Daimon. Revista Internacional de Filosofía, nº 89 (2023), pp. 149-162

ISSN: 1130-0507 (papel) y 1989-4651 (electrónico) http://dx.doi.org/10.6018/daimon.480391

Licencia Creative Commons Reconocimiento-NoComercial-SinObraDerivada 3.0 España (texto legal). Se pueden copiar, usar, difundir, transmitir y exponer públicamente, siempre que: i) se cite la autoría y la fuente original de su publicación (revista, editorial y URL de la obra); ii) no se usen para fines comerciales; iii) se mencione la existencia y especificaciones de esta licencia de uso.

Deflationary Truth, Ordinary Truth and Relative Truth

Verdad deflacionista, verdad ordinaria y verdad relativa

RAMIRO CASO*

Abstract. Horwich (Mind 123(491), 2014) has argued that only someone with inflationary tendencies could feel inclined to endorse truth relativism. In doing so, he argues that deflationism about truth entails the denial of relativism. If sound, Horwich’s argument could entail that truth relativism is incompatible with any conception of our ordinary truth predicate according to which there is some sort of equivalence between a ground-language claim that p and the corresponding claim that p is true. Arguably, any story the relativist might give about our ordinary truth predicate should entail some such equivalence. Hence, there is a problem for truth relativism that needs to be addressed.

Keywords: truth relativism; radical relativism; deflationism.

Resumen. Horwich (Mind 123(491), 2014) ha argumentado que solo alguien con tendencias inflacionistas podría sentirse inclinado a adoptar el relativismo de la verdad. Al hacer eso, argumenta que el deflacionismo acerca de la verdad implica la negación del relativismo de la verdad. Si el argumento de Horwich es sólido, podría de hecho implicar que el relativismo de la verdad es incompatible con cualquier concepción de nuestro predicado de verdad ordinario de acuerdo con la cual hay algún tipo de equivalencia entre una afirmación de que p del lenguaje de base y la afirmación de que p es verdadera. Cualquier historia que el relativista podría contar respecto de nuestro predicado de verdad ordinario debería implicar una equivalencia semejante. Por lo tanto, hay un problema para el relativismo de la verdad que requiere atención.

Palabras clave: relativismo de la verdad; relativismo radical; deflacionismo.

---

Recibido: 16/05/2021. Aceptado: 07/07/2021.

* Instituto de Investigaciones Filosóficas SADAF-CONICET, Assistant Researcher / Philosophy Department, Universidad de Buenos Aires, Jefe de Trabajos Prácticos / caso.ramiro@conicet.gov.ar. Research lines: truth relativism, deontic and epistemic modals, expressivity, slurs. Recent Publications: Caso, R. (2020), “A bidimensional account of slurs”, in Eleonora Orlando & Andrés Saab (eds.), Slurs and Expressivity, Lanham, Lexington Books, pp. 67-93; Caso, R. (2017), “Los compromisos normativos de la aserción relativista”, Análisis Filosófico 37(2), pp. 143-168.

 

1. Introduction

In (2014), Horwich concludes that ‘only those philosophers with initial attachments to one or another inflationary conception of ‘‘absolute truth and falsity’’ will have reason to say that certain propositions “can be true only relative to a perspective”’ (Horwich (2014, 736)). In arguing for this conclusion, he makes two separate claims: that certain inflationary conceptions of truth entail that, for a certain range of propositions, truth is not an absolute property, but a property whose possession depends on a perspective of assessment, and that relativism about truth is incompatible with the correct (i.e., deflationary) understanding of our ordinary truth predicate. Both subsidiary claims are required for establishing Horwich’s conclusion: the former provides the conditional, if you are an inflationist (of a certain type), then you have reasons for being a relativist; the latter provides the conditional, if you are a deflationist, then you must reject truth relativism—consequently, you can be a relativist to the extent that you endorse an inflationary conception of truth. I have no quarrel with Horwich’s claim that certain inflationary conceptions of truth entail truth relativism. I do take issue with Horwich’s argument for claiming that deflationism entails the negation of truth relativism, for it seems that the piece of reasoning that Horwich deploys in order to arrive at this conclusion could be generalized, so that Horwich’s objection could actually be more general than it seems: if sound, the argument could actually entail that truth relativism is incompatible with any conception of our ordinary truth predicate according to which there is some sort of equivalence between a ground-language claim that p and the corresponding claim that p is true. Since any story the relativist may give concerning our ordinary truth predicate should entail some such equivalence, there seems to be a real problem for truth relativism here.

