(unknown)

[arobur at CLUB-INTERNET.FR]

Le Laboratoire de Linguistique Formelle (LLF) et l'UFR Sciences du 
Langage de
l'Université Paris-Diderot vous invitent à une conférence du
Pr Rose-Marie Déchaine, University of British Columbia (Vancouver),
le lundi 6 avril 2009,
à 16h30.
UFR de Linguistique
30 rue du Château des Rentiers
75013. Paris
salle 124 ou 134.

Rose-Marie Déchaine (UBC) & Mireille Tremblay (U. de Montréal)
Presented by R.-M. Déchaine

Title: What should a theory of (syntactic) categorization account for?

Abstract:
We propose that the ontology of categorization reflects the logic of
individuation (requiring differentiation, discrimination and recognition) 
and set-formation (requiring identification,classification and labeling). We 
argue that categorization has specific correlates, as follows.
1. The ontogeny of categorization reflects a division between open- and
closed-class items which is the basis of proto-grammar.
2. The phylogeny of categorization reflects the distinction between 
Faculty of Language in the Broad Sense (FLB) versus Narrow Sense 
(FLN) (Hauser et al.(2002). Auditory categorical perception and 
vocabulary size instantiate FLB. Class partition arises via recursive 
application of Merge and instantiates FLN.
This clarifies the “logical problem of language evolution” (Christiensen &
Chater (2008): FLB reflects general-learning mechanisms; only FLN is
language-specific, reducing to recursion.)
3. The neurological basis of categorization with respect to neural 
information pathways is likely mediated by neural activity in the dorsal 
what stream and the ventral where stream. We suggest that 
object-recognition in the dorsal where stream corresponds to the 
cognitive act of individuation; syntactically, this yields an argument (DP) 
expression. Property recognition in the ventral what stream corresponds 
to the cognitive act of set-formation; syntactically, this yields a predicate 
expression.
4. Changes in categorization over time (diachrony) arise via the 
successor function in two ways: addition and subtraction. For Lexical 
categories this corresponds to vocabulary expansion or contraction. For 
Functional categories, this corresponds to sub-class addition(introduction 
of a novel Functional category), sub-class subtraction (elimination of a 
Functional category).
5. Typologically, all languages have a class partition between open- and
closed-class items (Gil 2005); so the lower bound on the number of 
word-classes is two. We show that although class partition is constrained 
by Merge, there is no upper bound to the number of word-classes.
6. Syntactic categorization displays cross-modal parallelism: the 
constraints that hold of the ontology, ontogeny, phylogeny, neurology, 
diachrony, and typology of syntactic categorization in spoken languages 
also holds of signed languages.




-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/parislinguists/attachments/20090330/252f8117/attachment.htm>