The actuation of sound change

Abstract

This dissertation is a sociophonetic study on sound change in progress. It addresses the actuation problem, i.e. the question as to why a particular change takes place in a particular language at a given time. The study is implemented in the framework of exemplar-based theories, which incorporates individual variation and the influence of the ambient language on individuals. Two sound changes in progress in the Dutch language are selected: the devoicing of initial labiodental fricatives (/v/ changing to [f]) and of initial bilabial stops (/b/ changing to [p]). The study is articulated around a series of experiments, which provide insight into the role of four aspects of linguistic competence involved in sound change: speech perception, speech production, phonetic imitation and language attitudes. These aspects were tested on the same pool of participants, and subsequently linked together both at the group and the individual level. Hundred participants were recruited from five regions of the Dutch-speaking language area, representing different stages of sound change: Groningen, South-Holland, Limburg, Flemish-Brabant and West-Flanders. The 20 participants born and raised in each region were 10 males and 10 females aged between 18 and 28 years old. Regional and individual differences appeared both in the production and the perception systems. Individual speakers differed not only in their realizations, but also in their perception of the concerned variables. Moreover, we found clear evidence for a link between the perception and production systems, even if these systems are thought to be separate entities. Furthermore, our results pointed towards the fact that change in speech perception precedes change in production. However, speech perception appeared to lag behind speech production when the sound change is reaching completion. Listeners can still hear the difference between sounds they cannot produce themselves any longer. We showed that language attitudes towards a variable undergoing change do not seem to be present from the beginning of the sound change onwards. The positive evaluation of a change seems to arise when the system of speech production is already undergoing the change. Language attitudes thus develop during the process of sound change and seem to be reinforcing this process. Phonetic imitation capacities were related to individual states of production. It was demonstrated that the best imitators are at the beginning of the change and show conservative production patterns. The further advanced in the change, the worse the imitation capacities. Bad imitators seemed to lead the change, because of their weak link between perception and production. It is argued that sound change is an iterative process in which an individual's production changes incrementally. The actuation of change happens within the individual every time speech perception and speech production make an attempt to align with each other. Actuation is directly linked to individual differences in exemplars stored in the perception and production systems and to differences in the strength of the link between these systems. Individuals use their phonetic imitation ability to convert the details of segments they perceive into production. Positive language attitudes associated with specific linguistic features reinforce this process.