Strawson argues that (1) the Truth-Condition theory must fully disclose the general notion of truth it is presupposing, but (2) the only plausible general analysis of truth is in terms of stating or asserting things, which is to say in terms of communication, which must be cashed á la Grice in terms of speakers' psychological states.  Thus although the letter of Davidsonian semantics may be unobjectionable, its spirit is broken, for it collapses into Griceanism rather than standing as a superior alternative.  (Strawson does not say it's useless to talk of truth-conditions; I don't know whether he privately thinks that.)
    First we must unpack (1).  Does Strawson mean that (a) we don't understand "true" well enough for the Truth-Condition theory alone to be informative or useful?  That's what's suggested by his phrase, "the concepts of meaning and truth each pointing blankly and unhelpfully at the other" (p. 116).  Or does he mean only that (b) although we understand "true" well enough for ordinary purposes, one who rests his theory of meaning on "true" bears a heavier burden, that of producing a plausible philosophical analysis of "true"?
    If (a) is what Strawson means, I am entirely unconvinced.  As I said in my textbook, the notion of truth is not like the made-up mystery concept "krum."  Whether or not we have an accepted general philosophical analysis of truth, we have the concept expressed by the ordinary English word "true," and we have the formal structure introduced by Tarski, that generates T-sentences.  If the notion of meaning can be reduced to that of truth, that is a theoretical economy even if we provide no further philosophical explication of truth.
    So let's suppose that (b) is what Strawson means.  For a while in class I complained about the principle that (b) seems to presuppose, the principle that if a philosopher explicates X in terms of Y, then the philosopher had better come up with a convincing theory of Y in turn.  (Notice that that principle leads to regress.)  But if Strawson is right about (2) above, he doesn't need to appeal to that principle, for he can argue as follows:  OK, Davidson need not offer any analysis of "true."  But (2) is correct.  And so, in explicating meaning in terms of truth, Davidson is really, in effect, explicating it in terms of Gricean psychological states.  Hence, his theory is no advance over Grice's.
    Let's here set aside the question I raised at the beginning of class, that of whether Grice's Stage I really just helps itself to Davidson and Stage II is merely a theory of referring.  Then it seems true that if (2) is correct, Davidson's theory  is no (further) advance over Grice's.  So we must ask whether (2) is right.
    And that is the weird feature of Strawson's argument.  As before, what about all the other general theories of truth that philosophers have offered over the past two thousand years--the Correspondence theory, the Coherence theory, the Pragmatic theory and more recent deflationary theories such as the Disquotational and the Prosentential?  Strawson simply ignores these.  As I said, I think he must be assuming that every such theory would somehow have to buy into the Gricean idea at some early stage:  For example, since beliefs are primarily what cohere or fail to cohere, the Coherence theory of truth would have to treat sentences only insofar as they express beliefs, etc.  But I do not see why we should grant that assumption (if it is Strawson's) just on his sayso.   On this interpretation, Strawson is in the position of saying, at bottom, "But surely some version of Griceanism is correct."  Beware the "surely" operator here and everywhere else in philosophy.