In what follows, I will briefly state Horwich’s case for the incompatibility between deflationism and truth relativism, pointing out how it can be generalized, and I will attempt to counter it in its generalized form. The question whether truth relativism is ultimately compatible with deflationism about truth, and to what extent (if any), turns out to be a complex question, for it can be read in one of two ways: whether the intuitive idea that truth is relative to an evaluative perspective is incompatible with the claim that our ordinary notion of truth is deflationary (Horwich’s concern in the first part of the paper), and whether the theoretical enterprise of providing a semantic theory based upon a theoretical notion of relative truth is compatible with deflationism (Horwich’s concern in the second part of the paper). In discussing Horwich’s incompatibility claim, I will provide an answer only the first concern, as I have dealt with the second part elsewhere (Caso (2016)).

2. An objection to truth relativism

Horwich starts by characterizing relativism about truth as the view that ‘within various domains of discourse […] there is no such thing as absolute truth, and truth applies only relative to a point of view’ (Horwich (2014, 733)). Hence, ‘although we can speak of a given assertion or belief as being true relative to this perspective but not relative to that one, there will be nothing in reality to make it simply true or simply false.’ Then, he argues that relativism about truth is incompatible with deflationism about truth. Since the terms in which the argument is stated are important, I will cite the passage in extenso:

However, as regards the concept currently deployed in everyday life, in the law, in logic, in science, and elsewhere, this characterization [i.e., that there is no such thing as absolute truth] appears to be quite wrong. For it can hardly be denied that our practice with the word ‘true’ is to treat the claim, ‘Blah blah blah’, (whatever it may be) as equivalent to ‘The proposition that blah blah blah is true’. Thus any statement that anyone ever makes is taken to amount to a statement to the effect that something is unqualifiedly true—and not merely true relative to a certain point of view. (Horwich (2014, 733–734))

That is, to endorse deflationism (which, for Horwich, is also the correct view on the ordinary truth predicate) involves holding (i) that our ordinary truth predicate is monadic—that is, that our ordinary truth predications are predications of truth simpliciter and not of truth relative to a perspective—, and (ii) that the ordinary truth predicate obeys an equivalence principle to the effect that it is always possible to move from the statement that p to the statement that the proposition that p is true and vice versa.1 Endorsing relativism requires us to reject a truth predicate characterized by these two features. Hence, relativism is, at best, a wrong philosophical view that springs from inflationary conceptions of truth and philosophical semantics.

A temptation that should be avoided right from the beginning is to attempt to solve this problem by arguing against deflationism. Dialectical destruction of this conception about our ordinary truth predicate would no doubt allow the relativist to avoid any need to show that relative truth is compatible with it. However, we could risk being drawn into a long discussion that is not going to be settled any time soon (or so it seems). An alternative way of dealing with this problem could be to hold that, insofar as the debate between inflationism and deflationism concerning truth is far from being over, the relativist would be entitled, dialectically at least, to ignore the challenge, should she be of inflationary tendencies. But this would not be a satisfactory answer either, not only because deflationism might be the right view to hold regarding our everyday truth predicate after all, but also because it is possible to generalize Horwich’s objection in such a way that, should it be correct, truth relativism would be incompatible not only with deflationism, but with a far larger range of conceptions about truth.

The sense in which Horwich’s objection might be more far-reaching is the following. The key ingredient of our everyday truth predicate that features in the above-quoted piece of reasoning is the idea that it obeys (some version or other of) an equivalence principle, which could be formulated as:

(EQ) The proposition that p is true if and only if p.

The distinctively deflationary contention, namely that this is (pretty much) all that can be said about our ordinary truth predicate, seems to play no role in the quoted passage. Moreover, it is not even required for the argument contained there to run that all instances of (EQ) are accepted: all that is required is that the equivalence holds for substituends of ‘p’ that contain no semantic vocabulary (which I will call sentences of the ground language).2 Hence, there is a way to generalize Horwich’s objection so that, should it be correct, it would entail that truth relativism is actually incompatible with any conception of our ordinary truth predicate according to which there is some sort of equivalence between ground language statements and the corresponding truth predications.

