This study focuses on Japanese and Chinese languages with regards to the possibility of omitting demonstrative adjectives. Iori (2007) points out that if a language has definite articles, such as English, the second time a noun appears in the context, some kind of marker is needed in front of the noun. However, if a language doesn't have definite articles, such as Japanese, the marker in front of the noun is not necessary. Chinese is similar to Japanese in this respect, also not needing definite articles before nouns. If Iori (2007) is right, we can predict that the demonstrative adjectives can also be omitted in Chinese. On the other hand, Si (2008) points out, it is not possible for demonstrative adjectives to be omitted in Chinese. The aim of this study is to determine if demonstrative adjectives in Chinese can be omitted. If the omission is possible what kind of situation occurs? Is there any difference between the Japanese and Chinese languages? As a result of the analysis, the following things became clear. (1) Because the omission of the demonstrative adjective was allowed in certain conditions in Chinese, the generalization of Iori (2007) is partially supported. (2) Both Japanese and Chinese languages tend to have a generic interpretation when an abstract noun appears in the context. The omission of the demonstrative adjective is usually not allowed with abstract nouns. The difference is, in the case of a common noun, the demonstrative adjective of Chinese can only be omitted when the antecedent is not limited by a singular quantifier such as "one".

This study investigates how advanced learners of the Japanese language also to be referred to as Chinese-speaking learner (CH) of Japanese, English-speaking learner (EN) of Japanese and native speakers (NSs) of Japanese manage the use of the referent markers in order to demonstrate topic continuity in Japanese oral narratives. Narrative discourse was produced by two groups of learners of Japanese, Chinese whose native language is typologically parallel to and English speakers whose native language is distant from Japanese.

The results showed that (1) both CHs and ENs used the indefinite marking (-ga) to encode referent introduction, as a target-like use. However some CHs and ENs still have difficulty using the indefinite marking (-ga) for this situation because they don't have such rules in their native languages. (2) ENs and NSs use zero anaphora to encode re-introduction of referents. In contrast, CHs tend to use full noun form followed by "wa". CHs attempted to avoid ambiguity and used full noun with "wa" when referents are re-established as the "on-going" topic of the discourse. (3) Both CHs and ENs performed zero anaphora as a target-like use of re-introduction of referents and referent maintenance even though zero anaphora is not the case in English.