Presuppose the Concept of Truth?

For an explication of truth, a specification of "ideal conditions" is problematic because it introduces a dilemma. As Robert Brandom puts it: there is

no way to specify the ideality in question that is not either question-begging (in implicitly appealing to a notion of truth) or trivial.

This dilemma assumes that the epistemic theories of truth in question must specify what makes some conditions more "ideal" than others. However, the present account escapes both horns of this dilemma by not specifying "ideal conditions" at all. The transcendental norms of belief and assertion are not selected because they are truth tracking, they are constitutive requirements of rational discourse.

The account of truth defended here: if rational inquiry, following the transcendentally necessary norms that constitute discourse itself, would render X indefeasible the infinite limit, then X is true. Does nothing to invoke ideal conditions, thus avoiding the circularity.

Perhaps something's subjunctive status as being "indefeasible in the limit" is secretly hiding the concept of truth? This allegation is usually made by considering a holist picture of meaning and adding additional background assumptions to falsify a claim. On a holist view, adding more auxiliary claims changes the meaning of another claim.

Suppose we have

Table salt NaCl  dissolves in water.

Now suppose we stipulate that this claim is indefeasible in the limit. To the holist, this means that if we add further background assumptions (which change the meaning of the claim), then we should still see it as indefeasible. We can add two kinds of claims, true ones or false ones. If we add false ones, so the objection goes, then the claim is defeasible because we can add a claim like "the water in this pot is not real water" (even if it is) and then the salt will dissolve and our claim is defeasible. What if we say "Ok, fine you can only add true claims"? Then we have clearly made a circular move. So it seems, says the objection, that in trying to make sense of "indefeasible in the limit" we had to invoke "truth". The objection is that in order to define "indefeasible in the limit" we have to stipulate that the beliefs-added are "true" (a circular move) not "false" or else "indefeasible in the limit" beliefs could wind up false.

This objection does not apply to the present account. It does not apply precisely because "would be indefeasible in the infinite limit" does not invoke ideal conditions. It merely claims that were inquiry (understood as it must be as rational discourse) to be carried forward infinitely, what would be indefeasible going forward would be truth.

The key to this objection is the question "When is it ideal conditions?" or "When is it inquiry in the infinite limit?", if the answer is "just when X is true" or "only when new information added is true", then the account is circular. To see why the objection doesn't work here, we need to look again at the subjunctive conditional explication of truth on offer here. Again that is:

*if* rational inquiry, ... , *would* render X indefeasible in the infinite limit, *then* X is true. 

We do not need to specify de dicto, when the infinite limit of inquiry is reached (that is impossible), we only need to specify de re the possible subjunctive conditions for it. If this de re specification itself baked in a substantive notion of truth, that would be a problem, but as we have seen above, it does not.

To the question "When is it inquiry in the infinite limit?", we can respond: that is when the constitutive norms of rational inquiry ($1 \& 2$) have been carried out in the infinite limit. All the present account needs to show is that norms $1 \& 2$ can be specified and applied as transcendental conditions for discourse (and carried forward in the infinite limit) without presupposing the concept of truth we're explicating. Since the correctness of the rational norms that underwrite the possibility of discourse cannot coherently be questioned, and hence neither can their carrying forward as an infinite regulative ideal, the apparent circularity reflects constitutive structure rather than a vicious theoretical regress.

A further counterobjection is that this kind of circularity is only a threat if a full-blooded analytic definition is being attempted. It is not incoherently circular to claim that "if an organism photosynthesizes, it turns sunlight into usable energy" as a kind of expository statement in teaching someone about photosynthesis. This claim, like $I$ $\rightarrow$ Truth, attempts only to be an explication of an important aspect of truth, among others. It does not claim to allow us to detect and identify true claims, it only claims to show what would partially constitute any true claim. All that is being shown in this note is that an important part of truth is made explicit by the $I$ $\rightarrow$ Truth conditional.

A Note On Deflationism

It is important to stake out my position in opposition to deflationary conceptions of truth. Since, on this view, claimings themselves presuppose truth, any attempt to reduce truth to merely claiming (disquotation, prosentences, etc) fails to see that truth is a necessary feature of claiming that reaches further than claiming. To claim is to claim-true, and to say of something "that it is true" is to endorse it, but that does not mean that there is not anything more to truth or that truth comes from some arcane place beyond our practices.

I would agree with most deflationists that "claiming" is prior in the order of understanding to truth. In being able to rationally claim something, one functionally already knows all one needs to know about truth. I would also agree with deflationists that there is no metaphysical element to truth that reaches beyond our practices. But can deflationism underwrite everything necessary for a normative and objective notion of truth?

Deflationism cannot ground the objectivity of rational discourse because it fails to see truth—born of the presuppositions necessary for rational discourse—as a contentful ingredient that was there from the start. Without the idealizing presupposition of a contentful truth, deflationism cannot account for the two important aspects of truth: normativity and the unconditionality of truths. Deflationism fails to explain these features as features of truth and so does not enable us to say that there is some determinate truth about things which can be found by constitutive features of our epistemic machinery. This is indicative of a deep mistake in my view. I am alleging that the mistake is attempting to make claiming do all of the epistemic work without seeing that a contentful notion of truth is presupposed by claiming.

