JAMES A. WOODBRIDGE
I maintain that considerable philosophical progress can be made on the topic of truth if we recognize that truth is a pretense. The claim that truth is a pretense may strike many as preposterous, but a few initial remarks might deflect some of this incredulity. To start, my position is not a radical relativism according to which being true is just a matter of being pretended true. Pretending it is true that the Mercedes parked across the street is yours does not make it true, and the police will be happy to let you know this if you try to take it for a drive. Pretending something is true involves applying an additional layer of pretense to what is expressed by some truth-attribution via the pretense that truth-talk already always involves in its basic linguistic functioning. This remark leads to my second point, namely, that the idea of truth being a pretense is really part of an analysis of truth-talk, the fragment of our talk (and thought) that employs the notions of truth, falsity, etc. My view is thus a kind of fictionalism regarding truth-talk.
This might raise some eyebrows since fictionalist analyses are commonly held to generate error theories. The problem with an error-theoretic reading of truth-talk is by now well known: an account of truth-talk based on the thesis that all its instances are false (or, more broadly, not true) presupposes an antecedent notion of truth-conditions, and so of truth. Moreover, the claim that all instances of truth-talk are false is itself an instance of truth-talk, and so would have to be false on this view. In fact, the problem is even worse; the error-theoretic thesis would be paradoxical, since it would say of itself that it was false. Understood this way, a fictionalist interpretation of truth-talk is a non-starter.
The fictionalist account I offer here avoids this incoherence by explaining truth-talk in terms of semantic pretense. This view is novel in at least two ways. First, most likely because a fictionalist account of truth-talk goes against conventional wisdom, the pretense approach has not been applied to truth-talk before. I claim that this approach can be used to construct a coherent fictionalist account of truth-talk because utterances understood in terms of pretense can still be used to make serious claims about the real world indirectly. Second, philosophers who apply the pretense approach elsewhere typically use truth-talk in such a way when doing so, that truth-talk might seem to be one place this approach cannot be applied. As I hope to make clear, however, the application to truth-talk is not only possible, but extremely illuminating.
The resulting account of truth-talk amounts to a new formulation of deflationism: although there are no such properties, making as if to describe things as having or lacking properties called "truth" and "falsity" extends the expressive capacity of our language by providing a way for us to make certain complicated, non-semantic, general claims we could not otherwise make. Through a pretense of property attribution, truth-talk allows us to generalize on embedded sentence positions, as in the move from
(1) If Bob says that crabapples are edible, then crabapples are edible.
(2) Everything Bob says is true.
What we are after in generalizing from (1) is a claim covering everything Bob might say, something to the effect of
(3) For all p, if Bob says that p, then p.
However, because (3) involves variables for sentences in use, we do not have the right sorts of logical devices in natural language to express this. If we try to use our ordinary devices of quantification and object variables (pronouns) we end up with the ungrammatical
(4) Everything is such that if Bob says that it, then it.
Of course, this is where truth-talk comes in. By adding the truth-predicate to the object-variables (pronouns) in (4) we get the grammatical
(5) Everything is such that if Bob says that it is true, then it is true.
The instances of (5) (in which the 'it' is replaced with a that-clause) then combine with the fundamental principles governing truth-talk, the instances of the equivalence schema
(ES) It is true that p iff p (= That p is true iff p).
In this way, truth-talk provides a natural language surrogate for the sort of generalization formally characterized in (3).
The pretense-theoretic formulation has certain advantages over other formulations of deflationism. In particular, it avoids the serious problems other deflationary views have in accounting for the generalizing role just mentioned, the very one they emphasize as truth-talk's central function. I'll return to this point later.
The idea of semantic pretense comes from Kendall Walton's work as presented in his book Mimesis as Make-Believe, but it has also been fruitfully extended to issues outside of aesthetics. The most familiar of these applications is that made to our talk ostensibly of what does and does not exist (henceforth, existence-talk). A pretense-based analysis of existence-talk is motivated primarily by the puzzle of negative existentials, such as
(6) Santa Claus does not exist.
