Communication in early infancy: three common assumptions examined and found inadequate.

Three taken-for-granted assumptions concerning the nature of communicative interaction between infants and adults are described. Their presence is demonstrated in three differently oriented theories ofearly interaction. The assumptions-that interaction is instrumentally oriented; that meaning is obvious and unitary; that change has an external source-are then put to empirical test, using narrative records from video-recordings made during a longitudinal study ofan infant girl and her mother. The outline ofa more adequate account of communicative exchanges is proposed. Language and communication are now 'placed in a central position in the philosophy ofthe social sciences. Language and the communicative interaction that it makes possible are seen as being the best candidates possible for the grounding of social scientific knowledge. Communicative interaction is the epi.t-temological foundation ofthe knowledge that people come to have both as members of a society, in their everyday practices, and as scientific investigators. example, each see discursive practices as playing this central role. Accordingly, there has been much empirical study of the communicative practices of persons in various roles and in a variety of types of situation. Yet little is known about the manner in which communicative practices develop before an individual achieves adult status in his or her society. There has been, of course, recent work in the field of psycholinguistics focusing on the pragmatic aspects oflanguage use in verbal children. The research to be reported here deals, however, with the pre-verbal child-the infant-with the intention of catching the development of communicative practices in their earliest stages. Adult-infant interaction has now become an area of attention for psychologists with varying interests; social development, attachment theory, psycholinguistics-each ofthese