Constraints on prosodic phrasing in spontaneous speech

The present paper makes some preliminary hypotheses about the constraints on prosodic phrasing in spontaneous Swedish speech. Within Optimality Theory, linguistic phenomena such as prosodic phrasing are explained through the interaction of constraints (McCarthy & Prince 1993). One may assume that there are a number of constraints that interact to determine what prosodic form an utterance is assigned given its syntactic and information structure. Among these are constraints that both align prosodic phrase structure with syntactic and information structure, as well as constraints that assign word and focal accents and delimit the prosodic phrases’ size and accentual content. The constraints on output representations are hypothesized to be universal. Languages differ only in the ranking order of the constraints. Therefore, it is interesting to investigate in what way constraints previously claimed for other languages can be assumed to interact in Swedish. In the present paper, results from a production data study will be presented. Taking a number of constraints suggested to exist in other languages as a starting point, a possible constraint hierarchy for Swedish is discussed