According to the pretense approach, claims like (6) can be genuinely true, but the explanation of how this is possible avoids prima facie paradoxical ontological commitments to nonexistent entities. Moreover, even though (6) functions in virtue of a pretense, it can still be used to make a serious assertion about the world because of the special kind of pretense involved.
The relevant kind of pretense is most familiar from children's games of make-believe. The interesting aspect of make-believe is that some of what is to be pretended by participants in the game--some of what is fictionally true or fictional in the game--depends on the state of the world outside of the game. Games of make-believe involve principles of generation, rules that determine the way actual circumstances combine with the game's stipulated pretenses to determine what else is to be pretended (which pretenses are prescribed). Within the context of a game of make-believe, then, there are two kinds of prescribed pretenses: those that are the stipulative ground of the game--what is expressly made-believe--and those that are generated from reality.
As an illustration of how this distinction functions in semantic pretense, consider the children's game of mudpies. In this game of make-believe, certain pretenses are stipulated: patty-shaped globs of mud are pretended to be pies, the hollow stump is pretended to be an oven, etc. Following this, certain other pretenses are prescribed depending on what happens in the world outside of the game. If someone puts a patty-shaped glob of mud into the hollow stump, it is to be pretended that she has put a pie in the oven. By including pretenses generated from reality, a game of make-believe establishes a systematic dependency between some of what is to be pretended and real-world conditions obtaining outside of the game.
As a result, a game of make-believe can provide a mechanism through which a speaker can, by making as if to say one thing, succeed in making quite a different, yet serious, assertion about the world. For example, if, in talking about two children playing the mudpies game, I say
(7) Dex stole one of Corey's pies out of the oven.
my utterance employs pretenses from the make-believe, but there is still a sense in which I make a serious assertion--just not one about pies or an oven. Sincerely uttering (7) as part of the mudpies game offers the pretenses displayed in the utterance as prescribed or appropriate. According to the rules of the game, they are if and only if certain real-world conditions obtain, namely those specified in
(8) Dex took a glob of mud that Corey had put in the hollow stump as part of a game out of the stump against her wishes.
An utterance of (7), therefore, expresses a commitment to the obtaining of these real-world conditions. In this way, (7) amounts to what I call a "partly pretend" claim.
Understood as a pretense-employing claim, an utterance of (7) makes a genuinely true, serious assertion whenever the pretenses displayed in the utterance are prescribed, that is, whenever an utterance of (8) makes a true assertion. Thus, a pretense-employing way of talking can serve as a way of making serious assertions indirectly, that is, as a way of engaging in "indirectly serious discourse." Even though (7) involves pretense, I can utter it for serious purposes, for instance, in order to explain why Corey is crying. So, an appeal to make-believe allows for, rather than undermines, the serious purposes served by a "way of talking." And if a way of talking is problematic when taken at face value, an appeal to pretense might explain how it serves any serious purposes at all. Certain philosophical problems might thus be solved by recognizing make-believe at work in ways of talking where it has not been noticed before.
Consider existence-talk again. Puzzles about negative existentials can be resolved by understanding existence-talk in terms of a game of make-believe governed by rules summed up on the handout. These stipulate pretending that every putative referring expression has a bearer, and that 'exists' is used to attribute a discriminating property. The serious purposes of existence-talk are explained by the principle of generation making it to be pretended that a (pretend) referent has the (pretend) property of existence iff the referring expression, as employed, really refers to something. Because of the dependency established, an utterance like (6) can be used to make a serious and genuinely true claim about how the world actually is (namely, that the kind of attempt to refer displayed in the utterance is always unsuccessful). (6) makes such a claim, even though it employs pretense to do so. An additional aspect of existence-talk is that it involves pretense intrinsically; pretense is essential to the very functioning of the expression 'exists'. In (7), the standard, pretense-free uses of the sentence and its components are exploited in the pretense; the pretense is extrinsic to the utterance, and the utterance is literally false. But there is no standard, pretense-free use of 'exists' being exploited by the pretense-involving use. The latter is the standard use. Because existence-talk involves pretense intrinsically it makes no sense to say even that existence-talk is always literally false. It cannot be taken literally.
