This Thursday: Elmar Unnsteinsson on Unarticulated Constituents

Our speaker this Thursday (March 22nd) will be Elmar Unnsteinsson, a PhD student in the CUNY Graduate Center’s philosophy program. Elmar will present a work in progress called ‘Propositions and Unarticulated Constituents’. The abstract follows:

In a recent paper, Adam Sennet poses an interesting problem for theorists who wish to postulate unarticulated constituents (UCs) of propositions. He takes note of sentences, such as (1), that seem not to be captured by standard definitions of UCs.

(1) Brooke comes to Vancouver when it snows.

On Sennet’s construal, the snow-location is a constituent of the proposition expressed by (1). But in contexts where the snow-location is Vancouver the constituent cannot be unarticulated since ‘Vancouver’ occurs in (1). Thus, Sennet argues, the UC-theorist must introduce occurrences of constituents in propositions and each occurrence must be coordinated with some bit of the sentence (hidden or not). Yet a central motivation for believing in UCs in the first place is that, syntactically speaking, they come for free. If Sennet is right, UC-theorists need to posit hidden syntactic structure just as much as anyone else.

In this paper I argue that Sennet’s argument fails. His examples don’t give rise to UCs at all. Most importantly, Sennet is conflating, like most others seem to do, the phenomena of linguistic underspecification and constituent unarticulation. Further, if one considers analogous cases his examples shouldn’t come as much of a surprise. Thus the issue of ‘occurrences’ is independent of whether one accepts UCs or not.

I also develop a Quinean argument against accepting occurrences. It seems almost impossible to find evidence showing that the same exact thing should occur twice in a single proposition. And this is especially pertinent with respect to Sennet’s examples, since one of the occurrences is not supposed to correspond to any linguistic expression. How can we be sure in such a case that the UC and the articulated constituent aren’t just co-instantiated, rather than identical? I give reasons to think that this possibility should be ruled out and conclude that occurrences should, at least in this case, be avoided.

Elmar has suggested that we may want to look at Sennett’s paper, ‘Unarticulated Constituents and Propositional Structure’, as background. You can download a copy here.

As usual, the workshop will take place at 7:00 in the third floor seminar room of NYU’s philosophy building.