These would indeed be bad news for the relativist: regardless of what the correct story about our ordinary truth predicate turns out to be, it will most likely entail (some version or other of) (EQ), at least when restricted to sentences of the ground language as suitable substituends for ‘p’. Hence, if this generalization of Horwich’s objection is sound, relative truth could not coexist with our ordinary notion of truth: given that our ordinary truth predicate applies to the propositions expressed by sentences belonging to the domains of discourse for which truth-relativistic treatments have been offered, endorsing relativism about truth would seem to require us to replace our ordinary concept of truth with another.

3. The elements of the objection

Well then, how are we to assess Horwich’s reasoning? Under closer scrutiny, it consists of five central claims:

1. Truth relativism claims that, within certain domains of discourse, there is no such thing as absolute truth.

2. Our practice with the truth predicate is to treat a claim to the effect that p as equivalent to the claim that the proposition that p is true.

3. Hence, a statement to the effect that p is equivalent to a statement to the effect that something (i.e., the proposition that p) is unqualifiedly true.

4. Hence, there is such a thing as absolute truth in every domain of discourse.

5. Hence, truth relativism is incompatible with our ordinary practice with the truth predicate.

The crux of this argument is the transition from step 3 to step 4, since everything depends on it. Is this transition sound? Which is the underlying inferential or conceptual move?

The first thing we should note is that Horwich seems to be moving somewhat swiftly from the fact that ordinary truth predications are unqualified to the preliminary conclusion that they are absolute. However, there are two relevant senses of ‘absolute’ that ought to be distinguished at this point: ‘absolute’ understood as without explicit relativization and ‘absolute’ understood as not subject to perspectival variation. Deflationism (or more generally, the claim that our ordinary truth predicate obeys (EQ)) entails the claim that ordinary truth predications are unqualified, hence absolute in the sense that they lack explicit relativization. Truth relativism, on the other hand, does not (primarily at least) involve the claim that truth predications are (or ought to be) explicitly relational, but the claim that truth is subject to perspectival variation (so that the correctness of truth predications may vary according to the perspective of assessment). So, if step 4 is to serve as an adequate ground for the claim rendered in step 5, ‘absolute’ ought to be interpreted in the latter sense, the one germane to truth relativism. However, if step 4 is to follow directly from step 3, ‘absolute’ ought to be interpreted in the former sense, the one linking it to the absence of explicit relativity.

Once this situation is brought to the fore, the inferential move from step 3 to step 4 should be regarded as problematic: How is it possible to move from the platitude that ordinary truth predications are unqualified, to the substantive claim that truth itself is not subject to perspectival variation? There are two directions in which it seems initially possible to argue in favor of an alleged connection between the unqualified character of our ordinary truth predications and their correctness being absolute or non-perspectival:

1. The monadic character of our ordinary truth predications and the endorsement of (EQ) entail that truth cannot vary according to the perspective of assessment.

2. The dependence of truth upon an assessment perspective entails that a monadic truth predicate that obeys (EQ) cannot be applied to the domains of discourse that are apt for a truth-relativistic treatment.

And the landscape becomes complicated by the fact that there may be up to three notions of relative truth (or three predicates that, more or less directly, could be associated with the idea of truth relative to a perspective) which could be seen as being targeted by Horwich’s argument: a properly semantic predicate, ‘true at ⟨w, s⟩’ (where w is a possible world and s a perspective of assessment), a predicate with immediate pragmatic relevance, such as ‘true as used at c and as assessed from c’’ or ‘correct as assessed from c’’ (where c is a context of use and c’ a context of assessment, in MacFarlane (2014)’s terminology), and the predicate ‘true relative to a perspective’ as it might be used in an informal characterization of relativism as the view that there are domains of discourse in which propositional truth is relative to an assessment perspective.

