[lingtalks] Bonnie Schwartz to speak at USC Monday

[Karma Dolma]

Monday, February 11, 2008  USC Linguistics Colloquium Series presents:

Bonnie D. Schwartz (work with Hyang Suk Song)
University of Hawai'i

3:00 pm, GFS 118.  Dinner to follow at 5:00 in GFS 330.


On the unfolding of nonnative language in children and adults

Research that systematically compares child L1, child L2 and adult L2 development has significant potential to probe fundamental similarities/differences in grammar creation capacities across these three populations (Schwartz 2004; Unsworth 2005).  Focusing on both development and convergence, this presentation reports on one such comparison, exploring the acquisition of Korean wh constructions.

While scrambling in Korean (an SOV wh in situ language) is generally optional, in the context of negative questions with a Negative Polarity Item (NPI, e.g. amwuto 'anyone'), (i) object wh phrases must scramble on the wh question reading (an Intervention Effect, Beck & Kim 1997), as in (1), and (ii) the non scrambled order has a yes/no question reading (exclusively), as in (2). These properties of Korean wh constructions with NPIs constitute poverty of the stimulus problems for (L1 children and) English natives acquiring Korean: This knowledge cannot transfer from English (a non wh in situ, non scrambling language), cannot be induced from Korean input, and is not the subject of L2 instruction.

(1)    a. *Amwuto    mwues ul    sa ci    anh ass ni?    SOV (Non scrambled)
        Anyone    what ACC    buy ci    NEG PAST Q
        (cannot mean 'What did no one buy?')

    b.    Mwues ul    amwuto    sa ci    anh ass ni?    OSV (Scrambled)
        What ACC    anyone    buy ci    NEG PAST Q
        'What did no one buy?'

(2)    Amwuto    mwues ul    sa ci    anh ass ni?    SOV (Non scrambled)
Anyone    something ACC    buy ci    NEG PAST Q
'Did no one buy something?'

Our study investigates how development in L1 Korean children (n=23, aged 5–7), L1 English child L2ers (n=10, Korean onset age=4–7) and L1 English adult L2ers (n=15) unfolds for these properties: (i) that scrambling, as in (1b), is required to obviate the NPI context Intervention Effect in wh questions; and (ii) that interpretation of wh phrases in non scrambled variants of these NPI context questions is restricted to wh indefinites (e.g. Nishigauchi 1990), as in (2).

Three experiments (elicited production; acceptability judgment; interpretation verification) test the 3 groups as well as adult natives (n=15). The results show that child L2ers and adult L2ers, independently assessed for Korean proficiency, follow the same route to convergence, including overcoming the poverty of the stimulus problems, a route differing from––yet subsuming––the L1 child route.

References
Beck, S. & S. Kim. 1997. On wh  and operator scope in Korean. Journal of East Asian Languages 6:339 84.
Nishigauchi, T. 1990. Quantification in the Theory of Grammar. Dordrecht: Kluwer.
Schwartz, B.D. 2004. Why child L2 acquisition? In J. van Kampen & S. Baauw, eds., Proceedings of GALA 2003, Vol. 1. Utrecht: Netherlands Graduate School of Linguistics (LOT), pp. 47 66.
Unsworth, S. 2005. Child L2, Adult L2, Child L1: Differences and Similarities: A Study on the Acquisition of Direct Object Scrambling in Dutch. PhD dissertation, Utrecht University.





-- 
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Michal Temkin Martinez
Linguistics Department
University of Southern California
www-scf.usc.edu/~mtmartin