Accent, Status, and Identity

2.3.1Accent and attitudes

It has been argued that accent is an integral part of everyone’s speech, both native and non-native English speakers have an accent. Our evaluation of those features of accent leads to assigning those characteristics to certain groups – creating stereotypes. There is a persistent idea that there exists a neutral, ‘standard’ accent which is accepted by the majority – making it superior, and anything not conforming to that standard is perceived with prejudices. Media, practices in the language classrooms, as well as nationally-held stereotypes create an image of a favorable accent variety, which is usually either the GB or GA accent. This accent is then assigned a high level of status, and is viewed as the most desirable. The social perceptions of non-native accents play an important role in the educational planning, and have a large impact on mutual understanding of two speakers.

Learners (…) may not realize the extent to which attitude and tolerance play a role in communicative success and failure. Consequently, teaching for linguistic tolerance, given the sociolinguistic reality of English for Expanding Circle users and learners, may be as relevant, if not more so, in terms of their needs (Berns, 2008).

2.3.2 Accent and language acquisition

Crystal argues that “the need to maintain international intelligibility demands the recognition of a standard variety of English, at the same time as the need to maintain local identity demands the recognition of local varieties of English.” (2001, p.57). Crystal is trying to come to terms with Kachru’s notion of World Englishes and the need for their acceptance as legitimate varieties. Most of the learners now study English for pragmatic reasons, such as an access to information, social mobility, job advancement, etc., but not for integrative ones. Crystal (2001) therefore suggests to turn from the prescriptivist concept of one dominant ‘correct English’ to a more liberal approach to allow for greater flexibility and a recognition of the importance of one’s identity.

A large emphasis has traditionally been put on the importance of native accent in language education. However, as noted in the previous chapter, acquisition of a native accent is very unlikely to occur with an increasing age of the learner, and it is generally very improbable unless one begins to learn it as a child. Many non-native speakers therefore never see themselves as users of English because they constantly compare their speech to that of native speakers. Not only the native speaker pronunciation model presents an unattainable goal for most of the learners, but it is also viewed as an inappropriate goal regarding the learners’ cultural identity. Giles and Smith claim that
retaining L1 accents might have a more positive influence while communicating with NSs than NNSs modeling themselves on their interlocutors. Copying the NSs too precisely, or in other words encouraging a greater convergence towards the interlocutors’ linguistic identity, might be perceived negatively as this might be attributed to the NNS’s projection of himself/herself (Giles and Smith, 1979).
Based on the presented research, a focus on a more pragmatic ability to communicate, with ‘international intelligibility’ as a central goal for the pronunciation should be part of the approach among ELT professionals in Europe in order to improve the learners’ communicative abilities.

2.3.3Accent and intelligibility

The speakers were university students from China, Indonesia, India, Japan, Papua New Guinea, the Philippines, Taiwan, the United States, and the United Kingdom. The listeners were divided into three groups, NNS, NS, and mixed group. The NNS in the first group were of various proficiency, and familiar with the Japanese, American, and British varieties, and with the Japanese speaker’s presentation topic. In the NS group, the subjects were familiar with the American speakers’ variety and topic. The subjects in the third group were highly proficient users of English and were familiar with several different national varieties of English. Recordings of the nine speakers’ interactions between each other in pairs talking on the same topic were produced, and served as a material for the ‘understanding’ tests. A cloze procedure was used to test intelligibility, a multipole-choice test for comprehensibility, and paraphrasing of a passage of the recording for interpretability.

The results of the study confirmed that there are differences in the three elements of understanding: intelligibility is easier than comprehensibility or interpretability. The familiarity with the topic didn’t prove to have significant effects on the results, but familiarity with the variety did. Those who were more familiar with the varieties achieved better score in the interpretability test. Language proficiency had the greatest influence on comprehensibility, less on intelligibility and interpretability. The results further revealed that the native speakers were not the most easily understood, nor were they best able to understand. Thus, being a native speaker was not as important as being fluent in English and familiar with the English varieties. Smith concludes that “the increasing number of varieties of English need not increase the problems of understanding across cultures, if users of English develop some familiarity with them” (p. 441). Smith also adds that more studies of interactions among Outer and Expanding circle varieties not involving Inner Circle users, to represent the current situation of the language use, are needed to reveal the factors which affect intelligibility.

In their much-cited study, Munro and Derwing (1995b) had come to a conclusion that accent doesn’t directly affect comprehensibility. When the listeners in their research made accentedness judgements of the NNS, they most probably judged the accents’ degree of deviation from their idea of a standard accent, but those which were rated as strongly accented didn’t receive a lower comprehensibility score. The following implications for ELT were drawn from their research: