Festschrifts

Two important claims about modals have been proposed in the past few decades. The first claim is that modals have uniform lexical semantics, but are interpreted as epistemic when located structurally above temporal elements and as root when located below them. The second claim is that, across languages, items that express root ability have episodic, non-ability-attributing interpretations in addition to generic, ability-attributing interpretations. With these two claims as background, this study examines verbal expressions of modality in Japanese. The study shows that (i) there are only three verbal expressions of modality in Japanese: the ability suffix -e/rare, the possibility suffix -e/uru, and the negative possibility suffix -kane; (ii) while these three expressions’ syntactic positions with respect to tense do not contradict the first claim, they do not provide strong support for it either; and (iii) Japanese sentences with the ability suffix -e/rare can have an episodic, non-ability-attributing interpretation and a generic, ability-attributing interpretation, providing novel evidence for the second claim. What makes the ability suffix -e/rare different from the ability expressions analyzed in previous studies is that the availability of the two interpretations with -e/rare is tied to two different syntactic structures it occurs in: while sentences with -e/rare in a dative-nominative structure only have the generic, ability-attributing interpretation, similar sentences in a nominative-accusative structure only have the episodic, non-ability-attributing interpretation. It is argued that these two alternative case-alignments are reflections of two different underlying structures of complements of -e/rare, making -e/rare a unique case of an ability expression with syntactic reflexes of episodic-generic ambiguity.