What does the deflationary theorist mean by saying that truth is not a genuine property of statements?
Is the deflationary theory itself true?
“[T]ruth is not the sort of thing one should expect to have a philosophically interesting theory about.” Richard Rorty(1)
“[W]hen we have pointed to certain formal features of the truth-predicate (notably its “disquotational” feature) and explained why it is useful to have a predicate like this (e.g. as a device for asserting infinite conjunctions), we have said just about everything there is to be said about truth.” Michael Williams(2)
Part 1 – Truth as a Property
The deflationary theory of truth is a family of theories which all have in common the claim that assertions that predicate truth of a statement do not attribute a property called truth to such a statement.(3) In order to make this claim, the deflationary theorist distinguishes between “logical” properties and “explanatory” (or in the title’s usage “genuine”) properties.
Consider the following two set of statements –
| London is the capital of England. | Liverpool is the capital of England. |
| The Earth revolves around the Sun. | The Sun revolves around the Earth. |
| Two plus two equals four. | Three plus two equals four. |
| Snow is white. | Snow is green. |
| This sentence contains five words. | This sentence contains nine words. |
The statements on the left all share an obvious similarity, while the statements on the right also share a closely related similarity. The similarity shared by the statements on the left we call “True”. The closely related similarity shared by the statements on the right, we would then call “not-true” or “false”. So obviously those on the left do indeed share a property that of being “True” that is not shared by those on the right. In this sense at least, it is obvious that “True” is unquestionably a property of some statements. And this sense in which truth is a property is not denied by the deflationary theorist.
However, when we say of two things that they are both F, we often mean to imply that they are both F in virtue of some common explanation as to why they are both F. For example, when we say that snow, light puffy clouds, parts of the Union Jack, hospital bed sheets, and this sheet of paper are all white, we imply that they are all white in virtue of some common explanation i.e. (in this case) their common optical reflectance behaviours. It is this explanatory implication that the deflationary theorist denies applies to the property “Truth”.
The property “exists” behaves somewhat differently from the property “white”. To say that London, the Earth, snow, light puffy clouds, the Union Jack, and hospital bed sheets all exist, we do not imply that they exist in virtue of some common explanation. There are many similar predicates. A “logical” predicate may be whatever you please in the sense that you can choose to group things together in any way that interests you, without relying on any explanatory sense of why they are the way they are. Hence, “mine” (in the sense that “These things are mine”), “close by”, “over there”, “exists” (according to some philosophers), are all logical predicates. They all collect together things that share some common feature, without implying any common cause behind that feature. It is in this sense that the deflationary theorist maintains that predicate “True” is a “logical” predicate. Whereas “white”, “loud”, “pungent”, “Sour”, “rough”, “fast” and other “genuine” predicates collect together what they are predicated of in virtue of some common explanation.
So the deflationary theorist can maintain that “Truth” is not a genuine property of statements by maintaining that “Truth” is a logical predicate, and not an explanatory predicate. The raison d’etre of the truth predicate is to permit the re-expression of possibly infinite sets of statements as a single statement e.g. “Everything Paul says is true.”
The deflationary theory of truth is, of course, a highly debatable position to maintain. On the one hand, common intuition and many philosophers argue that the deflationary theory is wrong, for various reasons. On the other hand, Paul Horwich in his 1990 work “Truth”(4) defended his minimalist version of a deflationary theory against 39 of these objections.
Part 2a – Truth and Meaning
The central thesis of the deflationary theory is that the notion of truth is implicitly defined by the T-schema of Alfred Tarski(5).
(T-Schema) <P> is true if and only if P.
Where <P> is the label of some statement or proposition, or standardly, the sentence or proposition in quotation marks, and P is the simple assertion of that statement or proposition, or standardly, the sentence or proposition without the quotation marks. (Hence, truth as a “device for disquotation”.)
As a particular example that appears constantly in the literature –
<Snow is white> is true if and only if snow is white.
But the real question is how this T-Schema is to be understood. If we understand the left side – <snow is white> to be a sentence, then the deflationary thesis is clearly false. For this sentence to be true not only must snow be white, but in addition “Snow” must mean snow, and “white” must mean white, and so forth. In other words, for any sentence to be true not only must the T-Schema be true, but it must be accompanied by an appropriate theory of meaning. More importantly, this theory of meaning must be framed in terms that specifically avoid the use of any explanatory sense of the predicate truth, or the deflationary theory becomes circular. In particular, this means that the deflationary theory of truth is definitely incompatible with any of the truth-conditional theories of meaning, and probably incompatible with verificationist theories of meaning. It is arguable whether it is incompatible with a referential theory of meaning.(4.4) Only a “use” theory of meaning appears to avoid the circularity.(4.5)
Alternatively, if we understand the left side as a proposition, then the deflationary thesis is rendered trivially true since the right side is also usually understood as a proposition. Now this may initially appear to be an advantage to the deflationary theorist, since he argues that “True” is not a complex predicate. But in this case, the T-Schema is rendered trivially true in virtue of the meaning of “proposition” rather than in virtue of the meaning of “Truth”. On a propositional interpretation, the T-Schema transfers all of its explanatory difficulties to the meaning of “proposition”, rather than letting them rest with “Truth”. So on this interpretation, the deflationary thesis is rendered uninteresting an “A is A” truism because the meaning of the right-hand side of the Schema becomes the same as the left.
