This article examines the Root Infinitive (RI) effect in consecutive child L2 acquisition of English. The data are drawn from a ten-month longitudinal study of three children (ages 6;4 - 8;1), of which two speak Russian and one speaks Japanese as L1. It is argued that the current theories of RIs (e.g., the Truncation Hypothesis of Rizzi (1993/94) and the Morphological Deficit Hypothesis (Haznedar and Schwartz 1997, Prévost and White 2000), do not account for the fact that finite root predicates predominantly consist of punctual eventives and statives, whereas non-finite predicates consist of non-punctual eventives in the L2 grammars of all three children. It is also argued that the Aspect-before-Tense Hypothesis (Antinucci and Miller 1976) does not explain the correlations between finiteness (or non-finiteness) and the aspectual characteristics of root predicates. The article proposes a new approach to the RI effect, arguing that the distribution of finite and non-finite predicates follows from the telicity semantics of a given verb. The core of the proposal is that early child L2 syntax is underspecified for syntactic telicity features, which are argued to reside in the AspP projections, as in Borer (1994).
Proceedings of the 6th Generative Approaches to Second Language Acquisition Conference (GASLA 2002): L2 Links
edited by Juana M. Liceras, Helmut Zobl, and Helen Goodluck
Table of contents