Towards an Empirical Model of Language Acquisition

Abstract

Recent research in primary language acquisition can be subdivided into two schools. Under the influence of the highly mentalistic theory of language known as Transformational Generative Grammar, an innate language acquisition device was postulated, and the first school tried to establish what the structure of this device might be on the evidence of its output, i.e. early child grammars. Subsequently a new school arose which concentrated on mothers' speech to young children, providing empirical evidence against the nativist view that primary language acquisition occurs more or less independently of the ambient language environment. In the present article a language acquisition ‘device’ is suggested which differs in essential respects from the nativist-transformationalist model. Instead of an innately endowed ‘black box’ in the head of the child, the ‘device’ is seen as a two-way transmitting and receiving system existing between, and jointly operated by, the language learning child and society.