@article{oai:stars.repo.nii.ac.jp:00007340,
 author = {有川, 康二 and Arikawa, Koji},
 issue = {13},
 journal = {桃山学院大学人間科学, HUMAN SCIENCES REVIEW, St. Andrew's University},
 month = {Sep},
 note = {The central issue of this study is the semantics and the syntax of temporal interpretations of the clausal complement of a verb. Based mainly on English examples, theories of tense hitherto have distinguished three types of the complement-clause-tense interpretation: simultaneous, past-shifted, and double-access. In the simultaneous construal, the denotation of embedded clause tense (∥Te∥) overlaps with the denotation of matrix clause tense (∥Tm∥). In past-shifted, ∥Te∥ precedes ∥Tm∥. In double-access, ∥Te∥ overlaps with both ∥Tm∥ and the utterance time (∥UT∥). A consideration of Japanese examples adds two more: past-double-access and triple-access. Past-double-access is a combination of simultaneous and past-shifted, i.e. ∥Te∥ includes both ∥Tm∥ and some prior time (∥PT∥). Triple-access is a combination of past-shifted and double-access, i.e. ∥Te∥ includes ∥PT∥, ∥Tm∥, and ∥UT∥. This article will attempt to answer the following questions: (i) Why is past-shifted more marked than simultaneous in English?; (ii) Why is past-shifted natural in Japanese?; (iii) How are the multiple-accesses explained?; (iv) Why does English lack past-double access and triple-access?; (v) Why is simultaneous possible in Japanese only when the embedded clause tense is nonpast?; and (vi) Why does stativity matter with simultaneous and multiple-accesses in both languages? We adopt a hypothesis in which temporal-argument phrases of the functional category T^0 interact with various principles and constraints of the generative grammar that have been found to motivate independent phenomena other than tense. [Areas of interest: semantics, syntax], 4, KJ00004276321, 論文, Article},
 pages = {45--84},
 title = {Tense, Subject, and Derivation},
 year = {1997},
 yomi = {アリカワ, コウジ}
}