Similar points hold for truth-talk, and this is how a pretense-based account avoids being an error-theory. Truth-talk can be understood as functioning in virtue of a game of make-believe governed by rules like the following.
I. It is to be pretended that expressions like 'is true' and 'is false' function predicatively to describe objects as having or lacking certain properties (called "truth" and "falsity").
II. The pretenses displayed in an utterance of '(The proposition) that p is true' are prescribed if and only if p.
III. The pretenses displayed in an utterance of '(The proposition) that p is false' are prescribed if and only if ~p.
Rules II and III are schematic principles of generation, each providing every possible instance of the given schema. These rules determine the real-world conditions under which particular truth-attributions make genuinely true claims indirectly. So, truth-talk is not automatically false. Rule I shows that truth-truth, like existence-talk, involves pretense intrinsically. Pretense is invoked in the basic functioning of the truth-locutions, so there is no pretense-free use of them. As a result, truth-talk cannot be taken literally and thus is not even automatically literally false. A pretense-based account of truth-talk is therefore no error theory. This placates the initial worry that a pretense-based account of truth-talk is incoherent.
Here is how the make-believe works in a basic instance of truth-talk, such as
(9) It is true that crabapples are edible.
This utterance belongs to the game of make-believe governed by the rules just sketched. An utterance of (9) invokes the pretenses that a certain object is picked out by '(the proposition) that crabapples are edible' and that this object (made the referent of the pronoun 'it' by the usual sort of anaphoric link) has the property attributed by the expression 'is true'. According to Rule II, these pretenses are prescribed if and only if crabapples are edible. Thus, (9) is a partly pretend claim; the serious claim it makes indirectly is
(10) Crabapples are edible.
More generally, the game of make-believe behind truth-talk generates all the instances of the equivalence schema
(ES) It is true that p iff p (= That p is true iff p).
This is an important result because these equivalences are the fundamental principles governing truth-talk--they are all necessary and a priori. My account gives them the right status since they follow directly just from the functioning truth-talk is given by the game of make-believe behind it.
I now turn to some advantages of my view. Because I take truth-talk to involve a pretense of property attribution, my view is logico-syntactically conservative. The logical form of any instance of truth-talk is just what it appears to be. This gives my view an advantage in accounting for the more complex forms of truth-talk in which it serves its real purpose: quantificational instances like
(2) Everything Bob says is true.
On my view, the logical form of claims like (2) involves objectual quantification (most likely restricted to propositions) and full-blown predication in which the objects from the domain satisfying a particular condition (here, being said by Bob) are described as possessing a particular property (truth). This all goes on in the context of a pretense, of course, since on my view there is no property of truth (and no propositions, for that matter). But the use of standard logical and linguistic devices in contexts of pretense is already familiar from claims like
(11) All of Harry Potter's relatives are mean.
In addressing quantificational instances of truth-talk like (2), the current formulations of deflationism offered by Paul Horwich, Robert Brandom, and Hartry Field all face serious problems. Anil Gupta has explained how Horwich's view cannot make sense of any generalizations involving the notion of truth since the Minimal Theory is just the collection of the instances of the equivalence schema, and a collection of instances does not amount to a generalization. A similar problem arises if one interprets the main quantifier in a claim like (2) substitutionally rather than objectually. (Disquotationalists like Field might be tempted to do this, and prosententialists like Brandom must do so because of that approach's reliance on anaphoric relations). Deflationists understand the universal substitutional quantifier as a means of encoding a (potentially infinite) conjunction of the instances of the schema the quantifier precedes. But a generalization is more than a conjunction even of all its instances.
