[Japanese | English]
Temporal interpretation of nouns has been studied by Enç (1981, 1986), Musan (1995, 1999), and Tonhauser (2002, 2006), among other researchers. Enç’s proposes that common nouns have temporal arguments and that they are time variables. As variables, they may be bound or free. But unless the traditional operator-based analysis of tense morphemes, Enç’s proposal allows the temporal interpretation of nouns to be a time other than the utterance time or the time of the main verb. If the temporal interpretation of nouns is determined by the value assumed by their temporal arguments, the time could be any time whatsoever. This idea explains the most natural interpretation of Every hostage will be at the party. But to say that the time argument for a noun is a variable is too liberal, and Musan and Tonhauser try to constrain the possible readings for nouns. Preston (2013) argues that the proposals offered by Musan and Tonhauser are not constrained enough and shows that there are some correlations between the main predicate tense and the interpretation of (subject) nouns.
In this presentation, I will present some additional complications and offer one possible solution. When a “strong” determiner such as every and most and an event-oriented common noun such as spectator and passenger occur in the subject position of a sentence, the quantifier must in question must quantify over individual-interval pairs and the intervals in question must be maximal ones. We could alternatively use events, but the same maximality requirement is also imposed. I believe that the stipulation about maximality is needed because common nouns behave like stative predicates when used predicatively and they should have the subinterval property.
(PDF version with references is available.)
In at least three environments -- de se binding, distributivity, and focus quantification -- the presuppositions of bound pronouns are known to not contribute to meaning in the expected fashion. More recently several cases of other presuppositions exhibiting the same behavior have been found. In this talk, I describe a general approach to capture the interaction of presuppositions. Then I discuss how the non-bound reading of the reflexives in (i) can be captured within the approach, as well as other Condition A phenomena.
(i) Only Romney expected himself to win.
Semantics Research Group
Last modified: 2013-07-04 17:25:48 JST