In order the appreciate this last point, allow me to delve a bit into relativistic theories of meaning. In the presentation I favor (Caso, 2014), relativism about truth—or, more properly, radical relativism—is a semantico-pragmatic approach to natural languages according to which the correctness of assertive utterances of sentences belonging to certain areas of discourses is sensitive to the perspective from which they are assessed. The properly semantic part of radical relativism aims to define, in the style of Kaplan (1989), a predicate ‘true at ⟨w, s⟩’ for propositional contents, where ⟨w, s⟩ is a circumstance of evaluation composed of a possible world w and a further, assessment-oriented parameter s (a standard of taste, a system of moral norms, etc.). The pragmatic part contains certain principles linking truth at a circumstance for propositions with correctness for utterances. For example:

(PR) An utterance of a sentence S made at c is correct, as assessed from c’, if and only if the proposition expressed by S at c is true at ⟨wc, sc’⟩,

where wc is the world of c and sc’ is the parameter relevant at the context of assessment c’. Thus, (PR) is a semantics-pragmatics bridging principle linking the semantic-theoretic notion of truth at a circumstance of evaluation for propositions with a notion of immediate pragmatic relevance, correctness as assessed from a context c’ for assertive utterances (cfr. also Kölbel (2008a, 2008b, 2009) for a similar articulation of a moderate version of relativism).

Of course, other presentations are also possible. MacFarlane (2014), for example, favors a purely semantic presentation of relativism (which justifies calling it truth relativism) according to which the definition of the propositional predicate ‘true at ⟨w, s⟩’ is part of the semantics proper, while the characterization of a predicate of immediate pragmatic relevance is the job of what he calls post-semantics—a part of a theory of meaning that yields the predicate ‘true as used in c and as assessed from c’’ as the predicate with immediate pragmatic relevance. The interface with pragmatics comes by linking this predicate with speech acts such as assertion and retraction by means of principles like the following:

Reflexive Truth Rule. At a context c, assert that p only if p is true as used c and as assessed from c.

Retraction Rule. An agent in context c’ is required to retract an (unretracted) assertion of p made at c if p is not true as used at c and assessed from c’.

And again, we have the schema of a (properly) semantic-theoretic predicate ‘true at ⟨w, s⟩’ and a predicate with immediate pragmatic relevance, this time for propositions, ‘true as used at c and as assessed from c’’.3

Now, the wording of Horwich’s objection might suggest that what is being targeted is the predicate ‘true relative to a perspective’ that could be used in providing an informal characterization of relativism. However, Horwich’s insistence on his target being the notion of relative truth deployed in relativistic semantico-pragmatic theories of meaning suggests that the targeted notion could be the semantic-theoretic predicate ‘true at ⟨w, s⟩’, or the predicate with immediate pragmatic relevance, ‘true as used at c and as assessed from c’’, or ‘correct as assessed from c’’, as the case may be.

Hence, these will be the elements of the objection I will be concerned with: three candidates for being the target of the incompatibility claim, a piece of reasoning that seems to exhibit a gap between steps 3 and 4 of the reconstruction, and two ways in which we might attempt to overcome that gap.

4. Monadic truth and absolute truth

Let us concentrate on the first way in which we could attempt to bridge the gap between steps 3 and 4:

1. The monadic character of our ordinary truth predications and the endorsement of (EQ) entail that truth cannot vary according to the perspective of assessment.

I must confess that I cannot see how an argument supporting this first direction could be spelled out. On the contrary, there is a good explanation of how it is possible for truth to be perspectival, once the issue is recast in terms of the correct application of our ordinary truth predicate. The explanation runs along the following lines. The claim that our ordinary truth predicate cannot be applied to assessment-sensitive propositions is not part of relativism.4 What is part of relativism is the claim that the correctness of ordinary truth predications is sensitive to an assessment perspective, so that the same assessment-sensitive proposition may correctly be said to be true when assessed from a given perspective, and false when assessed from a sufficiently different one. That is, relativism does not preclude the application of our everyday truth predicate to assessment-sensitive propositions. Rather, it only precludes that its application be correct regardless of the perspective of assessment. Moreover, far from conflicting with (EQ), it is the adoption of this principle that explains how our ordinary truth predications can be assessment-sensitive: on the one hand, according to truth relativism, we have assessment-sensitive practices within certain domains of discourse, which determine assessment-relative truth conditions for the contents expressed by declarative sentences belonging to those domains; on the other hand, our ordinary truth predications inherit this relativity to an assessment perspective in virtue of the equivalence that holds between a claim to the effect that p and the claim to the effect that the proposition that p is true. Hence, (EQ) and the assessment-sensitive practices that (according to relativism) are characteristic of certain domains of discourse entail that our ordinary, monadic truth predications within those domains are as assessment sensitive as the corresponding ground-language claims.5

