The fundamental difference between child and adult language acquisition: a longitudinal, naturalistic study of parameter resetting in Swedish interlanguage

This study addresses the question of Universal Grammar availability in second language acquisition within the framework of Bley-Vroman’s Fundamental Difference Hypothesis, contributing new empirical evidence from learners of Swedish. Johnson and Newport (1989) showed a strong negative correlation between age of arrival in second language environments and ultimate attainment, supporting the application of the Critical Period Hypothesis of Lenneberg (1967) to second language acquisition. One cogent explanation of this phenomenon offered by Bley-Vroman (1989, 1990) has not yet been adequately tested. Bley-Vroman’s Fundamental Difference Hypothesis attributes the differences between child and adult language acquisition (including uniformity in ultimate attainment, degree of success, and susceptibility to affective factors) to adult learners’ lack of direct access to the learning algorithms of Universal Grammar. But tests of the hypothesis to date have confined themselves to the use of metalinguistic grammaticality judgment tasks (DeKeyser 2000). The present study is an exploratory test of the Fundamental Difference Hypothesis that takes a novel approach; it examines longitudinal, naturalistic data on the acquisition of Swedish syntax by children and adults (n = 9). Subjects comprise native speakers of Iraqi Arabic and Finnish. Each subject was recorded three times over the course of one year during informal speech elicitation interviews. The analysis focuses on the setting of the Null Subject Parameter, the Head Parameter, the V2 or Verb-Second Parameter, and the Nominal Mapping Parameter (Chierchia 1998), as well as three other instantiations of L1-L2 syntactic transfer (null prepositions, null copulas, and resumptive pronouns). Special attention is paid to subjects’ learning curves and to the transfer of native language parameter settings into the target language. Results are found largely to support the Fundamental Difference Hypothesis, with some important exceptions. The dissertation concludes with the proposal that modified replications of this study are necessary in order to determine more conclusively whether Universal Grammar is operative in particular age groups.