The effect of variable input on children's acquisition of null subjects

Language is essential to almost every aspect of our lives and it is absolutely central to learning in school. The goal of first language acquisition research is to determine how children acquire their native language given the linguistic input to which they are exposed. Research has shown that children do not just imitate word-for-word their input; instead, they produce forms not found in their parents' speech, such as 'He eat-ed it' or 'I'm hang onto-ing it.' Moreover, the linguistic input to which children are exposed is quite complex, as utterances do not always match their underlying representation. For example, in sentences like 'Do you want to go?' speakers variably omit the first two words, producing: 'You wanna go?', 'wanna go?', depending on social context. This project examines the impact of this latter sort of variable input on the acquisition of language in children.

The empirical domain of this project is the acquisition of properties related to--what has traditionally been called--the Null Subject Parameter in two varieties of Spanish: Dominican Spanish (DS) and Chilean Spanish (ChS). While ChS shows properties of a null subject language, DS displays a number of syntactic properties that are infrequent in other non-Caribbean varieties of Spanish, including variable usage of overt expletive subjects, higher use of overt pronominal subjects in contexts of topic continuity, and variable usage of subject-verb (non)inversion in interrogatives. By comparing acquisition patterns in children exposed to two varieties of the same language, we can compare acquisition of specific grammatical structures in the context of two input types (e.g. consistent input v. variable input) while holding other properties of the language constant.

This project will collect 120 hours of naturalistic speech between DS caregivers and their children while they play and read together and 40 hours of Dominican adult-to-adult speech. Audio-recordings and transcriptions will be archived on the CHILDES database (http://childes.psy.cmu.edu) for use by future researchers and will provide much needed data on language development in children acquiring non-mainstream dialects of Spanish, which is important in the context of standardized testing, early education, and clinical intervention in the case of speech disorders. This project will also provide international research experience to both graduate and undergraduate students, who will collect data in the Dominican Republic.