Abstract
Rhotics, or r-sounds, are known to display a large amount of variation, both cross-linguistically and within particular languages. Dutch is an example of such a language. Even within what is generally regarded as the standard variety, there is variation in place of articulation (ranging from alveolar to uvular), manner of
... read more
articulation (trills, taps, fricatives, approximants and vowels), and r can be voiced, voiceless, or not there at all. This wide-ranging variation is governed by both linguistic and extra-linguistic factors, in a complex interplay of intra- and inter-speaker variation. The present study’s main empirical aim is to catalogue the variation and untangle the underlying factors, based on the collection and analysis of a corpus of acoustic speech data from over 400 speakers (~20.000 tokens) in ten cities in the Netherlands and Flanders. In addition, it presents a detailed articulatory (ultrasound) study of an innovative coda approximant variant. Results from the urban accent corpus show that around 20 different phonetic variants of r can be distinguished and that many of these variants occur in all of the cities in the study. Syllable onsets are more restrictive and allow around 10 variants, whereas around 16 variants occur in codas (with all 20 occurring in the Utrecht accent). The most frequent variants in onsets are the voiced alveolar tap, the uvular approximant, and the uvular trill; most frequent in codas are the retroflex or bunched approximant, the voiceless alveolar tap or trill with frication, and the uvular fricative. Speakers tend to have either alveolar or uvular consonantal variants in onsets, although 16% have variation between these two places of articulation; Belgian Dutch speakers also have trill, tap or fricative variants in codas, but speakers in the Netherlands prefer approximants and vowels, with the retroflex/bunched approximant most frequent. Results from the ultrasound study reveal that the coda approximant known as Gooise r in Dutch, which is currently rapidly spreading in the Netherlands, is produced as either a retroflex or a bunched palatal approximant, echoing similar results from studies of American r. They also show the abstract, phonological nature of r-allophony for these speakers, as their articulatory strategies for the coda allophones do not depend on their onset allophones (alveolar taps or uvular trills, which can both be combined with retroflex or bunched articulations). The theoretical contribution of the thesis lies in the development of a model of progressive sound change to account for the origins, development and current status of Dutch r-variation. In doing so, it shows the relevance of detailed sociophonetic analysis for answering phonological questions. A diachronic model of “family relationships” between the different variants of r is proposed, which shows how the appearance and distribution of particular variants can be explained by reference to casual speech processes, particularly that of lenition. This proposal expands on earlier models that showed the phonetic commonalities between different r-variants without making predictions related to the sound changes that could give rise to them. Making these family relationships explicit provides a way to establish the long sought for unity among r-variants. Importantly, this unity is found in the diachronic, not the synchronic, domain.
show less