What is a merger, and can it be reversed? : the origin, status and reversal of the 'NURSE-NORTH merger' in Tyneside English

Abstract

This thesis examines the apparent merger of the NURSE and NORTH lexical sets in Tyneside English. In order to determine its origin and status, whether reversal of the merger has taken place, and whether Wells (1982) is correct in his assertion that no hypercorrection has occurred, I examine two kinds of data: (1) traditional dialect phonetic transcriptions; and (2) an auditory and acoustic phonetic analysis of a socially stratified corpus of Tyneside English, the Tyneside Linguistic Survey (Pellowe et a/. 1972). Analysis of the first data-set suggests that there was indeed a merger of these two lexical sets. However, the sampling and elicitation methods employed mean that the real distribution of the merger within the speech community and within the speech of the informants themselves remains unknown. The second data-set is key, therefore, in that it reveals a range of speaker types, from those with complete merger to those with completely distinct NURSE and NORTH lexical sets. Additionally, there is evidence that hypercorrection of the NORTH lexical set has occurred, but on a narrower phonetic scale than Wells (1982) originally envisaged. In light of these divergent data and accounts of other'mergers', it is argued in this thesis that mergers are, in reality, diverse and complex sociolinguistic phenomena, so that questions such as Is Xa merger? and Has mergerXbeen reversed? only make sense when we know what kind of 'merger' X really was. Since the 'NURSE-NORTH Merger' is limited linguistically, geographically and socially, it is argued here that its reversal has been achieved with minimal disruption to the linguistic system in a way that would not be possible for other more widespread and entrenched mergers. Furthermore, it is suggested that it has been reversed in a phonetically gradual manner, such that obvious hypercorrection has been avoided. This possibility of phonetically gradual but lexically specific reversals of merger raises important questions for models of sound change.