意味論研究会2015年8月31日

In this presentation, I will entertain the hypothesis that in order to properly evaluate antecedents of counterfactual conditionals, we should reconsider the received assumption that the domain of quantification for the counterfactual operator would consists of (a small subset of) the worlds in which the antecedent is true. Facing with complex cases like “If I were you, …” “If Kazuo were a Caucasian”, “If John had given flowers to Mary tomorrow, …” we can obtain correct truth conditions by replacing an entity mentioned in the antecedent with a different entity that shares the same set of properties relevant for evaluating the counterfactual. Indeed, when we say “If I were you, …” we could paraphrase it into “If I were in your position, …” Japanese counterfactuals of the same type allow for exactly the same paraphrases. We can then find relevant worlds where the antecedent with some modification is true and this set is used as the domain. This account is not based on counterparts of individuals because the relevant individuals are not like each other other than the relevant properties. The proposal presents an antithesis to the proposals made by Ippolito (2003, 2006) but also covers data not having with tense.