Dissertation Defense: Mariko Uno
Candidate Name: Mariko Uno
Dissertation Advisor: Lourdes Ortega, Ph.D.
Title: Developing Question Constructions in Japanese as A First Language: The Roles of Discourse and Parental Input
Previous studies have shown that children start asking questions when they are as young as 2. Although many studies including those analyzing Japanese, investigated children’s use of questions, (1) they tended to focus on only one or two question constructions and there is a need to widen the scope of investigation; (2) methodological framework inspired by the Usage-based theory of discourse needs to be called for; and (3) studies investigating questions need to consider the asymmetrical relationship between children and the parents. The present study addressed the three issues by analyzing longitudinal data of the three Japanese children.
The present dissertation extracted 17,291 questions from the three Japanese children Aki, Ryo, and Tai and their mothers. The children’s age ranged between 1;3 to 3;0. Those data are available in the CHILDES database (MacWhinney, 2000; Oshima-Takane & MacWhinney, 1998). Their question constructions were coded for six different types: (1) yes/no questions including a sentence-final question marker (ka?, no?, kanaa?); (2) wh-questions; (3) wa-ending questions (4) one-word or phrase particle questions; (5) one-word or phrase questions without any overt lexicogrammatical marking; and (6) multi-word utterances without any overt lexicogrammatical question marking. By drawing on the Usage-based theories of discourse, each question was coded whether the referent carried given or new discourse information and whether it was present or absent at the time of speaking.
The results of the analysis of 17,291 questions showed that both children and their parents dominantly asked questions about referents that carried given information and were also present. Children tended to use one-word or phrase questions without any overt lexicogrammatical marking (Type 5) initially for those referents and later expanded their range of questions using more than one word or phrase to ask questions about those referents. Tai’s mother used the type 5 for those referents too, but Aki and Ryo’s mother used type 1,2,3,&6, and did not use the type 5 particularly frequently. Both children and the mothers produced much less questions asking about referents that were new and absent. When the children asked questions about those referents, they tended to use wa-ending questions (Type 3). However, their mothers used other types of questions such as yes/no questions including a sentence-final question marker (Type 1); wh-questions (Type 2); and multi-word utterances without any overt lexicogrammatical question marking (Type 6). By contrast, the use of wa-ending questions (Type 3) was not as distinctively frequent as in their children’s data.
The present dissertation argues that children were able to express cognitively complex concepts such as new-ness and absence of referents from their second birthday. When they do so, they tended to use a particular question construction. For Japanese children, type 3, wa-ending questions serve that function. The discrepancy between the children and their mothers in their use of questions for new and absent referents was explained by taking into account the asymmetrical nature of the relationship between young children and their mothers.
Thursday, July 20 at 11:00am to 1:00pm
Poulton Hall, 230
1421 37th St., N.W., Washington