DOES THE NOUN PHRASE ACCESSIBILITY HIERARCHY PREDICT THE DIFFICULTY ORDER IN THE ACQUISITION OF JAPANESE RELATIVE CLAUSES?
Yasuhiro Shirai,Hiromi Ozeki
DOI: https://doi.org/10.1017/S0272263107070106
IF: 4.73
2007-04-25
Studies in Second Language Acquisition
Abstract:Although Keenan and Comrie's (1977) noun phrase accessibility hierarchy (NPAH) has been shown to predict the difficulty order of relative clauses (RCs) in SLA, most studies of the NPAH have been on European languages. This paper tests the prediction for Japanese. Study 1 analyzes RCs in an oral interview corpus from 90 learners of Japanese at four different levels of proficiency (first language = Mandarin Chinese, English, and Korean; N = 30 for each). Analysis of 1005 RCs from nonnative data and 231 RCs from 15 native speakers (NSs) of Japanese revealed that even lower proficiency learners used direct object (DO) and oblique (OBL) relatives, suggesting that subject (SU) relatives are not easier than DO or OBL relatives for second language learners of Japanese. The learners (except Korean NSs) also made strong associations between SU and animate heads and between DO/OBL and inanimate heads. Study 2 employed a sentence-combining experiment. Fifty NSs of Cantonese studying Japanese in Hong Kong took the test, which controlled for the animacy of head noun phrases and arguments of the verbs. Results revealed no significant difference between SU and DO, which were both easier than OBL, with only a minimal effect of animacy. However, errors of converting DO and OBL target items into SU relatives almost exclusively involved animate-head items. The results suggest that the NPAH does not predict the difficulty order of Japanese RCs, and that learners use different types of RCs based on the animacy of the head noun.This paper is based on a paper presented at the Workshop on the L2 Acquisition of Relative Clauses (January 28, 2005, Cornell University). We thank the participants at the workshop (in particular, Stephen Matthews and John Whitman), Kevin Gregg, Zoe Luk, William O'Grady, Wataru Suzuki, and the two anonymous SSLA reviewers for their invaluable comments and discussions. Study 1 is part of the first author's doctoral dissertation, submitted to Ochanomizu University, and has appeared in Japanese (Ozeki, 2005b). Study 2 was conducted at the Chinese University of Hong Kong, supported by a Direct Grant for Research from the university. We thank Alexis Chu, Yumi Inoue, Zoe Luk, and Chi-Ming Ho for their assistance in data collection and/or the construction of the test material and the statistical consultants at Cornell University (in particular, Freedom King) for their assistance.
Psychology,Linguistics