Volume 35, Issue 2 - Special Issue: Papers in honor of Taro Kageyama: Selected papers from the 10th International Conference on Practical Linguistics of Japanese, Issue Editors: Masahiko Minami and Hisashi Noda

The studies presented in this paper connect the story-related quality and the language-related quality of narrative discourse. The term “coherence” refers to whether or not a text makes sense at a global level, whereas “cohesion” describes the linguistic relationships among clauses in a narrative, such as how the surface linguistic elements of a text are linked to one another at a local level. Using a content-based narrative analysis, a trilogy – a set of three independent but interrelated studies – introduced in this paper quantitatively analyze oral personal narratives through three lenses. As examples of devices for cohesion, the paper qualitatively examines the use of two linguistic devices, tense (past and non-past) and voice (active and passive), and tries to show how narrators deploy organizational strategies in the use of these linguistic forms. The paper (1) examines varied topics in different narrative contexts (genre, topic, oral or written), (2) reveals how both coherence and cohesion serve as the twin engines of narrative, and (3) emphasizes the significance of paying attention not only to the narrative content/structure but also to the appropriate use of linguistic devices so that we can fully grasp language-specific ways of expressing affective elements in narrative.

The present study examines how Japanese language learners use referential expressions in discourse, especially topicalized or non-topicalized subjects, in addition to whether the first language of a Japanese language learner influences the choice of referential expressions. The text of narrative stories, written in both the subject’s first language and second language (i.e. Japanese) by native Chinese speakers and native Korean speakers, as well as text written by Japanese native speakers, were analyzed. As a result, the first language influence and common difficulties were observed in the use of referential expressions by Japanese language learners. Using referential expressions is not simply a matter of negative or positive transfer.

This study investigates address practices among family members in a Japanese TV drama. It qualitatively analyses not only a speaker’s culturally normative use of address terms but also his/her deviations from the norm, including the use of terms that are atypical in Japanese communication. The study sheds light on dynamic aspects of interaction in Japanese in which the speaker creates meaning and constructs fluid relationships with other interlocutors through the use of address terms. The pedagogical implications of these findings are discussed, along with the value of using telecinematic discourse in teaching, to give learners of Japanese a socio-pragmatic understanding of address practices.

This paper examines the topic-specific words extracted from a contact-situation conversation corpus of Japanese. The corpus that was analyzed contains 38 original Skype conversation sessions of 9 pairs of Japanese learners and native speakers on preselected topics. After manually dividing the entire corpus (approximately 200,000 words) into 13 subcorpora by topic, many substantial words and some function words were extracted as topic-specific. Although previous research has suggested that function words are not topic-dependent, this study shows that Japanese function words do, in fact, have a tendency of occurrence. For example, tense and aspect markers occur frequently for the topic “pop-culture,” while nominative markers and existential sentences occur for the topic “town.” This tendency may represent a fundamental resource for developing materials and textbook for both topic-based ones and grammar-structural ones.

This study explored the organization of the multilingual lexicon in L3 learners of Japanese from an L1 Mandarin Chinese background. Using a masked-priming paradigm, it examined whether native-language translations of Japanese words facilitated the recognition of native-morpheme hiragana words and katakana -script loanwords to a similar degree. Participants performed a lexical decision task on a series of hiragana and katakana words, which were preceded by three prime types: noncognate translations, same-script duplicates, and unrelated words. Results showed an equal magnitude of priming from L1 translations for L3 Japanese targets in both scripts, suggesting that conceptual information is made rapidly available for word recognition via an L1 prime. However, priming in the same-script duplicate condition differed numerically between hiragana and katakana , indicating that lower-familiarity katakana loanwords are not activated as rapidly as words in the more-familiar hiragana script. Findings are discussed in relation to models of the multilingual lexicon.