Now consider the meaning of the right-hand side. What does it mean for snow to be white the P without the quotation marks. This side of the schema is clearly not intended to be a sentence. So is it to be understood as a proposition, as it usually is? If it is, then what is needed is another theory of meaning that interprets just what a proposition that snow is white is actually “Saying”. A theory of meaning that connects what the proposition is asserting to what the proposition is about. If the P without the quotation marks is not to be understood as a proposition, just how is it to be understood. If it is to be understood as an apt description of what is, what is the case, the facts, a state of affairs, the world, etc. then the T-Schema has evolved into a statement of the Correspondence Theory of truth. Other interpretations of what the unquoted P actually means would evolve the T-Schema into versions of the Coherence Theory. All of these consequences would entail that the deflationary theory of truth is wrong. The deflationary theory is lacking an adequate interpretation of just what it means to state (‘say”, assert, etc) that snow is white without the quotation marks.
Part 2b Truth as Conjunction
Consider the statement
“True beliefs about how to attain our goals tend to facilitate success in achieving them.”
This is one of the “laws of truth” proposed by Hillary Putnam(6). He offered these laws of truth as justification for concluding that truth is an explanatory predicate. The law explains our success in attaining our goals. The deflationary theory responds that these “laws” can be viewed as an abbreviated way of grasping the infinite conjunction of statements like that about snow being white. (Hence truth as a “device for infinite conjunction”.) However, this approach fails to distinguish between asserting the conjunction of particular instances and asserting the generalization. The generalization may incorporate an explanation of why it is a correct description of the world that is not implied by any one (or conjunction) of the particular instances. As Anil Gupta explains(7), providing an explanation of why each individual passenger on a boat died, is not the same as providing an explanation as to why they all died. “He drowned” might be accurate for each particular, but a conjunction of such reasons would not imply “The boat sank”. Conversely, “The boat sank” is not an acceptable cause of the death of any one particular. Hence, generalizations that are inductively concluded from conjunctions of particular instances can include information beyond that contained in the conjunction. So, contrary to the claims of the deflationary theory, it is not necessarily the case that
(a) “Everything that Paul said is true”
is equivalent to the infinite conjunction
(b) “If Paul said “P1” then P1, and if Paul said “P2″ then P2, etc.,”.
In any case where (a) incorporates more information than is implied by (b), the truth predicate has more to it than the deflationary theory allows. Conversely, we are not entitled to assume that for any similar use of the truth predicate, a conjunction (infinite or otherwise) of particular instances will be sufficiently equivalent to maintain that the truth predicate must have nothing further to say.
Part 2c Truth as Success
‘truth” is that target that beliefs, judgements, and assertions aim at. In the words of Bernard Williams(8) “Someone who asserts [a belief] to another standardly gives the hearer to understand that he can rely on the truth of what has been asserted, in particular that he may base his actions on that assumption.”
There is a fundamental conceptual difference between what we believe, and what we desire, wish, or speculate. There is a clearly recognised difference between a belief about something, and an opinion about something. Consider the proposition P=(“Pontius Pilate was a secret philosopher who really puzzled about the nature of truth”) I can utter the sentence p in a number of different mental “moods”. I can suggest p as a whimsical interpretation of the motivation behind his biblical (John 18:36 -38) question “What is truth?” I can say that I wish, want, or desire that P. I can hold or express the opinion that P. I can suggest that it is possible that P, given the absence of any evidence to the contrary. If I should express P in any of these moods, I am not pretending that my suggestion is really what motivated his question. If I should express P in any of these moods, I would not be intending you to understand that you can rely on the truth of P. But if I believe that P, and assert to you that it is my belief that P, then I am indeed intending to give you the understanding that P is true. The fundamental property that separates a belief that P from an opinion that P, a desire that P, or a whimsical suggestion that P, is that a belief that p intends to be true. So, what truth is, is that fundamental property that separates a belief from a whimsical speculation. And the deflationary theory has no contribution to this distinction between belief and the other possible moods of expression.
