Untangling Synchronic and Diachronic Variation: Verb Agreement in Palmerston English*. Issue 3 (2nd July 2016)

• Record Type: Journal Article
• Authors: Hendery, Rachel
• Is Part Of: Australian journal of linguistics. Volume 36:Issue 3(2016)
• Page Start: 429
• Page End: 450
• Publication Date: 2016-07-02
• Subjects: Verb Agreement -- English Dialects -- Creolization -- Regression Analysis -- Multivariate Statistics -- Sociolinguistics; Linguistics -- Periodicals; Australian languages -- Periodicals; 410
• DOI: 10.1080/07268602.2015.1134302
• Languages: English
• ISSNs: 0726-8602

Abstract

Abstract : This paper examines documents and recordings from the past 100 years to analyse the factors that affect the selection of verb form in Palmerston English. Palmerston Island is a small island in the Cook Islands group that was settled by a small group in the 1860s, which included an Englishman, a Portuguese or Portuguese-creole-speaking man, and a small group of Cook Islanders. The descendants of this population speak a dialect of English, and early documents from the island were primarily also written in vernacular English. As is the case in many varieties of English around the world, present tense verb endings do not exactly follow the Standard English pattern of -s following a third person singular subject and bare verbs elsewhere. The variation in -s marking, both historically and synchronically, has been abstracted away in previous descriptions of the dialect. This paper describes for the first time the variant verb agreement patterns found at each stage of the dialect's history, examines various grammatical and social factors that contribute to the distribution of the forms, and uses a logistic regression analysis to model the interactions of the factors and their contributions to the output. This contributes to a growing literature on the -s alternation in English dialects, and is one of only a few studies that has been able to include data from multiple individuals and also from a relatively long historical timeframe. I find that significant predictors of third person verb forms include subject type, subject number, whether the verb is be or not, and at least in the later data, the proximity of the subject to the verb. I also find that the amount of individual variation among speakers is large enough that it almost obscures diachronic variation. This is a finding that has methodological implications for future research into the phenomenon in other varieties, for example casting doubt on the usefulness of 'apparent time' studies.