LINGUIST List 14.2725

Fri Oct 10 2003

Diss: Semantics/Pragmatics: Taranto: 'Discourse...'

Editor for this issue: Takako Matsui <>


  1. taranto, Discourse Adjectives

Message 1: Discourse Adjectives

Date: Thu, 09 Oct 2003 19:15:37 +0000
From: taranto <>
Subject: Discourse Adjectives

Institution: University of California, San Diego
Program: Linguistics Department
Dissertation Status: Completed
Degree Date: 2003

Author: Gina C Taranto 

Dissertation Title: Discourse Adjectives

Dissertation URL:

Linguistic Field:	 Semantics 
			 Discourse Analysis

Dissertation Director 1: Chris Barker

Dissertation Abstract:

This thesis introduces Discourse Adjectives (DAs), a natural class
whose members include apparent, evident, clear, and obvious, as in

(1)	a.	It is clear that Briscoe is a detective.
	b.	It is clear to you and me that Briscoe is a detective.

Of primary concern are the semantics of DAs in sentences like (1a), in
which the conceptually necessary experiencer of clear is not expressed
syntactically, and is interpreted much like (1b), with the relevant
experiencers of clarity interpreted as the discourse participants -
that is, both the speaker and the addressee.

I argue that the meaning of utterances such as (1a) are highly unusual
semantically, in that they operate entirely on a metalinguistic level.
Interlocutors use such utterances to provide information about their
conversation rather than their world. Sentence (1a) does not provide
new information about Briscoe, rather, it provides information about
the interlocutor's beliefs about the designated proposition, in terms
of the current conversation.

My analysis begins with a Stalnakerian model of context-update, as
formalized by Heim (1982, 1983) and Beaver (2000). I augment this
model with Gunlogson's (2001) representation of individual commitment
sets of speaker and addressee within the Common Ground of a discourse,
and Barker's (2002) compositional theory of vagueness.

My proposal relies on the (vague) degree of probability that the
Discourse Participants assign to the truth of a proposition; the
context-update effect of an utterance of (1a) removes from
consideration those possible worlds in which the discourse
participants do not believe the proposition expressed by Briscoe is a
detective satisfies a vague minimum standard for 'clarity'. The
semantics of utterances with DAs are shown to depend directly on
probability, and only indirectly on truth. I argue that after an
utterance with a DA is accepted into the Common Ground, interlocutors
are licensed to proceed as if the designated proposition is true, if
only for the current discussion.

DAs are argued to have the ability to publicly commit all discourse
participants to the content of their complements. This is shown to
have a synchronization effect on the Common Ground of a discourse,
which explains how it can be useful to have an expression type that
has no normal descriptive content.
Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue