You are on page 1of 5

bs_bs_banner

2013 The Author Metaphilosophy 2013 Metaphilosophy LLC and Blackwell Publishing Ltd Published by Blackwell Publishing Ltd, 9600 Garsington Road, Oxford OX4 2DQ, UK, and 350 Main Street, Malden, MA 02148, USA METAPHILOSOPHY Vol. 44, Nos. 12, January 2013 0026-1068

A MINIMALIST THEORY OF TRUTH BRADLEY ARMOUR-GARB

Abstract: This article, after briey discussing Alfred Tarskis inuential theory of truth, turns to a more recent theory of truth, a deationary, or minimalist, theory. One of the chief elements of a deationary, or minimalist, theory of truth is that it replaces the question of what truth is with the question of what true does. After setting out the central features of the minimalist theory of truth, the article explains the motivation for opting for such a position. In addition, it provides some reasons for thinking that such a theory of truth is minimal or deationary in the way that contemporary truth theorists have claimed it to be. Keywords: deationism, meaning, minimalism, proposition, reference, Alfred Tarski, truth.

Introduction It is difcult to write about the development of the philosophy of language without returning to the revolution in logic and philosophy of language that occurred in the last century, beginning with the work of Gottlob Frege. Frege set out to show that the truths of arithmetic were analytic in nature, by deriving them from the axioms and denitions of logic. In order to carry out this logicist project, Frege needed to show that the theorems of arithmetic could be derived from the theorems of logic without appeal to any synthetic step. To show that his deductions achieved this goal, Frege devised a formal language that enabled him to carry out proofs. The formal language allowed for the characterization of a set of syntactic transformations, each of which was an instance of a purely logical inference rule. Freges concern was to place mathematics on the secure foundations of logic, and he appealed to semantics chiey in the service of this project. By contrast, the Polish logician Alfred Tarski set himself the task of setting semantics on the secure foundations of mathematics, by providing mathematical denitions of semantical concepts such as truth and logical consequence. In The Concept of Truth in Formalized Languages (Tarski 1983), he set out to show that, for many languages, one could consistently dene a truth predicate for that language. Without going into formal details, it is sufcient to note that Tarski set out a condition of adequacy
2013 The Author Metaphilosophy 2013 Metaphilosophy LLC and Blackwell Publishing Ltd

54

BRADLEY ARMOUR-GARB

for a denition of truth for a language. A denition of truth for a language was said to be materially adequate if, and only if, the denition has, as consequences, all instances of the following schema, where S is replaced by structural-descriptive names of sentences of the language, L, and p is replaced by translations of the sentence named in a different language, ML: (T) S is true if and only if p. Thus, Tarski shows how to dene the relative concept of Truth-in-L, for a specic language L. But there was a possible problem with Tarskis denition. He successfully showed that one could provide a denition of truth for a language, by appeal to the notions of reference and predicate satisfaction. But how are those notions to be understood? In particular, how are we to understand the reference relation, the word-world relation that stands between a name and its bearer? Although there was much work done on the notion of reference in the 1970s, in 1990 Paul Horwich made a surprising suggestion about how we might go about providing an account of truth. He agreed with Tarski that a denition of truth would be adequate only if it issued in all instances of (T). But rather than taking the instances of (T) to be consequences of a formal theory of truth, he proposed that we take those instances to exhaust our theory of truth. In so doing, he provided a deationary, sometimes called a minimalist, theory of truth. The theory is deated because it effectively denies that the truth predicate admits of a standard, or substantive, denition. In what follows, I describe the Minimalist Conception of Truth (henceforth, Minimalism) and briey set out its motivation.

Minimalism Minimalism comprises a theory of truth, the minimalist theory of truth, whose axioms are certain of the instances of the equivalence schema, (ES) <p> is true iff p. In (ES), surrounding an expression, p, by angled brackets produces an expression that refers to the propositional constituent expressed by what p says. So, <snow is white> (within angled brackets) is to be read as the proposition that snow is white. Note, also, that if and only if in (ES) is to be read as the biconditional, so that (ES) can be read as saying that if the proposition that p is true then p and if p then the proposition that p is true. Finally, just as the minimalist takes truth to be captured by the
2013 The Author Metaphilosophy 2013 Metaphilosophy LLC and Blackwell Publishing Ltd

