On the feasibility of the Danish model of intonational transcription: Phonetic evidence from Jutlandic Danish

Most of our knowledge of Danish f0 variation and intonation is based on the work of Gronnum and colleagues, who developed an a-phonological model in which a series of repeated "default" contours are superpositioned onto an overarching f0 slope. The current paper tests a range of predictions stemming from this model, most importantly the adequacy of analysing f0 modulations as a string of repeated contours differing in range but not in shape. To facilitate comparison with earlier work in the area, our material is based on read speech, 45 speakers of Jutland Danish participated in the experiment. Analyses of f0 in sentences of differing complexity supplied little evidence in favour of the existence of default contours. Instead, our acoustic data revealed an array of f0 shapes associated with various prosodic anchor points, which are influenced in both range and shape by positional context and the presence or absence of focus.