Also, note that this answer is independent of whether we take the target of Horwich’s argument to be the semantic-theoretic predicate ‘true at ⟨w, s⟩,’ the pragmatically relevant predicate ‘correct as assessed from c’’ (or ‘true as used in c and as assessed from c’’), or the informal predicate ‘true relative to a perspective’: all that matters is that relativism can secure the application of a monadic truth predicate that obeys (a suitably restricted version of) (EQ) to assessment-sensitive propositions.

5. Monadic truth and relative truth

Let us move on to the second direction in which we could attempt an argument in favor of the connection between the unqualified character of ordinary truth predications and their correctness being absolute, namely:

2. The dependence of truth upon an assessment perspective entails that a monadic truth predicate that obeys (EQ) cannot be applied to the domains of discourse that are apt for a truth-relativistic treatment.

We have already seen that a general case can be made in favor of a monadic, (EQ)-obeying truth predicate applicable to assessment-sensitive propositions. Now we will see that no matter how relative truth is understood (i.e., in terms of which predicate we understand the claim that, within certain domains of discourse, truth is relative to a perspective), relative truth does not preclude the application of a monadic, (EQ)-obeying truth predicate to assessment-sensitive propositions.

Let us start with the semantic-theoretic predicate ‘true at ⟨w, s⟩’. If by the claim that truth is relative to a perspective we mean that, in doing semantics, we employ the predicate ‘true at ⟨w, s⟩’ (hence rendering the claim that propositional truth is relative to a perspective as the claim that propositions are semantic objects whose extension is determined by reference to a circumstance of evaluation composed of a world and an assessment-oriented parameter), then the claim that relative truth entails that a monadic, (EQ)-obeying truth predicate cannot be applied to assessment-sensitive propositions risks to involve a somewhat imprecise description of the theoretical enterprise of relativism, as well as an under-appreciation of the distinction between levels of language with which relativistic semantico-pragmatic theories operate. For, from the point of view of the relativist’s semantic enterprise, the predicate ‘true at ⟨w, s⟩’ is a theoretical predicate introduced with the explicit aim of systematizing facts about the meaning of sentences of natural language, thus allowing for the recursive assignment of contents to natural language sentences in context. Hence, it is a metalinguistic predicate that the semantic part of relativism employs to specify facts about the meaning of the sentences belonging to the areas of discourse being studied. In contradistinction, the ordinary, monadic propositional truth predicate is a predicate that belongs to the language whose semantics is being studied, that is, it is an object-language predicate, and should not be assumed to share the language level of theoretical predicates such as ‘true at ⟨w, s⟩.’6 Thus, there appears to be no incompatibility between offering a relativistic semantic analysis of evaluative domains of discourse and holding that our ordinary notion of truth applies to the propositions expressed by declarative sentences belonging to those domains.7

So far, I have been focusing on the use the relativist makes of the theoretical predicate ‘true at ⟨w, s⟩’ in spelling out her relativistic analyses as the target of Horwich’s objection in its generalized form, but what about the natural language predicate ‘true relative to a perspective’ that we could use in an informal characterization of relativism? It seems to me that targeting this predicate does not ultimately help Horwich’s objection, although it requires a modification of the answer we give to it.

To see that the relativity of propositional truth, characterized by means of this informal predicate, does not preclude the application of our ordinary truth predicate to assessment-sensitive propositions, it suffices to notice that the deflationist herself has a right to the phrase ‘true relative to a perspective.’ And whatever the correct philosophical story about our ordinary truth predicate turns out to be, if the deflationist has a right to use the phrase, with even more reason will any non-deflationist who endorses (EQ) as characteristic of our everyday notion of truth.

