- NAJAKS – Nordic Association of Japanese and Korean Studies - http://www.najaks.org -

Analysis of L2 Japanese by Swedish learners: the effect of a study stay in Japan

Posted By admin On May 31, 2013 @ 9:18 am In Abstracts for 2013, Linguistics and Language Teaching | Comments Disabled

Yasuko Nagano-Madsen, Professor

Yuka Okamoto-Gustafsson and Yukiko Shimizu, Lecturers

Department of Languages and Literatures

University of Gothenburg, Sweden


This paper reports the results of an analysis for Swedish learners’ L2 Japanese with focus on the effect of a study stay in Japan. We compared total of nine students at intermediate/advanced level in which four have never studied in Japan, one has studied in Japan for one term, and the rest of four have studied in Japan for an entire year. The data consists of speech where each student explains a chosen theme such as ‘Swedish summer’ and ‘Swedish food’. Each student wrote his own draft with little or no correction by the teacher and made recordings. The average duration of each speech was three minutes.

The resulted speech was analyzed for the following three categories; 1) segmental quality, 2) prosody, and 3) sentence complexity. Segmental quality was evaluated auditorily for the following parameters: vowel quality, vowel quantity, consonant quality, consonant quantity, and prosody.  Three native teachers of Japanese evaluated the segmental quality for each student by giving three level ratings: 1) very unnatural, 2) unnatural, and 3) near perfect. Prosody was evaluated according to the generative phonological model of intonation proposed for Japanese (Pierrehumbert and Beckman 1988). Finally, sentence complexity was evaluated by using three parameters: 1) rate of complex sentences per speech, 2) the use of adnominal clauses (補足節、名詞修飾節、引用節), and 3) the use of connective clauses (連用修飾節).

The effect of the study stay in Japan was seen most consistently in the production of prosody while sentence complexity did not clearly demonstrate the effect of a study stay in Japan. This may be due to the type of data collected in the current study since a written draft was prepared first. Segmental qualities improved in general for the group who studied in Japan but the effect was more random compared to the prosodic development.

A typical intonation produced by the students with no experience in Japan was that of flat intonation where only the phrase boundary is marked by pause. The development of intonation starts by marking the phrase initial position by high pitch (%L, H-) , which is similar to the onset of accentual phrase in Japanese. This is followed by a lowering of phrase final by lower pitch (L%).  This is followed by the acquisition of lexical pitch accent (H*+L). However, the exact phonetic manifestation of the H*+L accent was not acquired by any of the students. The discourse parameter to reflect information structure such as focus appeared only in one student.

The results have lead us to hypothesize the followings: 1) sentence complexity may not be the best parameter in examining the communicative ability. Instead, such elements relevant to ‘information structure’ may be more suitable for speech communication. It includes the use of topic and contrastive particles wa and ga, ellipsis, word order, among others. 2) The generative model of intonation allows us to model the process of prosodic acquisition accurately. 3) Swedish students’ interlanguage concerning prosody takes as its first step a flat pitch pattern. This is followed by phrase marking by initial pitch rise and final pitch fall. The acquisition of lexical pitch accent comes somewhat later and it is difficult to produce it at the right place with right phonetic manifestation. 4) Discourse elements such as topic, contrast, and focus appear also at the latter stage.


Article printed from NAJAKS – Nordic Association of Japanese and Korean Studies: http://www.najaks.org

URL to article: http://www.najaks.org/?p=1060