Retour vers liste

Détail de la contribution

Auteur: Yasuyuki FUKUTOMI

Light Noun-in-situ in Japanese

Abstract/Résumé: The purpose of this presentation is to point out less-studied parallelisms and differences between English and Japanese indefinite pronouns. Kishimoto (2000) proposes a bimorphemic analysis of indefinite pronouns in English in order to explain the possibility of post-nominal attributive adjectival modification. According to his analysis, 'some' and 'thing' are introduced in the syntactic derivation as two independent lexical items, which are later merged into a single word in the post-syntactic morphological component. In the expression 'something delicious' the surface order obtains as the result of the overt N-raising of 'thing' across the adjective to the number head. With this analysis in mind, let us consider the corresponding expressions in Japanese. Like in English, attributive adjectives in Japanese are admitted only to the left of NP, not to the right. Significantly, however, Japanese indefinite pronouns in themselves cannot be modified pre-nominally or post-nominally. To obtain a grammatical expression, light noun such as 'mono' must occur in the position that follows an adjective, as in 'nani-ka oisii mono'. I will propose that what undergoes an overt movement in English stays in its base head position in Japanese like the contrast between English and Japanese wh-movements. The availability of overt N-raising can be attributed to the presence or absence of a number-related feature. Since Japanese nominals display the number morphology with a numeral and a classifier, the light noun itself does not move overtly to the number head, remaining in-situ. The following internal structure of Japanese nominals is assumed. Notice that wh-phrase with an indefinite marker 'nani-ka' cannot appear between the light noun and the case particle, as illustrated by the contrast between 'nani-ka oisii mono-o' and '*oisii mono nani-ka-o'. Following Watanabe's (2006) proposal, I assume that Japanese nominals have at least two functional layers above NP; CaseP and QP, that NP moves to the specifier position of CaseP, and that the wh-phrase with an indefinite marker itself is base-generated in the specifier position of QP. The current analysis implies that the bare wh-phrase with an indefinite marker is not of the category [+N] but a QP with a null head noun. Corroborating evidence comes from the fact that bare indefinite pronouns are not necessarily case-marked. Furthermore the adverbial nature allows them to co-occur with a referential noun phrase instead of a light noun. In summary, to the extent that the present analysis holds, the Japanese data provide further support for Kishimoto's bimorphemic analysis of indefinite pronouns and for the possibility or impossibility of overt N-raising in English and Japanese, respectively.