Let us start by noticing that the claim that certain propositions are true (or false) relative to a perspective is a way of generalizing over more specific relativizations, such as a relativization to a standard of taste, to a system of moral norms, etc. Now, the deflationist already has a right to use certain explicitly relativized truth predicates, such as those that relativize truth to a possible world and to a time. Of course, whether an explicitly relativized truth predicate is admissible from a deflationary point of view ultimately depends on whether its characterization involves deflationarily unacceptable resources (cfr. Field (2001, 2009)), and the very idea of a deflationarily acceptable resource is itself somewhat vague. However, the following biconditionals seem to be introductions of deflationarily acceptable, explicitly relativized truth predicates:

(1) The proposition that dinosaurs are extinct is true at a world w if and only if dinosaurs are extinct at w.

(2) The proposition that Socrates is sitting is true at a time t if and only if Socrates is sitting at t.

These biconditionals may be seen as introductions of the explicitly relativized truth predicates ‘true at w’ and ‘true at t,’ respectively. And these predicates should be acceptable to a deflationist, since they amount to grammatical transformations warranted by how our everyday truth predicate behaves when ascending from explicitly relativized ground language claims.8 But, if this is so, then I see no reason not to accept predicates such as ‘true relative to a standard of taste’ and ‘true relative to a system of moral norms’ (hence, the predicate ‘true relative to a perspective’) as deflationarily acceptable, for the biconditionals:

(3) The proposition that apples are tasty is true relative to a standard of taste S if and only if apples are tasty relative to S.

(4) The proposition that stealing is wrong is true relative to a system of moral norms n if and only if stealing is wrong according to n.

should be as admissible as (1) and (2), since the resources employed in introducing the new predicates are not fundamentally different from the resources employed before. Insofar as the idea of apples being tasty relative to a standard of taste, and of stealing being wrong according to a system of moral norms, are intelligible to the deflationist (and they seem to be), the new predicates should be deflationarily acceptable. With even more reason they should be acceptable to anyone with non-deflationary tendencies.

Thus, we have considered the semantic-theoretic predicate ‘true at ⟨w, s⟩’ and the informal predicate ‘true relative to a perspective,’ and none provided warrant from the inferential move from step 3 to step 4 in Horwich’s argument. The only predicates that are left are the predicates with immediate pragmatic relevance, ‘correct as assessed from c’’ and ‘true as used at c and as assessed from c’,’ depending on the presentation chosen for the relativistic theory of meaning. Are these predicates incompatible with the application of a monadic, (EQ)-obeying truth predicate to the propositions expressed by sentences belonging to assessment-sensitive discourses? The short answer seems to be: no. The reason depends on how we conceptualize these predicates. If we take them to be philosophical terms of art, the considerations that applied to the semantic-theoretic predicate ‘true at ⟨w, s⟩’ seem to apply to these predicates as well: they would be predicates introduced for the purpose of semantico-pragmatic theorizing, and should not be considered to share the same level of language with the object-language truth predicate. Rather than conflicting with it, they help in spelling out in which sense the correctness of our ordinary truth predications is itself perspectival. If, on the other hand, we take them to be predicates drawn from an informal way of understanding the relativity of propositional truth or of utterance correctness, then they may be thought of as sharing the same language level with our everyday truth predicate. However, in that case, the situation seems to be parallel to that of the explicitly relativized truth predicates ‘true according to a standard of taste,’ ‘true relative to a system of moral norms,’ and more generally ‘true relative to a perspective’: they can be obtained by means of resources that ought to be deflationarily acceptable. All that we need at this point is that the practice of assessing the correctness of an utterance (or the truth of a proposition) from a context different from the context of utterance (or of use of that proposition) be intelligible to a deflationist—and I see no reason to expect it not to be. And if this practice is intelligible, then the idea that an utterance may be deemed correct or incorrect in an act of assessment (hence, that an utterance may be correct or incorrect as assessed from the context in which the assessment takes place) should also be intelligible (and the same goes for propositional truth).

Moreover, there are interesting relations between these explicitly relational notions of correctness and truth and the ordinary concept of truth that may be of help in understanding them better. For one of the key elements of relativism is the claim that an assertion is correct as assessed from a context c’ just in case its content is true relative to (the world of the context of utterance and) the assessment-oriented parameter relevant at c’. This is indeed the import of a principle in the line of (PR), once it is applied to the special case where the utterance in question is assertive. And the considerations in Evans (1985) have taught us that principles of assessment in the line of (PR) are nothing more than the explicit relativization of a principle linking correctness simpliciter with truth simpliciter, which could be expressed as follows (leaving aside indexical sentences):

(C) An utterance of S is correct if and only if S is true.

