Nel, Joanine Hester2022-07-132022-07-1320222022*S2022https://repository.up.ac.za/handle/2263/86151DOI: 10.25403/UPresearchdata.20205116Thesis (PhD (Linguistics))--University of Pretoria, 2022This study investigated the use of articles, modal verbs and selected discourse markers (so, well and but) in Zimbabwean English (ZE) to determine whether there are variations and innovations in their use. Different morphosyntactic features that are reported by Kortmann, Lunkenheimer and Ehret (2020) to show innovation were used as the basis for analysis. A comparative analysis was done on the ZE corpus totalling 356 007 words and the International Corpus of English-Great Britain (ICE-GB) totalling 608 235 words in order to determine whether there are statistically significant variations in the frequencies of articles, modal verbs and discourse markers (so, well and but). A log likelihood test was used to check whether the observed variations between the corpora were statistically significant. The spoken and written registers and the different genres were compared for variations. The statistically significant variations observed in the frequencies of articles, modal verbs and DMs (so, well and but) in the spoken and written registers, and in different genres, pointed to the fact that there are differences in the frequencies of these constructions in the ZE corpus and the ICE-GB. No statistically significant variations in the spoken and written registers and in different genres in these constructions means that the frequencies of these constructions are not different in the two corpora. The statistically significant variation of these constructions in functions of these constructions point to the fact that ZE speakers and BrE speakers do not use the functions of articles, modal verbs and DMs (so, well and but) in a similar way. Where no statistically significant variations were recorded in the frequency of the functions of articles, modal verbs and DMs (so, well and but), this points to the fact that ZE speakers and BrE speakers use these functions the same way. These reported variations and no variations will aid in the determination of the stage of ZE in the DyM. In the ZE corpus, there was attested absence of the use of the definite article where StE has indefinite article (feature 60), use of indefinite article where StE has definite article (feature 61), use of definite article where StE favours zero (feature 64), indefinite article one/wan (feature 66), and non-standard use of modals for politeness reasons (feature 127). In relation to the use of zero article where StE has definite article (feature 62), use of zero article where StE has indefinite article (feature 63), use of indefinite article where StE favours zero (feature 65), double modals (feature 121), and present tense forms of modals used where StE has past tense forms (feature 123), L2 users of English who speak Shona as their L1 seemed to deviate slightly from L1 English conventions. This is because the features exist but are extremely rare. The use of demonstratives for definite article (feature 67) was attested to be neither pervasive nor extremely rare. Results from this study support the idea that ZE is at both stage 2 and stage 3 of Schneider’s (2003, 2007) dynamic model.en© 2022 University of Pretoria. All rights reserved. The copyright in this work vests in the University of Pretoria. No part of this work may be reproduced or transmitted in any form or by any means, without the prior written permission of the University of Pretoria.UCTDZimbabwean EnglishDynamic modeNativisationCorpus-based analysisWorld EnglishesThesisu12167160https:doi.org/10.25403/UPresearchdata.20205116