Are word-stress variants in lexicophonetic corpora exceptional cases or regular forms?
In: Recent Advances in Corpus LinguisticsSearch for other papers by Franck Zumstein in
Current site
Google Scholar
PubMed
Purchase instant access (PDF download and unlimited online access):
Abstract
This study of word-stress variation relies on a sub-corpus of more than 2,000 word entries in which stress variants appear. They were extracted from a computer-searchable version of John Wells’ first edition of the Longman Pronunciation Dictionary (henceforth LPD1, Wells 1990). With the help of a selection of examples, I want to demonstrate that word-stress variation is the result of conflicting rules that are indicators of simplification in the phonology of British English. The theoretical framework adopted here is Lionel Guierre’s study of word stress (Guierre 1979) that includes a close examination of exhaustive results automatically obtained from a computer-searchable version of Daniel Jones’s twelfth edition of the English Pronouncing Dictionary (Jones 1963). More often than not, Guierre’s Normal Stress Rule (hence NSR) is involved in the conflicts.1 A close examination of the variations shows that NSR stressing appears to be the new variant that challenges traditional stress patterns. We may call this process “NSR regularisation”. Word stress variation is undoubtedly symptomatic of ongoing changes that are phonetically abrupt, as exemplified by the poll preferences in LPD, and lexically gradual since some words in identified lexical paradigms are not affected by the changes. In this study, directions of the changes are determined with the help of diachronic data extracted from pronouncing dictionaries of the 18th and 19th centuries. After such identifications of stress variations within the lexicon, further research needs to be carried out to put dictionary data to the test.