What's left in early L2 architecture

• Published in: SECOND LANGUAGE Vol. 5; no. May; pp. 3 - 26
• Main Author: SCHWARTZ, Bonnie D.
• Format: Journal Article
• Language: English
• Published: The Japan Second Language Association 01.05.2006
• Subjects: Functional Heads; Interlanguage; Linguistic Interference; Native Language; Second Language Learning; Second Language Learning Theories; Transfer
• ISSN: 1347-278X; 2187-0047
• DOI: 10.11431/secondlanguage2002.5.0_3

Evidence presented by A. Vainikka & M. Young-Scholten (1994) & by R. M. Bhatt & B. Hancin-Bhatt (2002) for the hypothesized absence of the complementizer projection (CP) domain in early stages of interlanguage is found wanting, as the former consists of the absence of specific left-peripheral structures & the latter depends on a criterion of target-like performance. Neither of these types of evidence contradicts Schwartz & R. A. Sprouse's (1996) full transfer/full access hypothesis of second-language (L2) acquisition, which holds that the entire native grammar constitutes the initial state of an interlanguage & is supported by results of three studies that attest the native CP domain in early interlanguage: (1) U. Bohnacker's finding that native speakers of Swedish, a verb-second (V2) language, who do not know English immediately converge on the V2 property of L2 German whereas those who already know English allow alternative word orders; (2) T. Gruter & S. Conradie's (2004) evidence that the syntactic ambiguity of German wh-questions is immediately perceived by Afrikaans-speaking learners as opposed to English-speaking learners; & (3) A. Hulk's (1991) demonstration that Dutch speakers accept French word orders robustly only if they are also possible in Dutch. Tables, References. J. Hitchcock