@article{oai:repository.ninjal.ac.jp:00002022,
 author = {小林, 英樹 and KOBAYASHI, Hideki},
 journal = {日本語科学, Japanese Linguistics},
 month = {Apr},
 note = {群馬大学, Gunma University, 項関係の複合語は,これまで項構造レベルで分析されることが多かったが,項構造レベルでは上手く分析できない項関係の複合語が存在する。(a)企業を意識改革していこう。(b)*大規模小売店法を法律改正する。(c)日産がアメリカに工場進出した。(d)*日産がアメリカに工場を進出した。(e)日産の工場がアメリカに進出した。非主要部と主要部が項関係で複合しているにもかかわらず,目的語が統語的に現れる(a)の「意識改革(する)」,非主要部が非能格自動詞の主語であるように一見みえる(c)の「工場進出(する)」は,項構造レベルでは上手く分析できない。本稿は,項構造レベルでは上手く分析できない「意識改革(する)」や「工場進出(する)」を語彙概念構造レベルで分析し,すべての項関係の複合語に統一的な説明を与えることができる主要部位置への名詞概念の代入を提案する。, Argument-relation compounds have been analyzed on the level of Argument Structure. In this paper we give examples, such as ishiki-kaikaku (sense-improve) and kojo-shinshutsu (factory-expand), which cannot be analyzed on the level of Argument Structure. Ishiki-kaikaku takes an object syntactically. If we analyze ishiki-kaikaku on the level of Argument Structure, we cannot answer why ishiki-kaikaku takes an object syntactically, because argument-relation compounds usually cannot take an object syntactically.  Kojo-shinshutsu is also one of argument-relation compounds which cannot be analyzed on the level of Argument Structure. Shinshutsu is unergative, not transitive, so we lead to think of kojo as the subject of shinshutsu, but compounding with the subject of unergative verb is prohibited by First Sister Principle, which says verbal compounds are formed by incorporation of a word in first sister position of the verb. If we analyze kojo-shinshutsu on the level of Argument Structure, we cannot answer what kojo of kojo-shinshutsu is. We analyze ishiki-kaikaku and kojo-shinshutsu on the level of Lexical Conceptual Structure (LCS), and propose an insertion of a nominal concept into a head position in LCS, by which we can explain all argument-relation compounds, including ishiki-kaikaku and kojo-shinshutsu., application/pdf},
 pages = {7--25},
 title = {語彙概念構造レベルでの複合},
 volume = {5},
 year = {1999},
 yomi = {コバヤシ, ヒデキ}
}