#4 6 月 29 日（木曜日）
柴谷 方良 氏
ON DATIVE SUBJECT CONSTRUCTIONS
The so-called dative subject constructions, where what appears to be a subject is marked by a dative and other oblique case as in the Latin example, Mihi est liber 'I have a book,' have been a center of focused attention for more than the last two decades, especially among the specialists of South Asian languages, Japanese, Icelandic, Quechua, and others in which a similar type of construction exists. The past analyses assume that the construction-type in question is transitive, or at least at some level of representation as in Relational Grammar, and that the dative-marked experiencer/possessor nominal is the subject of a simplex clause. I claim that these past efforts are misguided - literally misguided by the structure of Modern English, in which the possessors/experiencers of the near-synonymous expressions are encoded as grammatical subjects in the transitive frame. In this paper I endeavor to show that the so-called dative subject constructions are similar in structure and meaning to the double subject construction, instantiated, for example, by the so-called external possessor or possessor ascension construction; e.g. Japanese [Taroo ga [atama ga ookii]] (Taro NOM head NOM big) 'Taro has a big head,' Nepali [ram-ko [taauko dukheko cha]] (Ram-GEN head.NOM hurt be) 'Ram has a headache'). Namely, the constructions in question have a complex structure with two subjects, "large" subject and "small" subject. Semantic motivations for the distributional pattern of subject properties over these two subjects are also explored.