Argument realization in child Inuktitut
One of the characteristic features of child language is its high frequency of argument omission. To explain this phenomenon, various proposals have been offered, including grammatical, performance and discourse-pragmatic accounts. Recently, the discourse-pragmatic approach has been shown to successfully predict the form of the argument based on its informativeness features (Clancy 1993, 1997; Allen 2000). The results of a study on argument realization in child Inuktitut suggest that informativeness features, such as the presence or absence of a referent in the physical context or its discourse status (i.e., whether it is discourse new or discourse old), have a strong effect on the form in which an argument appears: the more informative the argument is, the more likely it is to be expressed by an overt argument. Conversely, the less informative an argument is, the more likely it is to be expressed by a null argument (Allen 2000). However, this account fails to explain the instances of null arguments that are informative and overt arguments that are uninformative.
To account for these contradictory examples, I extend the discourse-pragmatic view to include a social-cognitive component that has not been systematically explored in this context previously, joint attention, a triadic social activity wherein the speaker and the interlocutor are both focused on the same referent while aware of each other’s attention (Tomasello 1999). Specifically, I will show that the production of null informative arguments can be explained by the presence of joint attention, whereas the production of overt uninformative arguments can be explained by the absence of joint attention. I will review the results of a study on spontaneous speech data from four monolingual children acquiring Inuktitut (2;0-3;6) videotaped in naturalistic communication situations. I will argue that argument form choices in child language reflect not only the child’s knowledge of syntax and evaluation of the discourse context, but crucially, they require and are determined by the speaker’s ability to assess the knowledge state of the interlocutor.