Once this link is pointed out, (PR) may be seen merely as making explicit the principle of assessment expressed by (C).

Of course, there are interesting (and difficult) issues regarding the relativistic notions of immediate pragmatic relevance: once speech acts such as assertion are understood in terms of these notions, it is not clear whether we may still think of them as aiming to, or as being (constitutively) normed by, truth or correctness (Evans (1985), MacFarlane (2005), García-Carpintero (2008), Field (2009)), so that it is not clear whether we can still make sense of assertion as a norm-governed practice (Caso (2014), Greenough (2011), MacFarlane (2014)) or we have to make sense of assertion under a different way of conceptualizing it—e.g., in terms of commitments rather than aims (MacFarlane (2005)). More recently, Gariazzo (2016, 2019) has claimed that assessment-sensitivity can be made sense of only piecemeal, taking one area of discourse at a time, and that usual ways of doing so are not successful. However, even if thorny, these issues are orthogonal to the issue of the compatibility of truth relativism with the claim that our ordinary truth predicate obeys one (suitably restricted) version of (EQ) or other.

6. Last line of resistance

In the last section, we claimed that the deflationist has a right to use the phrase ‘true relative to a perspective,’ and that this is enough to warrant the relativist’s use of the concept of relative truth. A last line of resistance that the deflationist might essay against the idea of relative truth could consist in holding that, even though she has a right to the phrases that the relativist employs in her characterization of truth relativism, the philosophical content that the relativist attributes to those phrases is off-limits for the deflationist. Of course, this worry goes beyond the problem of assessing the compatibility of the notion of relative truth with the claim that our ordinary truth predicate obeys one (suitably restricted) version of (EQ) or other. However, examining this worry will allow us to have a firmer grip on what is the minimal philosophical content required by the idea of perspectival truth, and it will allow us to ascertain to what extent truth relativism is compatible with a properly deflationary, or at least minimalist (in the sense of Wright (1992)’s and Kölbel (2002)’s metaphysically lightweight) conception of truth.

Surprisingly, it is not that clear that there is a substantive philosophical content behind the idea that propositional truth is relative to a perspective—or, more precisely, whatever substantive content there is in the idea of relative truth does not seem to be a content that has to do specifically with the notion of truth. For, as we have already remarked, the claim that propositional truth is relative to an assessment perspective may be seen as a way of performing two operations of generalization at once. First, there is the generalization into the assessment-oriented parameter position, since ‘perspective’ may be taken to be a blanket term for more specific relativizations of propositional truth, such as a relativization to a standard of taste, to a system of moral norms, etc. Thus, the claim that propositional truth is relative to a perspective is simply a way of compacting into a single claim the more specific claims that propositional truth, in matters of taste, is relative to a standard of taste, that propositional truth, in moral matters, is relative to a system of moral norms, etc. And these relativizations, in turn, reduce, in the presence of (EQ), to claims that can be cast purely in terms of the ground language (with the well-known loss of expressive power). And this is the second operation of generalization, this time, into sentential (or propositional) position: the claim that the proposition that apples are tasty is true relative to standard of taste s1 and false relative to standard of taste s2 simply reduces, or is equivalent, to the claim that apples are tasty according to s1, and not tasty according to s2 (mutatis mutandis, the same applies to sentences and propositions belonging to any discourse for which a distinctive form of assessment-oriented relativization is being proposed). Thus, the formulation in terms of propositional truth being relative to an assessment perspective is a convenient way of compacting into a single claim the numerous claims that could otherwise be cast in purely non-semantic vocabulary.9 That is, the minimal philosophical content involved in the idea that propositional truth is relative to a perspective comes down to a ground-language relativism: the deployment of a formulation in terms of truth, rather than the use of a schematic formulation or the endorsement of a plurality of formulations cast exclusively in terms of the ground language, has only an expressive function that ought to be acceptable to the deflationist.

