Dynamic Systems Theory: Reinterpreting "Prosodic Bootstrapping" and Its Role in Language Acquisition
The strong position that prosody could reveal complete phrase struc ture trees was never really endorsed by the framers of the "prosodic bootstrapping hypothesis" (Gleitman & Wanner, 1982; Morgan & New port, 1981). Rather, they suggested that prosodic information m ight provide useful and im portant cues to form class, major constituents, open class versus closed class units, and so forth. This volume is testi mony to the fact that there are a num ber of informative cues in the pro sodic input; cues that may take the learner a long way towards discov ering the units of language. The chapters in this volume, however, also highlight the need for ways to examine how children use prosody in conjunction with syntactic, semantic, social, and morphological cues to solve the "logical problem of language acquisition" (Baker & McCarthy, 1981). This chapter presents one way to think about how children m ight use m ultiple input sources to solve this complex acquisition prob lem. By introducing the m etaphor of dynamic systems theory (see Thelen & Smith, 1994), we offer a new way of thinking about the prob lem space and offer concrete suggestions on how to empirically investi gate the role of m ultiple inputs on the child's developing grammatical system.