25-MONTH-OLD CHILDREN DO NOT HAVE A GRAMMATICAL CATEGORY OF VERB

This study investigated experimentally the nature and development of c hildren's early productivity with verb-argument structure and verb mor phology. Twenty-two to 25-month-old boys and girls were, in the contex t of playing a game over a several week period, exposed to eight novel verbs modeled with experimentally controlled argument structures and verb inflections. The question was whether, when, and in what ways the children would become productive with these verbs in their spontaneou s speech, going beyond the particular linguistic forms they had heard. In terms of verb-argument structure, the results showed that children most often followed the surface structure of the model, regardless of the argument they were trying to express. Thus, when children had hea rd an argument expressed for a verb, they almost always marked that ar gument correctly in their own utterances; when they had not heard an a rgument expressed for a particular verb, their correct marking dropped to chance levels. The children showed no signs of productive verb mor phology, but they did use the newly learned verbs in some creative way s involving noun-like uses and the appending of locatives. Results are discussed in terms of Tomasello's (1992) Verb island hypothesis.