Tense, aspect, and event representations in English and Japanese.

摘要

This study attempts to defend a theory of temporal reference that unifies the notions of tense and aspect, and apply it to the tense and aspect systems of English and Japanese. It argues that adopting Klein's (1994) system of temporal reference allows for the crosslinguistic differences between the English and Japanese tense and aspect systems to be accounted for in a simple manner with no extra stipulations.;The present study shows that the complications with the interpretation of tenses in different types of embedded clauses can be ascribed to the presence or absence of the intensional context, a shift of the deictic center for the tense interpretations, and the aspect of the predicates that appear with the tense morphemes. As a result, the semantics of tenses can be maintained consistent across different constructions and the tense system can be much simpler than those that have been proposed by some existing theories of tense. Both the English past and the Japanese -ta encode TT TU as a tense marker, while both the English present and the Japanese -ru encode TT ⊃ TU for the present tense, though the Japanese - ru as a non-past tense marker additionally encodes TU TT. The seeming differences between the English and the Japanese tense systems are attributed to the fact that Japanese employs the relative tense system on top of the absolute tense system. The relative tense is obligatorily used in the verb-complement structure in Japanese, while it is optional in toki-clauses and relative clauses. This is because only complement clauses may involve the indirect speech context. The presence of SOT in English and its absence in Japanese in the verb-complement constructions can be explained by the fact that the complement past in Japanese is always evaluated with respect to the matrix past, and therefore, may not be able to overlap with the matrix event. Thus, there is no need to posit a SOT rule for English nor different semantics for the past tenses in English and Japanese.;This study also accounts for the puzzling behavior of the Japanese Verb-te-iru form, which can have both a perfect and a progressive interpretation. While the English present participial encodes TT ⊂ TSit and the English past participial encodes TSit TT, -te of -te-iru simply introduces a topic time without specifying how it relates to the situation time. This allows the TT of the sentences with -te-iru to be placed either within the situation time or in the post time of the situation, depending on the context. The existential meaning of the present progressive and the present perfect of English and the Japanese -te-iru is provided by their auxiliary verbs. The tense components of the auxiliary verbs of these aspectual forms encode TT ⊃ TU, and provide the current relevance of the described eventuality. The study shows that Klein's system clarifies the respective contributions of tense and aspect components of these complex temporal morphemes in English and Japanese in a simple manner without extra stipulations.