Page 86 - Encyclopedia of Philosophy of Language

Language and Mind
behavior in infants 12-24 months old. Merrill-Palmer
Quarterly 11:129-48
Sugarman S 1983 Children's Early Thought. Cambridge Uni-
Vygotsky L S 1962 Thought and Language. MIT Press, Cam- bridge, MA
Wittgenstein L 1951 Philosophical Investigations. Blackwell, Oxford
versity Press, Cambridge
The major thrust of studies of children's language development from the perspective of linguistics has involved the grammatical analysis of spontaneous speech samples obtained from the child's con- versations with their mothers and other interlocutors. Phonological, lexical and discourse issues have also been pursued, but it is at the grammatical level that much of the research energy has been directed, in the main because it is here that links between linguistic theory and language acquisition are most directly made. Also, spontaneous speech from children is rela- tively easy to collect, and furnishes extensive corpora of utterances in naturalistic settings. These samples are used to estimate a child's grammatical status at successive stages of the developmental process. Cer- tain utterances by the child have proved of particular importance for researchers. These are often referred to as 'errors.' More accurately, they are non-adult forms which the child produces, and they often pro- vide a window into the child's construction of gram-
mar which would not otherwise be available. The most frequently cited example of this in English is over- regularization of past tense. Productions by the child of forms like corned, bitted, and buyed, in place of the irregular forms came, hit, and bought are evidence of grammatical immaturity, certainly, but they also indicate clearly that the child has mastered the rule for regular past tense formation. While the focus of enquiry has not changed over the 30 years or so of linguistic studies of language acquisition, metho- dological advances and theoretical reformulations have had significant effects on the field. The following sections will concentrate on the major trends in the methodology and theory of child language studies, so far as they relate to grammatical development.
1. Methodology and Theory: The First Phase
The modern history of the study of children's language development begins in the early 1960s. Not surpris- ingly, for any branch of linguistic research in the second half of this century, Noam Chomsky was a formative, though initially indirect, influence. The first research project of the modern era was planned and directed by Roger Brown at Harvard University from
1962, following a five-year period Brown had spent at the Massachusetts Institute of Technology. The project, written up in Brown (1973) and numerous theses and research papers, was designed to determine the stages of grammatical acquisition in English- speaking children. In its concentration on the child's construction of grammar, independently of the con- text in which it was acquired, using as a database longitudinal samples of children's speech over the whole of the preschool period, Brown's project is a model which other researchers in the field have copied, modified or reacted against. This section therefore begins by looking in more detail at the methodology of this important study, as a starting-point for a more general consideration of methodological and theor- etical issues in the field.
1.1 Sampling the Data
The study of children's development over time (hence 'longitudinal') was not new with Brown. The late nine- teenth and early twentieth century saw a number of 'diary studies' on the acquisition of various languages (see Ingram 1989: 7ff for a review). Typically an inter- ested parent would note the child's utterances, from the emergence of recognizable words onward, and provide a commentary on what appeared to him/her to be interesting features. The frequency of entries, the timespan of the child's development covered, and the features of interest noted, were somewhat unpre- dictable from diarist to diarist. A major handicap for the diarist, and later researchers who wanted to use the information contained in them, was the inevitable selectivity imposed by the method of handwritten records of increasing quantities of speech once the child passed the second birthday.
In his study, which examines the preschool devel- opment of three children (code-named Adam, Eve, and Sarah) Brown had the inestimable advantage over his diarist predecessors of being able to tape-record spontaneous speech samples from his subjects. This technological advance made a major difference to the data available to researchers. A permanent record of what the child and his interlocutors said was now available, which could be used to provide reliable writ-
64
Language Acquisition in the Child P. Fletcher