Can Hamblin's alternative semantics replace unselective binding?
It is well-known that Japanese wh-words have no quantificational force of themselves. Their quantificational force is supplied by the operators such as ka and mo. Dare-ga kimashita-ka. (Who-Nom came-Q: Who came?) Dare-mo-ga kimashita. (Who-MO-Nom came: Everyone came.) They also can be associated with operators crossing complex NP islands. [Dare-ga shidooshita gakusee]-mo shoo-o totta. (Who-Nom taught student-MO prize-Acc got: Every student who was taught by anyone got a prize.) There are two major analyses to account for the long distance association. Nishigauchi in his dissertation proposes the unselective binding approach, and Shimoyama in her dissertation applies Hamblin's alternative semantics. In this paper, we argue for Shimoyama's framework based on the data from Mandarin Chinses.