Field's official strategy actually avoids this problem and takes the main quantifier in a generalizing instance of truth-talk to be objectual (perhaps restricted to sentence-tokens). However, in addition to problems generated by taking the things that people assert, believe, etc. to be sentences, Field still has non-standard logical devices like substitutional quantification and/or schematic sentence variables at work in the underlying logic of generalized truth claims. This leaves us with the question of why this underlying logic comes to us in such a misleading surface form, making it simply a brute fact that it does. My view, on the other hand, explains why truth-talk comes in the form it does, and also how these apparently unsuitable claims, employing nothing but familiar logical and linguistic devices like objectual quantification and predication, manage to incorporate into our language a surrogate for complex, non-standard logical devices like substitutional quantifiers and sentential variables.
The role pretense plays in my view also gives it an advantage in dealing with the Liar Paradox. The apparent possibility of generating a strengthened liar sentence in response to any proposed elimination of the paradox strongly suggests that inconsistency cannot be eliminated from truth-talk without imposing severe artificial constraints on it. This is one of my main reasons for denying there is a property of truth attributed by truth-talk. A merit of the pretense-based view is that in rejecting any such property, it does not need to eliminate truth-talk's inconsistency. Because I take truth-talk to involve pretense intrinsically, a putatively paradoxical claim, such as
(L) The proposition expressed by the sentence labeled '(L)' is not true
turns out to be what I call a "purely pretend" claim--in contrast with unproblematic instances of truth-talk, such as (9), which amount to partly pretend claims. As an intrinsically pretense-invoking claim, (L) necessarily looks to some "other" claim as the source of any serious content it puts forward. But the self-referential aspect of (L) makes (L) itself the "other" claim it looks to for its serious content. But then no serious content ever attaches to (L). This situation is my view's analog of the notion of ungroundedness. Since truth-talk's problematic inconsistency only arises in purely pretend claims, it is entirely contained within the pretense; there is no serious inconsistency. The reason the contained inconsistency does not then trivialize the pretense the underwrites truth-talk is that, since a pretense is a kind of fiction, its logic is paraconsistent--the notion of entailment involved is non-explosive. Thus, the inconsistency (contradiction) manifested in (L) and its ilk does not entail everything.
The paraconsistent nature of the logic of fiction is already familiar from cases like the Holmes stories in which Watson's sole war wound is in his leg and not in his leg (if we take the stories all together). Although the stories contain this inconsistency we are not thereby invited to conclude that in the stories everything is true. Explosion is blocked. Similarly, since truth-talk is based on pretense, the ineliminable inconsistency generated by its fundamental governing principles (the instances of (ES)) is also non-explosive; the logic of truth-talk is paraconsistent.
I have time to consider only one objection, so I will address the most serious challenge facing my view. Although I have already blocked the error-theoretic circularity worry, circularity threatens again at a deeper level. The charge is that truth cannot be explained in terms of pretense because pretense is explained in terms of truth. A natural seeming account of pretending is as follows: pretending that a is F involves regarding the proposition that a is F as being true, whether or not it is true. Since I want to explain truth (and propositions) in terms of pretending, this account is obviously unavailable to me.
My response is to offer the following non-truth-theoretic account of pretending. Start by semantically descending from the above account and explaining pretending that a is F in terms of regarding a as being F, whether or not it is F. For this to cover cases in which 'a' is taken as an empty or fictional name it has to be modified as follows: pretending that a is F involves regarding the serious subject of the sentence 'a is F' as being as it is seriously described in that utterance, whether it is that way or not. This account applies generally, both to situations in which pretending that a is F involves applying extrinsic pretense to something expressed directly by a claim that does not involve any pretense, as well as where it involves applying extrinsic pretense to something expressed only indirectly by a claim that itself invokes pretense intrinsically (e.g., cases involving the use of name as empty or fictional).
The pretense approach can thus be used to construct a novel and interesting account of truth-talk. This account offers certain merits over other formulations of deflationism, and thereby allows us to take advantage of the benefits deflationism offers (e.g., not having to eliminate truth-talk's inconsistency) without accruing the costs imposed by the current formulations of deflationism.
