[lingtalks] Linguistics Colloquium 11/21

[Katie McGee]

I am pleased to announce the speakers for our next colloquium on Monday, November 21st!

Date: 11/21
Time: 2-3:30
Place: 003 CSB

Speakers:

1) Gabriela Simon-Cereijido,  Joint Doctoral Program in Language and Communicative Disorders.
2) Ezra Van Everbroeck, Department of Linguistics.

Abstracts are given below.

Gabriela Simon-Cereijido

Spanish monolingual and bilingual speakers with language impairment (LI)
have shown deficits in clitic production in spontaneous language and
elicited tasks However, it is unknown whether these difficulties relate to
general problems of clitic knowledge or deficits with the coordination of
specific grammatical and pragmatic constraints. Clitic doubling (CD) and
clitic climbing (CC) are two clitic-verb constructions that demand
increased grammatical and pragmatic proficiency.  This study aimed to
determine whether children with LI have significantly more difficulties
with CD and CC compared to their peers. Thirty typically developing
children and 30 children with LI matched by age (mean: 64 months) produced
narratives based on wordless picture books, which were coded for CD,
infinitival clauses and CC.  The findings indicate that children with LI
use CC and CD less frequently than their peers. However, only CD presented
significant difficulty.  The observed problems with clitics could not be
attributed to insufficient knowledge of the clitic system but to a limited
capacity to coordinate grammatical and pragmatic constraints in complex
contexts.



Ezra Van Everbroeck

       SVO LANGUAGES AND PRO-DROP: COMPUTATIONAL MODELING

  I present data from several computational experiments which
  investigate the impact of null subjects on the learnability of SVO
  languages. The simulations varied the degree of morphological
  marking on nouns and verbs.

  The network results show that the effect of pro-drop on language
  learnability is very limited, at least as long as some morphological
  marking is present. Contrary to theoretical assumptions, rich
  agreement did not play a greater role in pro-drop languages than
  case marking or verbal tense marking. In the absence of morphology,
  however, pro-drop leads to severe learnability problems for the
  networks, which suggests that such a language type should be
  unattested among natural languages.

  I discuss two groups of natural languages which seem likely to
  produce counter-examples: i.e. Romance-based creoles, and the
  isolating languages of South-East Asia (exemplified by Mandarin
  Chinese). A closer look reveals that the former have lost the
  pro-drop phenomena found in Spanish and Portuguese. Mandarin has
  clear pro-drop but also a number of linguistic features which
  compensate for it. Tellingly, many of these features are acquired
  early by children learning Mandarin.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://ling.ucsd.edu/pipermail/lingtalks/attachments/20051118/178c72fd/attachment-0001.htm