Latest Update: March 20, 2008

Reading Group: Current Topics in Formal Pragmatics

Mission

Meet biweekly in order to discuss one or several recent(ish) papers concerned with formal approaches to pragmatics, broadly construed.

Time and Place

TBA in Margaret Jacks Hall (B460), Room 127B ("Chair's Office").

Mailing List

If you want to receive email announcements/reminders of the meetings, you can subscribe yourself to the mailing list.

Upcoming

Date: Wednesday, April 1st, 2009
Time: 11.00 am
Place: Margaret Jacks Hall, Room 126 ("Greenberg room")
Speaker: Raj Singh (MIT)

Symmetric and Interacting Alternatives for Implicature and Accommodation

Standard theories of presupposition projection (eg. Karttunen and Peters 1979, Heim 1983, Beaver 2001, Schlenker 2007) predict that the sentences in (1) and (2) should presuppose that if John flies to Toronto, he has a sister:

  1. If John flies to Toronto, his sister will pick him up from the airport
  2. Mary knows that if John flies to Toronto, he has a sister

But this prediction seems correct only for (2). Sentence (1) seems to presuppose something stronger, namely, that John has a sister whether or not he flies to Toronto. I will assume, following Beaver (2001), von Fintel (2008) and others that this so called proviso problem (Geurts 1996, 1999) is a problem for the theory of accommodation (in the sense of Lewis 1979): Why, when we hear sentences that presuppose the same information, do we accommodate such different kinds of information? Although several proposals for dealing with the proviso problem exist (eg. Beaver 2001, van Rooij 2007, Singh 2007, von Fintel 2008), a general, fully predictive solution has not previously been developed.

I will propose a solution to the proviso problem that I hope will achieve the desired generality by developing a single (unified) theory for accommodation and scalar implicatures. Such a proposal has obvious roots in Gazdar (1979). However, the details of my proposal will be quite different. First, in response to any given sentence, two candidate sets are computed based on formal principles: a set of potential implicatures (as standardly assumed since Horn 1972), and (and this is the innovation) a set of potential accommodations. Second, these two candidate sets interact with each other along lines previously proposed for the theory of implicature (e.g. Sauerland 2004, Fox 2007): If there are candidates (from either candidate set) that are inconsistent with each other, they will cancel each other out so that neither one will become actual. I will argue that variance in such logical properties of candidate sets is the key to solving the proviso problem, in addition to a few other problems (in the theory of implicature) that I will identify.

Past Meetings

Contact

Sven Lauer
sven DOT lauer AT stanford DOT edu
Margaret Jacks Hall, Room 105
(650) 723-9019