Kripke on empty names
I (somewhat randomly) picked up Kripke 2011 the other day. This is Kripke's first engagement with the problem of empty names. What struck me is the biased selection of examples. Most of the paper is concerned with names of fictional characters like 'Sherlock Holmes', and Kripke only seems to consider simple utterances in which they figure as the subject, like (1).
He argues, plausibly enough, that an apparent assertion of (1) should be understood as a pretend assertion, which only requires that it has a content in the context of the pretense.
Kripke also points out that it's hard to evaluate (1) at counterfactual scenarios: what would have to be the case for (1) to be literally true? It's not enough, he says, that the descriptive claims of the Sherlock Holmes stories are true at the scenario. If we identify the "proposition expressed" by an utterance with its pattern of truth-values at counterfactual scenarios (what Jackson 2004 calls the "C-intension" of the utterance), our inability to evaluate (1) at counterfactual scenarios supports the idea that it expresses no proposition.
I assume Kripke would make similar points about (2).
Since Vulcan (the hypothetical planet) doesn't exist, it is not obvious what would have to be the case for it to be smaller than Mercury. Moreover, an utterance of (2) is intuitively defective: the speaker presupposes that 'Vulcan' refers, and this presupposition fails. (We understand what the speaker is trying to say, Kripke might suggest, by acquiescing to her error, but she isn't actually saying anything.)
But let's look at some other examples.
Isn't this simply false, since I've had breakfast on Earth?
More seriously, we can use well-known devices to block the projection of presuppositions:
These are odd in a context in which the speaker knows that Vulcan doesn't exist. But suppose she doesn't. She clearly doesn't presuppose that 'Vulcan' refers. As a result, there's nothing defective about (4) and (5). The speaker of (4) or (5) does not commit an error. We don't understand them by acquiescing to the error, evaluating them on the pretense that 'Vulcan' refers.
I also think Kripke was too quick to conclude that we can't evaluate (1) or (2) at counterfactual scenarios. Granted, (1) isn't true at a scenario even if all the descriptive claims of the Sherlock Holmes stories are true at the scenario. But this is compatible with the hypothesis that (1) is false at every scenario. On this account, (2) may also be false at every scenario, but (4) is true at every scenario, and (5) is true at some scenarios and false at others. Isn't this a more accurate picture?
(Of course, these C-intensions are not plausible candidates for what is said, but that's a general fact about C-intensions.)