Seminars
Thursday 16th March2006 (5.30 pm)
Dr. Celia Jakubowicz
CNRS, LPP, Université Paris
Derivational complexity and “spell out” : Formulation and interpretation of Wh-questions by
typically developing French-speaking children and
children with Specific Language Impairment
Adopting the Minimalist Program (Chomsky 2001, in press) I will propose (a) that target deviant structures used by young typically developing children and by children with SLI are the result of developmental constraints that are external to the computational
component of the Language Faculty and (b) that these constraints are sensitive to the “Computational Complexity” of the derivation (CCd) that is determined by the nature and number of narrow-syntax operations (Jakubowicz & Nash 2001, Jakubowicz 2003, 2004).
I will show that young typically developing children and children with SLI have no difficulties with respect to Agree or Merge (External or Internal Merge), but are penalized by:
(i) the iteration of Merge (merging a n+1 times is more costly than merging a
n times) and
(ii) the number of elements that within the same derivation undergo Merge (merging an+1 is more costly than merging an ).
Both (i) and (ii) predict that shorter derivations are in general less costly and hence more suitable than longer derivations for the interfaces. I will propose that this is particularly the case in the presence of working memory and automatisation limitations that may affect
immature production systems. While for typically developing children these limitations disappear in a regular way through development, for children with SLI they can remain alive for a considerable or indefinite time.
To illustrate these claims I will present data from recent research on question formation by monolingual French-speaking typically developing children (henceforth TD),
children with SLI acquiring French and adults (henceforth Ad). Results from three different tasks will be discussed: An Elicited Production task (EP) of root and long distance questions, an Interpretation task (IN) of long distance ambiguous dative questions and a comprehension task of hidden false-clause questions.
On the one hand, I will show that the EP results do not contradict these predictions.
Namely, results (a) through (e) comply with clause (i) above, while the result indicated in (f)
complies with clause (ii).
(a) In simple root sentences, 3TD and SLI use more in-situ than fronted-wh, particularly for object questions; in the other groups (4TD, 6TD and Ad) fronting is more frequent than in situ. Note that in situ root questions without an echo interpretation are legitimate in French.
(b) Only 3TD, 4TD and SLI use the in situ strategy for long distance questions
(c) Although in the SLI group in root questions wh in situ outranks fronting, in long distance questions a relatively high number of “Simple Partial Movement” questions
(ungrammatical in French) appear, where the wh-word is pronounced in Spec of the intermediate C (see (1))
(1) a. *tu penses quoi que les enfants ils mangent (TD4)
You think what that the children eat?
b. *tu penses quoi _ je mange dans les fêtes? (SLI, 10;1)
You think what I eat in the parties?
c. *tu crois où _ elle a trouvé les images ? (TD4)
You believe where she found the pictures?
d. *tu crois où que j’ai caché le bebé (SLI, 9;4)
You believe where that I hide the baby?
(d) TD as well as SLI children produce other types of “exceptional” long distance questions (copy questions and partial movement questions) previously reported in the literature by Thornton (1990) for English, van Kampen (1997) for Dutch and Gutierrez (2005) for Spanish.
(e) Only Adults produce an important number of long-distance questions.
(f) Only Adults produce Subject- Verb Inversion in root as well as in long distance questions. Note that in French, Subject-Verb Inversion is obligatory for the clitic wh word “Que”; otherwise it is optional.
On the other hand, I will show that although in the EP task TD3, TD4 and children with SLI avoid long distance questions, produce root, (simple) partial movement and copy questions instead and optionally omit the complementizer, these children show preserved capacity for long distance extraction in the IN task and they correctly interpret the hidden false questions in the comprehension task.. Considering these two last facts I will suggest that constraints (i) and (ii) above affect the developing production system more in particular.
References
Chomsky, N. (2001). Derivation by phase. In M. Kenstowicz (ed.), Ken hale: a life in Language, MIT PressChomsky, N. (in press) Three factors in the design of language. Linguistic Inquiry
Fanselow, G. (to appear) "Partial Movement« . SYNCOM (the Syntax Companion), in Everaert, M & Van Riemsdijk (in prep.. The LingComp Foundation)
Gutierrez, J. (to appear).Acquiring Long Distance wh questions in L1 Sapnish: a longitudinal investigation. In Torrens, V. & Escobar, L. (Eds). The acquisition of syntax in Romance languages. Amsterdam: John Benjamins.
Jakubowicz, C. & L. Nash (2001). “Functional Categories and Syntactic Operations in (Ab)normal Language Acquisition”. Brain and Language 77, p. 321-333.
Jakubowicz, C. (2003). “Hypothèses psycholinguistiques sur la nature du déficit dysphasique”. Y. Brun & Ch.L. Gérard (eds.), Les Dysphasies. Paris: Masson.
Jakubowicz, C. (2004b) Production and Interpretation of Long Distance Questions in French SLI: Narrow Syntax l or Interface Deficit? PP presentation, Workshop: From syntax to SLI, University of Massachusetts, Amherst, November 2004.
Thornton, R. (1990). Adventures in Long Distance Moving: The acquisition of Complex Wh- Questions, Phd University of Connecticut
Van der Lely, H.K.J. and Battell J. (2003) “Wh-Movement in children with grammatical SLI: a test for the RDDR hypothesis”, Language, 79/1, 153-1981.
Van Kampen, J. (1997). First steps in wh-movement . Poonsen and Looijen. Wageningene