7. Conclusion

The idea that, within certain domains of discourse, truth is relative to a perspective does not conflict with the claim that our ordinary truth predicate—which obeys (a suitably restricted version of) an equivalence principle to the effect that a claim that p is equivalent to the claim that the proposition that p is true—applies within those discourses. Regardless of whether the style of semantico-pragmatic theorizing deployed by relativists is ultimately acceptable to someone with deflationary tendencies, the most worrisome aspects of Horwich’s objection (that the philosophical insights of relativism require the revision of our practice with our everyday truth predicate and that the style of semantico-pragmatic theorizing chosen by relativists precludes the existence of truth simpliciter within assessment-sensitive discourses) seem to pose no real threat to relativism: the different notions of relative truth deployed in relativistic theories of meaning are compatible with our ordinary notion of truth, and the endorsement of a relativistic conception of certain areas of discourse does not force upon us the rejection of our ordinary notion of truth as inapplicable within those areas. As for the question, to what extent is the idea of relative truth compatible with deflationism, it seems that the informal insight that there is no absolute truth and falsity within evaluative domains of discourse survives a deflationary understanding of our ordinary notion of truth.

References

Caso, R. (2014), «Assertion and relative truth», Synthese, 191(6), pp. 1309–1325.

Caso, R. (2016), «A second opinion on relative truth», Manuscrito, 38(2), 65–88.

Evans, G. (1985), «Does tense logic rest upon a mistake?», in: Evans, G.: Collected papers, Oxford: Clarendon Press, pp. 343–363.

Field, H. (2001), «Disquotational truth and factually defective discourse», in: Field, H.: Truth and the absence of fact, Oxford: Oxford University Press, pp. 222–258.

Field, H. (2009), «Epistemology without metaphysics», Philosophical Studies, 143(2), pp. 249–290.

García-Carpintero, M. (2008), «Relativism, vagueness, and what is said», in: García-Carpintero, M. and Kölbel, M. (eds.): Relative truth, Oxford: Oxford University Press, pp. 129–154.

Gariazzo, M. (2016), «Can we make sense of MacFarlane’s relative truth?», Manuscrito, 40(2), 39–70.

Gariazzo, M. (2019), «Assertion and assessment sensitivity», Kriterion, 143, 355–376.

Greenough, P. (2011), «Truth-relativism, norm-relativism, and assertion», in: Brown, J. and Cappelen, H. (eds): Assertion, Oxford: Oxford University Press, pp. 197–231.

Horwich, P. (2014), «An undermining diagnosis of relativism about truth», Mind, 123(491), pp. 733–752.

Kaplan, D. (1989), «Demonstratives: an essay on the semantics, logic, metaphysics, and epistemology of demonstratives and other indexicals», in: Almog, J., Perry, J., and Wettstein, H. (eds): Themes from Kaplan, Oxford: Oxford University Press, pp. 481–563.

Kölbel, M. (2002), Truth without objectivity, London: Routledge.

Kölbel, M. (2008a), «Truth in semantics», Midwest Studies in Philosophy, 32, pp. 242–257.

Kölbel, M. (2008b), «Motivations for relativism», in: García-Carpintero, M. and Kölbel, M. (eds.): Relative truth, Oxford: Oxford University Press, pp. 1–38.

Kölbel, M. (2008c), «‘True’ as ambiguous», Philosophy and Phenomenological Research, 77(1), pp. 359–384.

Kölbel, M. (2009), «The evidence for relativism», Synthese, 166(2), pp. 375–395.

Lewis, D. (1980), «Index, context, and content», in: Kanger, S. and Öhman, S. (eds): Philosophy and grammar, Dordrecht: Reidel, pp. 79–100.

MacFarlane, J. (2005), «Making sense of relative truth», Proceedings of the Aristotelian Society, 105, pp. 321–339.

MacFarlane, J. (2014), Assessment sensitivity: Relative truth and its applications, Oxford: Clarendon Press.

Williams, M. (2005), «Meaning and deflationary truth», in: Armour-Garb, B. P. and Beall, J. C. (eds): Deflationary truth, Chicago: Open Court, pp. 301–320.

Wright, C. (1992), Truth and objectivity, Cambridge MA: Harvard University Press.

Wright, C. (2008), «Relativism about truth itself: Haphazard thoughts about the very idea», in: García-Carpintero, M. and Kölbel, M. (eds.): Relative truth, Oxford: Oxford University Press, pp. 157–185.

---