ABSTRACT

Research that aims to examine mothers’ speech in different social groups and its effects on children’s language development has to draw from at least two research traditions. First, developmental psycholinguistic studies have accumulated an enormous amount of knowledge on mother-child interaction, the characteristics of input, and its possible effects on language development. Investigations, though, have been limited to young children, to data usually gathered in naturalistic settings, and to English-speaking populations such as mainstream groups in North America, Australia, and Britain. The second area of interest encompasses studies of SES differences. Apart from differences in points of departure and basic theoretical issues, studies of SES differences have concentrated on older subjects, investigated mother and child speech (either separately or in dyads) often in artificial situations (Bereiter & Engelmann, 1966; Cook Gumperz, 1973; Hawkins, 1969; Hess & Shipman, 1965; Robinson & Rackstraw, 1972; Turner, 1973), and only rarely in naturalistic settings (for the latter, however, see Wootton, 1974; Tizard & Hughes, 1984). Like input studies, most SES studies have been limited to the English-speaking population of Britain and the United States. Studies connecting these two domains, that is, investigations examining social variation in input language to young children in naturalistic settings (e.g., Snow, Arlmann-Rupp, Hassing, Jobse, Joosten, & Vorster, 1976; Wells, 1980; Ninio, 1980) have been rather uncommon until recently. A closer connection between these two domains of research should, however, prove useful to both fields. In studies of input, individual variation in maternal style has been exploited to scrutinize effects of particular features on children’s language development. As Wells and Robinson (1982) pointed out, studying input in sociologically con­ trastive social groups is consistent with this approach. In a sense, it could be

198 REGER

considered as an extension of the range of individual variation, a range, Wells and Robinson (1982) wrote, “within which individual differences combine to yield differences across socially based categories” (Wells & Robinson, 1982, p. 42). The study of SES differences, on the other hand, can extend the scope of analysis to the ontogenesis of later socially based differences and to the subtle mechanisms of early interaction by which ways of speaking and related social knowledge are transmitted from generation to generation (Heath, 1982, 1983).