 See Boghossian (1990), pp. 167, 174-175 for a fuller account of this kind of objection to such a view.
 Consider, for example, the use of 'fictional' or 'fictionally true' in the explanations of the pretense approach offered in Walton (1990), Chapter 11 and Crimmins (1998), as well as the pretense-based accounts of existence-talk they discuss. Truth-talk plays a similar role in the applications of the pretense-approach made to attitude ascriptions and identity-talk in Crimmins (1998), and the application made to possible-worlds-talk in Yablo (1996).
 Chapter 11 of Walton (1990) lays out the sort of pretense-theoretic account of existence-talk sketched here. Slightly different accounts are developed in Evans (1982), Chapter 10 and Kroon (1996).
 A way of talking is a loosely bounded fragment of discourse (and thought) centered around some expression (concept) or family of expressions (concepts)--e.g., modality, numbers, truth--or around some mode or figure of speech--e.g., metaphor, irony, hyperbole.
 Velleman (2000), p. 251.
Alston, William P. (1996) A Realist Conception of Truth.
Boghossian, Paul. (1990) "The Status of Content." Philosophical Review 99: 157-184.
Brandom, Robert. (1994) Making It Explicit.
Chihara, Charles. (1979) "The Semantic Paradoxes: A Diagnostic Investigation." Philosophical Review 88: 590-618.
Crimmins, Mark. (1998) "Hesperus and Phosphorus: Sense, Pretense, and Reference." Philosophical Review 107: 1-47.
Evans, Gareth. (1982) The Varieties of Reference, John McDowell (ed.).
Field, Hartry. (1986) "The
Deflationary Conception of Truth." In G. McDonald and C. Wright (eds.) Fact, Science and Morality,
Field, Hartry. (1989) Realism, Mathematics and Modality.
Field, Hartry. (1994) "Deflationist Views of Meaning and Content." Mind 103: 249-285.
Field, Hartry. (2001) Truth and the Absence of Fact,
Grover, Dororthy. (1977) "Inheritors and Paradox." Journal of Philosophy 74: 590-604.
Grover, Dorothy L., Camp, Joseph L., Jr. and Belnap, Nuel D., Jr. (1975) "A Prosentential Theory of Truth." Philosophical Studies 27: 73-125.
Gupta, Anil. (1993) "A Critique of Deflationism." Philosophical Topics 21: 57-81.
Horwich, Paul. (1998) Truth, Second Edition.
Kripke, Saul. (1975) "Outline of a Theory of Truth." Journal of Philosophy 72: 690-716.
Kroon, Frederick. (1996) "Characterizing Non-Existents." Grazer Philosophische Studien 51: 163-193.
Priest, Graham. (1979) "The Logic of Paradox." Journal of Philosophical Logic 8: 219-241.
Priest, Graham. (1998) "What Is So Bad about Contradictions?" Journal of Philosophy 95: 410-426.
Schiffer, Stephen. (1996) "Language-Created, Language-Independent Entities." Philosophical Topics 24: 149-167.
Soames, Scott. (1999) Understanding Truth.
Tarski, Alfred. (1944) "The Semantic Conception of Truth." Philosophy and Phenomenological Research 4: 341-375.
Velleman, J. David. (2000) "On the Aim of Belief." In The Possibility of Practical Reason,
L. (1990) Mimesis as Make-Believe.
Walton, Kendall L. (1993) "Metaphor and Prop-Oriented Make-Believe." European Journal of Philosophy 1: 39-56.
Yablo, Stephen. (1996) "How in the World?" Philosophical Topics 24: 255-286.
Yablo, Stephen. (1998) "Does Ontology Rest on a Mistake?" Proceedings of the Aristotelian Society, supp. vol. 72: 229-261.
Yablo, Stephen. (2000) "Aprioricity and Existence." In P. Boghossian and C. Peacocke (eds.) New Essays on the A Priori,