We can approach the same result from another direction. Our beliefs serve a definite purpose in our daily lives. We survive on a day to day basis because our beliefs guide us to wait for the traffic to pass. Without appropriate beliefs about the dangers of jaywalking across a busy street, many of us would perish in pedestrian encounters with vehicles every day. It is on the basis of our beliefs that we can discriminate things that we can eat from things that tend to eat us. “Truth” therefore is that property of beliefs that renders them more successful than their negations when those beliefs become the foundation for our expectations of future events or as the basis of our behaviour. If one’s belief is that there is no tiger behind that bush, then one will proceed on that basis. And one will either become lunch or enjoy lunch according as to whether the belief was true or false. Again, the deflationary theory can answer this functional notion of truth only by claiming that the success criteria can be specified without using the truth predicate as an infinite conjunction of the propositions that if our goal is to obtain F then one should do P. But we saw above that the generalization can contain more than the conjunction of particulars. And in this case what is missing from the conjunction of particular instances is any notion of which set of particulars is to be conjoined, and why those rather than others. The significance of this “Success attribute” of “Truth” is that it is clear that there is something other than what we believe that renders beliefs true. There has to be something more than what we believe.
Part 2d Truth and Metaphysics
It is frequently argued that the most obvious thing we can say of truth is that it consists of a correspondence to the facts. For example, <snow is white> is true if and only if [it is the case that / the facts are that / the state of affairs is that / an apt description of the world is that / etc.] snow is white. In other words, the common intuition (the “folk philosophy”) is that <snow is white> is true somehow because snow is white. These common intuitions about truth are based on a realist “folk metaphysics” understanding of what truth is. The metaphysical realist claims that a statement is true in virtue of some sort of correspondence relation to the world; this relation makes the statement true; it is because of this relation that the statement is true. The correspondence theory of truth provides a substantive response to the question “In virtue of what is it true that <snow is white> iff snow is white?”
The deflationary theory, however, is based on a linguistic analysis of how we use the predicate “True” in English sentences. It is a theory about linguistics about the words “True” and “Truth. The deflationist therefore rejects all such “in virtue of” reductive explanations of truth. The deflationary theory of truth is therefore metaphysically anti-realist about truth. Whereas the correspondence theory draws upon a metaphysical theory of what exists to explain what it is that truth means, the deflationary theory appeals to the meaning of “True” to explain why no explanation is necessary. Hence, part of the process of adjudicating between the deflationary theories of truth and other theories of truth, is resolving the dispute between the metaphysical realist and the metaphysical idealist. If one is a metaphysical realist, the deflationary theory of truth is a simple non-starter. If one is a metaphysical idealist, then the deflationary theory of truth becomes more interesting.
In either case, in addition to providing the “correct” theory of the meaning of truth, the deflationary theorist must also provide an intuitively acceptable explanation as to why our common sense “folk philosophy” intuitions about truth are so wrong.
Part 2e Deflationism and History
One final observation. Over the history of philosophy, from Plato and the Pre-Socratics down to the latest in developments, common sense intuitions have almost always prevailed over more esoteric and revisonistic philosophical musings. Modern philosophical thinking dismisses as unacceptably wide of the mark such notions as Plato’s Forms, Berkeley’s Mind of God, Leibniz’s monadology, Kant’s Innate Categories, etc., etc. If the deflationary theory needs to argue that the commonsense notion of “Truth” as an explanatory “genuine” predicate is unacceptably erroneous, then so much the worse for the deflationary theory of truth.
Notes & References
(1) Rorty, Richard; Consequences of Pragmatism, University of Minnesota Press, Minneapolis, Minnesota. 1982. Pg xiii.
(2) Williams, Michael; “Epistemological Realism and the Basis for Scepticism”, Mind 97 (1988); p g424
(3) Wikipedia contributors, “Deflationary theory of truth”, Wikipedia, The Free Encyclopedia, 28 April 2008, 14:41 UTC=<http://en.wikipedia.org/wiki/Deflationary_theory_of_truth> [accessed 10 September 2008]
(4) Horwich, Paul, Truth Second Edition, Clarendon Press; New York, New York, 1998.
ISBN 0-19-875223-7.
(4.4) Schwartz, Stephen P. (ed.); Naming, Necessity, and Natural Kinds, Cornell University Press, Ithaca, New York.1977.
(4.5) Horwich, Paul, “Meaning, Use and Truth: On Whether a Use-Theory of Meaning Is Precluded by the Requirement that Whatever Constitutes the Meaning of a Predicate Be Capable of Determining the Set of Things of Which the Predicate is True and to Which It Ought to be Applied”, Mind, New Series, Vol. 104, No. 414 (Apr., 1995), pp. 355-368.
(5) Tarski, Alfred; Logic, Semantics, Metamathematics: Papers from 1923 to 1938, 2nd edition Corcoran, J., ed. Hackett Publishing Company. 1st edition edited and translated by J. H. Woodger, Oxford University Press. Oxford, England. [1956] 1983. ISBN 0-915-14475-1.
(6) Putnam, Hillary; Meaning and the Moral Sciences, Routledge Kegan & Paul, New York, New York, 1979. ISBN 0-710-00437-0.
(7) Gupta, Anil; “A Critique of Deflationism”, The Nature of Truth, Lynch, Michael P. (ed.), The MIT Press, Cambridge, Massachusetts. 2001. pg 534-539.
(8) Williams, Bernard; Truth and Truthfulness; Princeton University Press, Princeton, New Jersey 2002; ISBN 0-691-10276-7; Pg 79.