A Subset Analysis for Stereotypical Expressions in Japanese

In this paper, we will discuss the semantics of stereotypical expressions in Japanese focusing on the adjectival suffix -rashii, which takes an NP and forms an adjective. We will propose that -rashii involves the process of subset taking, which intuitively corresponds to strengthening. Stereotypicality obtains by taking a subset of all properties associated with an individual. Furthermore, it will be argued that the same semantics can be applied to other stereotypical expressions in Japanese such as Contruction Reduplication (X X si-ta Y) and X in X constructions (X-no naka-no X), which involve the process more explicitly as reduplication. Not only is our semantics motivated by overt counterparts, it also can explain wider range of data than previous studies such as McCready and Ogata (2007). Specifically, the obviation of the P(x) presupposition, in which the modified NP is in the extension of the modifier NP, is correctly predicted under our analysis. Finally, we touch upon Contrastive Focus Reduplication in English, making a speculation on the relation between subsets and stereotypicality.