ABSTRACT
Clearly if we are to close the gap between studies of adult and child production, the possible role of optimal linguistic forms and cross-linguistic patterns must be further explored in both domains. Perhaps a framework for such investigations is a hypothesis that has been offered at different times and in different ways by several researchers (e.g., Donegan & Stampe, 1979; Stampe, 1969; Stemberger, 1989; Vihman & de Boysson-Bardies, 1994). On this view, the process of becoming a fluent speaker of one’s native language is one in which universal tendencies (such as the one for CV syllables) gradually come to coexist with structural and typical properties of the target language (e.g., the grammaticality of initial /sp/ clusters in English). On such a view, canonical or optimal linguistic forms might play a role for both children and adults, perhaps with the patterns observed in adult casual and impaired speech reflecting a maximum adherence to optimal forms. Differences between children and adults might reflect the fact that children’s optimal forms show a greater influence of universal tendencies, while adults’ optimal forms show a greater influence of the grammar and of frequently produced patterns.
