Cognitive categorisations of language: how EFL students’ (mis-)identifications of three British accents engender stereotypic attributions

ABSTRACT

Language attitudes involve two sequential cognitive processes: categorisation and stereotyping. Against this empirically underresearched postulate, this study illuminates the influence of self-reported accent identifications on speaker evaluations. Using the matched guise technique, attributions to three L1 varieties of British English were examined among Austrian university students (n = 268) who are learning English as a Foreign Language (EFL). Their judgments towards the EFL reference accent Received Pronunciation (RP), Standard Scottish English (SSE) and Estuary English (EE) reflected two factor-analytically confirmed constructs, i.e. competence and warmth. The manipulated phonetic cues significantly affected the guises’ evaluations in both dimensions, as well as their perceived news reader qualifications and the raters’ intentions to affiliate with the guises. A recognition of RP correlated with an upgrading of competence and news reader suitability, while the combination of EE features and a ‘south-eastern England’ categorisation resulted in downgrading in these dimensions. Conversely, when SSE was identified as ‘Scottish’, warmth ratings became more favourable and students expressed a greater affiliation intention. Altogether, the data underlines the importance of social categorisation by demonstrating that evaluations of a single speaker vary according to inferred regional group memberships by message receivers, which suggests the activation of distinct stereotypes.

KEYWORDS:

• Speaker evaluations
• categorisation and stereotyping
• variety identification
• matched guise technique
• language attitudes
• accents of English

Disclosure statement

No potential conflict of interest was reported by the author.

Notes

1 The Austrian constitution additionally recognises several minority languages including Austrian Sign Language and the languages of autochthonous minorities; see Dalton-Puffer, Faistauer, and Vetter 2011.

2 This brief discussion focuses on attitudes to spoken varieties of British English. To find out how English is embedded into the wider Austrian linguistic landscape, see Soukup (2016).

3 Preliminary between-subjects ANOVAs revealed no significant effects of the four ‘definition-congruent’ categories on competence, F(3, 144) = 0.27, p = .851, ${\eta }_p^2$ = .005, warmth, F(3, 144) = 0.36, p = .783, ${\eta }_p^2$ = .007, news reader, F(3, 144) = 0.57, p = .638, ${\eta }_p^2$ = .012 or affiliation, F(3, 144) = 0.23, p = .874, ${\eta }_p^2$ = .005, scores.

4 Preliminary Kruskal-Wallis H tests showed no significant effects of the four ‘definition-congruent’ categories on competence, χ²(3) = 2.65, p = .450, warmth, χ²(3) = 0.48, p = .924, news reader, χ²(3) = 2.22, p = .528 and affiliation, χ²(3) = 0.30, p = .959, scores.