ABSTRACT
One of the main goals of the study of second language acquisition is to provide an explanation of the different outcomes of language learning in adulthood and in childhood. One crucial problem involves ascertaining how much of the observed differences depend on the nature of the grammatical representation of non-native speakers and how much they depend on performance factors. Generative work has favored performance-based arguments for nonconvergence, primarily in the domain of morphology. In 2003, Montrul and Slabakova's attainment study included a truth-value judgment task investigating habitual and one-time events. The two approaches reviewed lead to different developmental perspectives. In the morphosyntactic approach, interpretable features enter into the syntactic computation, determining the aspectual interpretation. In the selectional approach, interpretation is processed compositionally by the elements involved. Schmitt's results suggest that a system that is rigidly morphology based will not suffice to describe observed language variation.
