LINGUIST List 14.788

Tue Mar 18 2003

Diss: Syntax: De Cat "French Dislocation"

Editor for this issue: Anita Yahui Huang <>


  1. cdc3, Syntax: De Cat "French Dislocation"

Message 1: Syntax: De Cat "French Dislocation"

Date: Tue, 18 Mar 2003 09:51:03 +0000
From: cdc3 <>
Subject: Syntax: De Cat "French Dislocation"

Institution: University of York
Program: Department of Language and Linguistic Science
Dissertation Status: Completed
Degree Date: 2003

Author: C�cile, De Cat

Dissertation Title: French Dislocation

Linguistic Field: Syntax, Pragmatics, Language Acquisition

Subject Language: French (code: 1843)

Dissertation Director 1: Bernadette Plunkett

Dissertation Abstract: 

This thesis provides a comprehensive analysis of dislocations in
spoken French. In dislocated constructions, an element appears in the
left- or right-periphery of the sentence and is usually resumed by
another element inside the sentence. Most dislocated elements are
topics. Topics usually encode what the sentence is about:

(1) Tintin, il est connu. 
 Tintin he is famous

The foundations of this research lie in the establishment of solid
diagnostics for dislocated constructions and a detailed description of
the distribution of dislocated elements in a corpus of spontaneous
speech. This required resolving two controversial issues in the study
of spoken French: (i) I argue that subject clitics are arguments in
the most common varieties of spoken French (against the claim that
such elements are agreement morphemes, e.g. Auger 1994); and (ii) I
show that prosodic diagnostics traditionally used in syntactic
analyses to identify left-dislocated elements are partly unfounded.

This thesis develops a theoretical, modular account of dislocated
constructions in spoken French in terms of syntax and pragmatics. I
provide arguments against previous assumptions on Clitic Left
Dislocation (as in (1)) and demonstrate that dislocated elements are
base-generated. The restrictions on the distribution of dislocated
elements are shown to be pragmatic in nature, and the interpretation
of topics not to be exclusively encoded in syntax, contra Rizzi
(1997). Dislocated topics are argued to be adjoined to a
`performative' maximal projection and interpreted as the frame within
which the predication holds true (following Erteschik-Shir 1997).

The picture is completed by investigations into the acquisition of
dislocated structures in child French. These investigations
demonstrate among other things that (i) what have been claimed to be
target-deviant non-nominative subjects (Schutze 1997) are in fact
target-compliant dislocated subjects with a missing resumptive clitic;
(ii) the presence of dislocated topics expressing the subject has no
effect on the realisation of subjects during the null subject stage;
(iii) the pragmatic knowledge required to encode XP topics is
available from the outset of word combinations, hence children display
signs of "pragmatic" competence earlier than previously assumed
(e.g. Wexler 1998).

Ultimately, this research suggests that Information Structure (the
module responsible for topic encoding and decoding) is an integral
part of Universal Grammar and not part of a separate component.
Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue