Donal McEvoy wrote: "1. Whether a statement is true or corresponds with facts depends on, and only depends on, whether the 'statement' corresponds with the 'facts'. Eg. 'Snow is white' is true if and only if [or 'iff.'] the snow is, in fact, white." Since Donal is interested not in the sentence 'Snow is white' but the statement 'Snow is white', this "theory of truth" must be epistemic. After all, what is a statement but a sentence asserted to be true? But then the problems just pile up. How does an assertion "correspond" to anything? To determine what "fact" is being fixed on by the sentence 'Snow is white', one cannot turn to the sentence itself but rather the assertion of the sentence. Yet, Donal's account of Tarski's T-Convention has the sentence, not the assertion, corresponding to a fact. Finally, how is a correspondence between a linguistic phenomenon, the 'statement', and a presumably non-linguistic entity, the 'fact', possible? I would suggest that this is not a helpful approach to a "theory of truth". Donal continues: "1.1.2.3 Tarski's theory denies that a statement, even one we 'know' or 'believe' to be true, can ever be true (despite how much we feel we 'know' or 'believe' it) if it does not correspond to the existence of the facts that the statement posits as existing." There is a far simpler reading of Tarski's T-Convention that does not require dubious references to 'facts' and 'existence', namely, that a sentence in a particular language is true iff it is the sort of sentence native speakers would agree is true. A: "It's raining" B: "Yup" And a theory of truth just accounts for all such instances. Sincerely, Phil Enns Toronto, ON ------------------------------------------------------------------ To change your Lit-Ideas settings (subscribe/unsub, vacation on/off, digest on/off), visit www.andreas.com/faq-lit-ideas.html