| Abstract: | Past research shows that young language learners know something about the different category organizations of animals, objects and substances. The three experiments reported here compare Japanese‐speaking and English‐speaking children's novel name generalizations for two kinds of objects: clear instances of artifacts and objects with ambiguous features suggestive of animates. This comparison was motivated by the very different nature of individuation in the two languages and by the boundary shift hypothesis that proposes that entities that straddle the individuation boundary of a language are assimilated toward the individuated side. The results of the three experiments support the hypothesis. An explanation in terms of mutually reinforcing correlations among language, perceptual properties and category structure is proposed. |