This content need not be identical or identifiable in all true claimables as Davidson points out in "Truth Rehabilitated":

Truths do not come with a "mark", like the date in the corner of some photographs, which distinguishes them from falsehoods.

Is it intelligible to claim that "Cleopatra sneezed an odd number of times in her life" with there being no way in principle to settle the question? If there is necessarily some way in principle of settling a given truth-apt question, then there is something more substantial to truth that points to the correct application of presupposed epistemic norms. If truth is just disquotational or prosentential, then there is no robust sense in which inquiry aims at mind-independent facts. This can only be, at best, an incomplete account of truth-norms. Deflationists can be equally committed to objectivity, and indeed many are committed to a similar conception to what I have outlined here. However, I think it is a staggering oversight not to see objectivity as inherently a part of an epistemic correctness that points us towards truth.

Not to call the products of that transcendentally required correctness "truth" seems to be splitting hairs and would amount to little more than an allergic reaction to "truth talk".

Deflationism fails to account for objectivity, not because it lacks some metaphysical notion of representation, but because it cannot account for the necessarily intertwined nature of objectivity and truth as threads in one braid. Where I part with deflationists is in either saying prosentences or disquotation are "all there is to truth". Or in saying that truth can be reduced to—or extracted only out of—claiming. Claiming is not exhaustive of all there is to truth any more than "locating" is exhaustive of all there is to space. Claiming presupposes truth. Locating presupposes space; it does not comprise it. The correct locating of a thing in space is inextricably linked to what it means for a thing to have location.

A Social Objectivity, Not a Metaphysical One

The above transcendental requirements give us truth as an infinite limit or infinite horizon concept, one which we need never actually attain, but which is determinate and transcends mere social agreement. The objectivity of claims, then, does not require us to access some "view from nowhere" or metaphysical correspondence to mind-independent facts. It emerges from the structure of transcendental norms of inquiry and concept-application—norms that are not optional but constitutive features of rationality itself.

When we engage in the practice of giving and asking for reasons, we are already committed to there being facts which stand free of our attitudes. The only objectivity we should want emerges from following these inescapable norms of rational inquiry, not from some unexplained representational relation.

The crucial move is recognizing that objectivity can be instituted by our practices while remaining objectively binding on us in the strongest possible sense. We cannot coherently opt out of these requirements while continuing to engage in the practice of making claims at all.

A Pragmatist's Truth & Objectivity

This gives us all the objectivity we need: claims that reach beyond any finite community's attitudes, that constrain what we can rationally believe, and that allow for genuine progress—all without requiring unexplained correspondence to a reality independent of the practices through which we make such judgments intelligible and without mistakenly equating causal natural-world properties with epistemic ones.

This account has powerful advantages over alternative accounts, chief among them being that it can secure objectivity without divorcing that truth from our practices of giving and asking for reasons.

Does Fallibilism Mean We Never Know the Truth?

The present account comes with a heavy helping of fallibilism. If only what is indefeasible in the infinite limit is true, then it seems we cannot know the truth truly. The $JTB$ account's $T$ condition could never truly be satisfied.

An objection to the present account might go that since it cannot account for knowing that "the cat is on the mat", it has taken a wrong turn somewhere. I do not find this objection convincing because I think it assumes that there is some deeper, arcane grasp one could have on things beyond our most indefeasible (so far) ideas.

The $T$ condition of $JTB$ on the present account can only ever be a hope, as it may yet be rendered defeasible. Surely I know that the cat is on the mat?! But what can I possibly mean by this, which could extend infallibly beyond our practices?

Unmoved Movers

For any substantive account of truth, there must be an unmoved mover somewhere. There must be some hinge on which everything turns. For correspondence accounts, this is the unexplained representation/reference relation. On the present account, it is the discourse-transcendental features we must commit to implicitly to give/ask for reasons at all. These are not correctable claims themselves, but inalienable features of anything that could count as giving and asking for reasons at all. Thus, the unmoved mover here is only what is required to make sense of our practices.

If the unmoved mover here is our practices, can't those change as well? They could, but this would not imply on the present account that truth would change. Quite the opposite is true. As long as there are practices with the $1\&2$ features above, then there is truth and objectivity baked right into them. If the practices changed to no longer accomodate those norms, then those practices would not intelligibly count as truth-apt at all.

Is this view a kind of justificationism then? It is not, because it does not state that there are unjustified claims that ground truth and objectivity, but rather that there are transcendentally necessary propositions that do so, without which one cannot make any claims.

Footnotes

¹ This is simply not how it works in the real world (we constantly give and ask for reasons for things). The problem with shoving justification (giving reasons for claims) aside and countenancing only claims against is a resulting radical misunderstanding of the essential practice of abduction. Theories are not picked from a vacuum and equally weighted until evidence rolls in. These objections have been historically devastating for anti-justificationist views. But I think the practical problem goes deeper. When we only allow reasons against in combination with something like the "infinite limit" proposed here, we have essentially generated an epistemic "get out of jail free" card to believe whatever we want with the hope that it is a bold conjecture that will withstand all reasons against it (just give it more time).