(This continues the line of thought from my last post.)
4. If we were to stick rigidly to Davidson’s views in all respects, we’d be even worse off, for then we’d be lacking even a way to formulate this. That’s because Davidson insists that there is no such thing as metaphor in semantics at all: what is communicated in language are only literal meanings. (What counts as metaphor in imagistic or poetic language is, according to Davidson, entirely an aspect of linguistic use, a dimension of pragmatics. It can cause hearers to change perspectives on something, or even suggest perceptions, but there is no further content communicated — nothing over and above, that is, the literal meaning of the linguistic expressions involved. Cf. his “What Metaphors Mean”; TI 245-264.)
Now we have already seen that Cassandra doesn’t need to even formulate (let alone utter) a linguistic expression such as “Damocles has just died.” — nor indeed one like “This clock has stopped.” But even if she did formulate (even utter) the latter, it couldn’t have the metaphorical meaning of the former (according to Davidson). The only thing it would communicate is its literal meaning, anything additional that might cause a hearer to perceive more than just the immediate physical fact (the clock’s stopping) would be caused by the utterance, given extralinguistic contextual elements (the general situation, Damocles being of declining health, him having been admitted to the hospital, and so on). But obviously, there is no utterance here, and there is no hearer. We have no situation where someone causes additional psychological processes over and above a communication of linguistic meaning. None of these elements are given here. We only have Cassandra’s perception (that the clock has stopped) and her sudden realization (that Damocles has died). We do have what looks very much like a metaphorical connection; or, to use a consideration Davidson himself often deploys with (some) argumentative weight, we have an intuition that there is some conceptual connection here (and “metaphorical” seems apt, since it conceptualizes something in terms of something else). But that’s not something Davidson even admits in his theory. On the other hand, we have no causal connections, which Davidson would admit: for the truth-making event does not cause the belief; and the event that does precisely that (the stopping of the clock) has no semantic connection at all with the content of the belief. (Or at least none that Davidson would admit.)
But that leaves it mysterious how there can be a true belief here at all. Compare what Davidson insists is the paradigm case:
Communication by language is communication by way of literal meaning […]. The theory of truth deals with the literal sense. […] Cleaving to the literal, then, someone speaking English will make a true statement by uttering the sentence ‘It’s Tuesday’ if and only if it is Tuesday in his vicinity when he speaks. (“True to the Facts”, TI 45)
Similarly then, Cassandra’s belief that Damocles just died is true if and only if Damocles just died; and the “in the vicinity” clause here must mean at least that there is something close by that both triggers and justifies that belief formation (or at least contributes to its justification). And of course there is: in the doctor variation, it’s his facial expression which Cassandra perceives; and in our original example it’s the stopping of the clock. But the latter seems to be admissible only by virtue of its metaphorical connection; dismissing that, we have nothing here “in the vicinity”, and we’re back to making it a mystery how the belief can be formed (and justified). Davidson’s interior design isn’t just too sparse by itself: he also shuts the door for anything useful to come in.
Now, obviously working with a Davidsonian framework does not require us to subscribe to every theoretical item Davidson himself held, and clearly his theory of metaphor (which denies any form of metaphorical meaning and declares metaphor to be entirely a matter of pragmatics) is supremely unhelpful in our case. But we’re not much better off if we were to accept “metaphorical meaning”. If we allowed that, pace Davidson, we’d still have a missing link: Cassandra, on seeing that the clock has stopped and grasping the metaphorical meaning of that event, might form the belief, understandably so, in that moment — so we have a trigger. But how is it justified? And how is it connected with the event which actually does make it true? (There is a gap in the justification, for as we have seen, there is no regularity in common sense experience that connects deaths and stopping clocks; there is a gap in truth-making, for the events of the death and the stopping clock are unrelated, merely coincidental.) Thus even allowing “metaphorical meaning”, we’re still left with a situation that is theoretically indistinguishable from the “wild guess” scenario we discussed earlier, while our example clearly is not of that type. To bridge the gap, there would have to be a connection between the two events on a deeper level than even semantics: something that plays the role otherwise played by causal relations between events, that is, something factual.
