1769. According to the analysis on which . . .

SCANNED PDF

According to the analysis on which I have long relied, "true" is simply "the most general adjective of commendation" pertaining to matters of belief (Toulmin). Thus, to call a statement "true" is to commend it as credible, in the sense of being worthy of belief or acceptance. This means that "true" functions similarly to such other familiar terms of commendations as "good" or "beautiful," in that, like them, it is, as Toulmin says, a "gerundive"; as such, it does not express a merely subjective preference or state of mind, but has the force of claiming objective worth—specifically, of assessing a given statement or belief as in itself worthy of credence. But this entails that "true" is responsibly used only when it is possible to specify at least some criterion or criteria of credibility, in the sense of worthiness to be believed. Because when we say that a statement is "true" we do not mean simply that we believe it but that it is worthy of being believed, we commit ourselves in saying it to stipulating the standards that govern our so using "true" in the case of this or any other statement of the same logical type. It further follows that, although the meaning of "true" is constant, the standards governing its use in different cases are variable—or, as we may also put it, in Toulmin's terms, although the "force" of "true" is "field-invariant," the "criteria" necessarily implied in properly using it are "field-dependent."

There clearly seems to be a certain convergence between this analysis of "true" and that offered by Searle in The Construction of Social Reality, according to which "'true' and 'false' are evaluative terms used to describe certain kinds of success and failure." Specifically, "[t]hey are used to assess [the] success or failure of statements (and beliefs) in achieving . . .the word-(or mind)-to-world direction of fit" (208). Thus "'truth' applied to statements is a term of assessment implying trustworthiness" (210). "We need a metalinguistic predicate for assessing success in achieving the word-to-world direction of fit, and that term is 'true'" (219). On Searle's analysis, then, "true" applied to statements is simply "a general term for the feature of trustworthiness as it applies to statements." Similarly, "fact(s)" is simply "a general term for what makes statements trustworthy, what it is in virtue of which they are reliable" (210 f.). And so, also, with "correspondence," which answers to our need for "a verb for describing the relations between the statements and the facts when the statements are true" (213). "Just as we need a general term for all the different features of the world that can make statements true [viz., 'facts'], so we need a general term for naming the ways in which true statements can accurately represent how things are in the world, and 'corresponds to the facts' is just such a general characterization (213). This means that, although "both statements and facts are specified propositionally, 'the fact that...' and 'the statement that...,'" still "facts are not thereby linguistic in nature." "Facts are not the same as true statements" (220).

But now how, exactly, do these two analyses fit together? The connection between them, I judge, can be seen by distinguishing between: (1) commending a statement as true, i.e., credible or worthy of belief, or, in Searle's terms, as trustworthy or reliable, because it satisfies a certain criterion or criteria of truth; and (2) the necessary condition of its being able to do this and of its therefore being credible. If the invariable force of "true" is to commend a statement as credible because it satisfies at least some variable criterion or criteria, the invariable condition necessary to the possibility of its doing this, and thus to its being worthy of belief, is that it corresponds to things as they really are. No statement could be credible or trustworthy that did not thus correspond with what is really the case independently of the statement. "Statements are assessed as true when they are trustworthy, i.e., when the way they represent things as being is the way that things really are" (219).

In sum: Searle's analysis significantly supplements the one on which I have hither relied by clarifying the necessary condition of the possibility of a statement's being worthy of belief and therefore commended as "true." A statement can be properly assessed as "true," and hence as worthy of belief, if, and only if, it succeeds rather than fails in achieving "the word-to-world direction of fit." In this sense, a statement is true if, and only if, it corresponds to the facts—or, as I should prefer to say, allowing for the strict meaning of "facts," if, and only if, it corresponds to reality, or to things as they really are.

January 1996; rev. 8 August 2002