Perceiving Identity through Accent: Attitudes towards Non-Native Speakers and their Accents in English

Abstract

This book explains how variation in the pronunciation of English can affect the development of attitudes towards the speaker. There have been concerns about positive and negative discrimination between people who speak with different accents. This book is one of the first to examine attitudes towards non-native speaker (NNS) accents of English which could lead to discrimination and it is the first to compare two different kinds of NNS accents of English (German and Greek). \nThe main part of this book is based on a study which examined the effect of variation in Greek and German accents of English with varying degrees of influence from the first language (i.e. with ‘stronger’ and ‘weaker’ Greek and German accents). The aim of this study was to find out 1) what the general attitudes towards NNS accents of English are, 2) whether NNS of English feel a sense of solidarity with speakers from their own first language background and 3) what types of consonantal variation could influence attitudes towards NNS of English.\nIn the first experiment, the attitudes towards these NNS accents of English are compared to attitudes towards native speaker (NS) accents of English. The results indicate that our listeners (who are German and Greek NNS of English and English NS) tend to assign high prestige to a Received Pronunciation (RP)-like NS accent. In general, NNS listeners do not consider their own NNS accent of English to reflect their identity and issues of status and prestige seem to be of more importance to them than aspects of solidarity with their own speaker group. \nIn a further experiment, listeners from the same groups as before were asked to point out the ‘foreign’ sounding features of those speech stimuli of the NNS accents which received the lowest ratings in the previous experiment. The results reveal that variation in the realisation of /r/ and variation in sibilants (e.g. /s/, /z/) were most obvious to the listeners and seemed to influence the perception of the speaker. Further variation such as final devoicing and non-velarised /l/ in word-final and post-vocalic positions (e.g. the realisation of /l/ in “call” as opposed to “like”) were not detected by the listeners which could mean that listeners did not recognise these variations.\nIn a final experiment, the aim was to find out whether the consonant features that were isolated in the previous experiment can influence attitudes. Overall, consonantal variations were more positively rated the more similar they were to the RP variants, indicating that the RP accent was the mental model that NNS accents were compared to. Additionally, the results show that variation in consonants was important beyond the intelligibility of accents since they directly influenced the perception of the speaker.