A MINIMALIST THEORY OF TRUTH

55

instances of (ES), the minimalist takes falsity to be captured by certain instances of the falsity schema, (FS) <p> is false iff ~p.1 The minimalist theory regards the relevant axiomsthe instances of (ES)as both epistemologically and explanatorily fundamental. The instances of (ES) are epistemologically fundamental in that we do not arrive at them, or seek to justify them, on the basis of anything more obvious or immediately known. Instead, its claimed that, as linguistically competent language users, we are disposed to accept pretty much any instance of the schema. The instances of (ES) are taken to be explanatorily fundamental in two senses. First, it is claimed that our underived acceptance of them is the source of everything else we do with the truth predicate. Thus, the minimalist is committed to explaining our acceptance of sentences containing true by reference to our underived inclination to accept the instances of (ES), together, perhaps, with our acceptance of propositions not involving the concept of truth. Second, its claimed that minimalism sufces to explain all of the facts about truth. The claim of explanatory fundamentality, when combined with a particular view of meaning, yields the Minimalist Thesis of Truth. The minimalist thesis about truth is that the meaning of true is xed by our disposition, as linguistically competent language users, to accept instances of (ES). So, the Minimalist Thesis about truth xes the meaning of truth: it is xed by our inclination to accept instances of (ES). And the Minimalist Conception of Truth tells us that the instances of (ES) tell us everything there is to say about the meaning of the word true and the role and function of the truth predicate in a language like English. But if the instances of (ES) tell us everything there is to say about truth, why have the truth predicate in a natural language in the rst place? What role, if any, does it perform? According to the minimalist, the truth predicate enables us to express certain claims that we could not otherwise express, and therein lies the utility of truth. An example will help. Suppose that you accept everything that Barack Obama thinks and says, and that you want to express your commitment to all of the propositions that he thinks and says, so that someone else might come to accept your thought and speech, too. How would you communicate this to this other person? Well, the rst thought is that you could say, I am committed to all of the propositions that Obama thinks and says. But thats not particularly helpful, for the other
1 This is to be read as: The proposition that p is false if, and only if, it is not the case that p.

2013 The Author Metaphilosophy 2013 Metaphilosophy LLC and Blackwell Publishing Ltd

56

BRADLEY ARMOUR-GARB

person might well wonder whether all of the things you are committed to are correct. What you are really after, in order to communicate that you are inclined to accept each of the propositions that Obama thinks and says, might be to list each of the relevant propositions in question. So, you might start reeling off the following: (1) If Obama thinks or says that snow is white, then snow is white; and if Obama thinks or says that education is important, then education is important; and if Obama thinks or says that art is dead, then art is dead; and so on, ad innitum. Now, the problem here is obvious: There are far too many propositions for you to utter, in order to convey what you intend to convey. After all, you are, I assume, a nite being, and, being nite, you could not possibly assert the set of propositions included in the conjunction, (1), mentioned abovethere just isnt enough time for that. What we need, in order for you to convey your commitment to all of the propositions that Obama thinks or says, is a nite expression that will manage to express precisely what you intend to convey, were you able to afrm the relevant conjunction of propositions. Heres where truth comes in. For, (2) The proposition that art is dead, for example, is taken, by the minimalist, to be equivalent to (3) The proposition that art is dead is true. And similarly for all of the other relevant instances. Accordingly, the potentially innite series of conjunctions mentioned above can be converted into the form: If Obama thinks or says that p then that p is true. The minimalist will then gather these instances together to yield (4) Everything Obama thinks or says is true. This proposition is nite, and asserting it has the desired effect of committing you to each of the propositions that Obama thinks or says. Thus, this gets at the minimalists central point: it enables us to express commitments that, given our limitations, we could not otherwise express. This, according to the minimalist, is why we have the truth predicate. What this means is that, unlike most other predicates, is true is not used to attribute to certain entities (statements, beliefs, and so on) an ordinary sort of propertya characteristic whose underlying nature will account for its relation to other ingredients of reality. Therefore, unlike most other
2013 The Author Metaphilosophy 2013 Metaphilosophy LLC and Blackwell Publishing Ltd

A MINIMALIST THEORY OF TRUTH

57

predicates, is true should not be expected to participate in some deep theory of that to which it refers. In closing, I should note that I nd minimalism to be one of the most promising suggestions regarding the nature and structure of truth to have been made since Tarskis work. Of course, the view has been challenged, but it has set the agenda for much recent work on truth and is seen by many to be very appealing, given its minimal metaphysical commitments. Department of Philosophy University at Albany, SUNY Albany, NY 12222 USA barmour-garb@albany.edu References Horwich, Paul. 1990. Truth. Oxford: Oxford University Press. Alfred Tarski, Alfred. 1983. The Concept of Truth in Formalized Languages. In Logic, Semantics, Metamathematics: Papers from 1923 to 1938, edited by John Corcoran, 152272. Indianapolis: Hackett.

2013 The Author Metaphilosophy 2013 Metaphilosophy LLC and Blackwell Publishing Ltd