Mereological Nihilism

I have been thinking a bit more about the (B) theorist¹ in my previous post. It looks as if allowing the content of one’s propositions to be determined by one’s intentions can leave one open for liar sentence analogs. Consider this sentence: ‘I am intending with this utterance to express a proposition different from the one determined by the sentence which I am expressing.’ In this case, if p₁ is that which we would normally assign to the sentence, the proposition expressed is different, p₂, and p₁≠p₂. On the other hand, p₂ then becomes the proposition assigned to the sentence which he expressed, thus making it not in fact the proposition intended. An easy response would be that utterances like this have no propositional content, but I think this is somewhat embarrassing for the (B) theorist.

If we take the no-content view seriously we then allow one’s underlying theoretical commitments to dictate when a proposition is actually expressed by a sentence. What would happen, then, if someone who held a theoretical view that propositions were not dictated by one’s intentions were to utter the liar-analogous sentence? If the sentence expressed a proposition in this case, then we would have weird cases where someone who wavered between the two views would on some occasions express propositional content with a particular sentence and on other occasions not. If, on the other hand, the (B) theorist were to hold that propositions are dicated by intentions generally, and one who does not hold this view would express no content on liar-analogous utterances as well, then it would seem that any percieved gap between the (B) theorist and the (A) theorist would close, as the (B) theorist would be passing the buck, simply positing a wide-ranging error-theory on a different level.

Over the course of the last semester, I have had the chance to have morning coffee with Dr. Kvanvig and Ryan Byerly. During the last few weeks, we have had some interesting conversations on mereological nihilism, specifically regarding what exactly is expressed when a purported nihilist engages in conversation with the folk. It seems that there are two general views on what kind of proposition is expressed when making utterances regarding things such as chairs and tables: either (A) the nihilist would propose that the propositions expressed in folk-contexts are in fact propositions quantifying over some form of simples, attributing a wide-ranging error theory, or (B) the nihilst would propose that the propositions she asserts are of the (A) form, but when one does not have such a theoretical view undergirding one’s assertion, one is simply quantifying over the composites.

I find the (B) line of argument troubling, as it seems to presuppose that one may intentionally manipulate the proposition one expresses. Language is a social phenomenon. The (B) view here seems to imply that, rather than a shared language, we, as speakers, are all speaking individual idiolect which happen to share a surface form and structure in such a way as to accidentally match up such that we manage to understand one another. This is risky business. I do not see how any proper semantics could be developed on such a view, for it would be quite possible that two subjects might seem to communicate (as the surface structure of their idiolect align), yet their intentions fundamentally diverge in their communication-acts. To provide an example, suppose my idiolect is bisimilar to that of a random strangers in all respects, except for the directions left and right. I have these reversed. I get out of a taxi in Chicago, and ask said stranger for some directions to a particular building I am interested in. The stranger, knowing nothing of Chicago, provides a random set of directions, which, according to his idiolect, happen to be completely wrong. But as I have the directions ‘left’ and ‘right’ reversed, the set I interpret from his utterances happen to be correct. I go off on my merry way following the directions I received and manage to successfully make it to the building in question. The question remains however, as to the semantic value of ‘left’ and ‘right’ in this example (assuming that the semantics is compositional, which is pretty standard these days).