Deconstructing stress

In this article, I will show that, despite its wide use, the term ‘stress’ is too ambiguous and therefore must be replaced by a set of other terms that together cover the various usages of ‘stress’. Firstly, we need to be able to distinguish between the formal representation of ‘stress’ and the correlates of ‘stress’. Within the formal representation of ‘stress’ I will show that we have to separate four notions: accent, edge prominence (EP), rhythm and weight. Accordingly, we also need to reckon with potentially different correlates. Here, I will use stress as a general label for phonetic correlates and distinguish stressA, stressEP, stressR and stressW. The article will then mainly focus on accent and its correlates. I will first propose central parts of a theory of accent. Then, I will address the question as to whether we need to single out pitch (or tone) as a special correlate of accent, i.e. one that is not included in the correlate stressA, which would entail a currently controversial distinction between what is traditionally called stress-accent and pitch-accent. The article concludes with a discussion of refinements, problems and extensions of the accentual system proposed here (including the deconstructing of ‘tone’).