Propositions and Proto-attitudes

I have been going through Jeff King's book The Nature and Structure of Content in my metaphysics seminar and wanted to put down some thoughts (time permitting).

Very generally, King's account of structured propositions is as follows. Propositions are certain kinds of facts whose constituents are structured in the same way as the sentences which express them are structured at LF. Consider, for instance, the proposition that Rebecca swims. On King's account, this proposition is analysed as:

The fact of there being a (possible) context c and lexical items a and b of some language L such that a has as its semantic value in c Rebecca and occurs at the left terminal node of the sentential relation R that in L encodes the instantiation function and b occurs at R's right terminal node and has as its semantic value in c the property of swimming.

It is clear from reading the book (though the issue is not addressed directly) that King intends his language variable to range over languages-in-use (rather than, for example languages as abstract objects). Consequently, on King's view languages are mind-dependent entities which do not exist prior to the development of language.

This gives rise to the following worry. On the most widely accepted views of what it is for a language to be the language of a given community, this will involve various highly complex propositional attitudes (intentions, beliefs, etc.) on the part of the members of that community. But it seems clear that one cannot have those propositional attitudes unless, well, there are propositions. So it appears that King is commited to the claim that there must be propositions prior to language. But this is apparently inconsistent with his view of propositions.

King notes that he has two options (i) adopt the Language of Thought Hypothesis or (ii) appeal to protobeliefs/protointentions. The second option is important to King because he doesn't want to hang is hat entirely on the LOT hypothesis. But I am doubtful that this is a genuine option for him. Call the "contents" of the protobeliefs/intentions "protopropositions". Whatever else we want to say about protopropositions, it seems clear that if they are going to do the work that King wants they had better be both truth-evaluable and structured. But these are two of the primary characteristics of propositions which King's theory was supposed to capture. If he is now commited to giving a theory of structured protopropositions with these characteristics, it is likely that whatever theory is put in place can be adapted to handle propositions directly. If that is right, then King is more heavily committed to LOT than his discussion in the book suggests.   

Comments

[Robbie Williams]

Hmmm... I can't quite recall King's discussion, but do you think he needs to think that protopropositions are structured? E.g. Lewis and Stalnaker's stuff on how beliefs and intentions come into play in fixing what language we're using (otherwise very different from each other) both explicitly appeal to unstructured contents. It's not obvious to me what's wrong with positing such coarse-grained proto-attitudes---and if they can do the work, you've got a non-circular route in.

[marc moffett]

Hi Robbie. Good point! I hadn't really thought this part through completely.

First, a quick point about the Lewis/Stalnaker view. I take it that they allude to propositions and propositional attitudes in their articulation of theory. It is a further and independent question whether or not propositions are as they say they are. Given that I am deeply skeptical of their view of propositions (as, obviously, is Jeff), it is not uncontroversial that they can do what they want to do w/unstructured entities.

Now one of the reasons Jeff gives us for adopting his view of structured propositions is that it makes transparent why propositions are truth-evaluable. So at the very least, it would be a bit perplexing for Jeff to claim that it is nevertheless possible to transparently explain the truth-evaluability of protopropositions if they are unstructured. Second, Jeff claims in the first chapter of the book that the demand that propositions need to be recursively specifiable also gives us a major reason for adopting structured propositions. But I think that protopropositions will also need to be recursively specifiable in more or less the same way. And so, one would think that his prior reasoning would apply to protopropositions as well.

So, in effect, if we spot Jeff his reasons for thinking that propositions are structured, those same considerations will apply mutatis mutandis to protopropositions. Conversely, it is hard to see why any plausible view of unstructured, truth-evaluable protopropositions could not be applied to propositions proper. So I guess I am suggesting a kind of dilemma for Jeff: either he can give a theory of unstructured protopropositions or he can't. If he can, then arguably he can give a theory of unstructured propositions; if he can't then (LOT to one side) he is going to have to give an independent theory of protopropositions which arguably can be applied to propositions. Either way, his own "linguisticky" theory of propositions becomes superfluous.