Cresswell's Colleague TLM

In an illuminating criticism ([ 1] ) of an article of mine ([2]), Cresswell sketches the semantical views of a Traditional Logician known as TLM, whose final initial must I suppose stand for Montague, and who, it is claimed, also shares the outlook of David Lewis, David Kaplan, Dana Scott, and perhaps A. N. Prior. I agree that the views of TLM correspond reasonably well with those of Montague, though I think Prior's differ in certain respects; the others must speak for themselves. On internal evidence, it is likely that the views of TLM are a good representation of those of Cresswell. To motivate discussion, I should say that it seems to me that the project of a dialectical reduction of logic is an extremely interesting one, and one that someone ought to be working on. If we knew how to program a computer to speak English well enough to be able to take part in conversations with ordinary English users, not only language but also logic need have no further mysteries for us. But although such a machine would have to be capable of registering various features of its context, it would not need an accurate built-in clock or calendar, since humans do not. Consequently, logic should not need to make reference to real times and dates. It is of course already accepted by most logicians that logicality is consistent with faulty sense-organs, but the point apparently needs special argument for the case of time, which Kant thought was perceived by an "internal sense". Naturally, the truth of a tensed statement such as "It's Monday" cannot be determined by someone who has no information about the time of its utterance. But truth, unlike necessary truth and logical relations such as incompatibility, is not a logical property and should enter logic at most hypothetically. Now when someone says "It's raining today" on 21st March 1973, a day which for short I shall refer to as day D, his