A Cognitive View on the “Deaccenting” Phenomenon Within Utterances

This article presents a new perspective on the “deaccenting” phenomenon. Phonological and semantic approaches to deaccenting refer to its association with a change of the sentence accent position. This article suggests that information structure differentiates between two related utterances of the same sentence: with deaccenting and without deaccenting. Thus, we propose a cognitive perspective on the structure of their F0 contours. We then present the cognitive model of information structure by defining the categories from which constituents take their functions within the cognitive and prominence structures of the prosodic relations of utterance trees. We later apply the cognitive categories and nucleus identification rules in order to correctly describe the prosodic unit hierarchies and the related prominence hierarchies of all utterances related to the sentences presented in the Introduction. The cognitive analysis of information structure within the utterances discussed, led to the conclusion that deaccented constituents have cognitive predicate function, while the same constituent, in the paired utterance without deaccenting, has cognitive argument function. Thus, we can claim that deaccenting is in essence a cognitive “predicativization” phenomenon because it changes the cognitive argument function of one constituent in a sentence into the cognitive predicate function in another utterance of the same sentence when the semantic context generates this phenomenon.