Intonation: Theory, Models, and Applications

Athens, Greece
September 18-20, 1997


Empirical Evidence for the Contrast Between L* and H* in Dutch Rising Contours

Carlos Gussenhoven, Toni Rietveld

Department of Language and Speech, University of Nijmegen, Nijmegen, The Netherlands

A Dutch accent-lending rising intonation contour can be realised either as a rise that begins low and ends mid-to-high (henceforth 'low rise'), or as a rise that begins mid and ends high (henceforth 'high rise'). These two contours can either be seen as the extremes on a continuum of phonetically different realisations of the same phonological contour L*H H%, or as phonetic realisations of two phonologically different contours, L*H H% and H* H%. One property that potentially distinguishes L-tones from H-tones is their behaviour under variation in pitch range. H* is generally claimed to be higher as the pitch range increases, and at least some researchers have claimed that L* is lower as the pitch range is increased. Our approach to the issue of the phonological status of high and low rises in Dutch took this hypothesis as its point of departure. To this end, we manipulated the F0 contours of four utterances, so as to produce a range of stimuli for each of them, which represented high rises and low rises in a number of different pitch ranges. Listeners were asked to indicate the degree to which each of these stimuli expressed 'Surprise' The results lend support to the hypothesis that the low rise and the high rise are categorically distinct contours of Dutch.

Full Paper

Bibliographic reference.  Gussenhoven, Carlos / Rietveld, Toni (1997): "Empirical evidence for the contrast between l* and h* in dutch rising contours", In INT-1997, 169-172.