Results: attitudes to labels

Have attitudes to different named accents changed over time?
When asked about their attitudes to accent labels, the UK public has shown remarkable consistency over the past 50 years. Our results confirm that labels such as “Queen’s English” and “Own English” fare well, and that certain local (e.g. “Birmingham”) and ethnic (e.g. “Afro-Caribbean”) accent labels fare poorly.

Overview

Our research compared current attitudes to accent labels with those shown by comparable surveys 15 years ago and 50 years ago. We found that, despite some changes, public attitudes to different accents and their related stereotypes have remained largely unchanged over time. 

Has public perception of accents changed over time?
Despite some change, a clear hierarchy of accents has remained consistent for the last 50 years.
Historical graph showing accents rating over time

1970: identifying accent bias

In the late 1960s, Howard Giles set out to map British listeners’ evaluations of 16 different accent labels, including ‘a standard accent of English’, various non-native accents (‘French’, ‘German’), various ethnic accents (‘Indian’, ‘West Indian’) and various regional UK accents (‘Northern English’, ‘Liverpool’, ‘Somerset’).

Respondents were asked to rate each accent label for how prestigious and how pleasant they think it sounds.

Giles (1970) reported a very clear pattern of evaluation of the different accent labels: ‘standard English accent’ received the highest ratings for prestige and pleasantness, while various non-standard urban vernaculars (Liverpool, Cockney, Birmingham) were rated lowest. Ethnic varieties were also rated low for prestige, while non-native varieties were rated highly. 

Finally, Giles found that the label ‘an accent identical to your own’ was rated high for prestige for all respondents. Giles’ findings provide a snapshot of attitudes to accents in Britain at the time.

2004: patterns of bias remain consistent

In 2004, Hywel Bishop, Nikolas Coupland and Peter Garrett, in cooperation with the BBC, replicated Giles’ study, this time asking over 5,000 British respondents to rate 34 accent labels.

Their results were remarkably similar to what Giles had found 35 years earlier: ‘Standard English’ and ‘Queen’s English’ were top-rated for prestige, immediately followed by ‘accent identical to your own’. The ‘Birmingham’ accent again received the lowest rating, with ‘Black Country’ and ‘Liverpool’ doing only slightly better. Ethnic varieties of English were also poorly rated, with ‘Asian’ and ‘Afro-Caribbean’ accents both in the bottom decile of evaluations.

The findings of Bishop et al.’s (2005) study illustrated a striking consistency in evaluations of accent labels across three decades. Despite the major changes in British society between 1970-2005, beliefs about which accents are the most prestigious had remained largely the same.

Today: attitudes are largely unchanged

Our examination of attitudes to accent labels provides a third historical time point, allowing us to compare present-day attitudes to those from 50 years ago (as reported in Giles 1970) and 15 years ago (as reported in Bishop et al. 2005). 

We expanded the accents examined to 38 by including newer accents that have emerged over the past two decades (‘Estuary English’, ‘Multicultural London English’) and well as some additional non-native accents that have become more prominent in the UK (‘Chinese’). 

Our results show an extraordinary level of continuity in evaluations of accent labels. Like Giles 50 years ago, we find that a ‘standard’ accent is rated the highest for prestige, with ‘Edinburgh’, ‘New Zealand’, ‘Australia’ and ‘Estuary’ all appearing within the top 10. Like in the two previous studies, ‘Birmingham’ is once again rated at the bottom, with ‘Afro-Caribbean’, ‘Indian’, ‘Liverpool’ and ‘Cockney’ all in the bottom 10. 

Conclusion

There is some evidence that differences between accents are reducing: while standard accents are still ranked highest and urban and ethnic vernaculars ranked lowest, the quantitative distances between the top- and bottom-rated accents are smaller in our study than in either 2004 or 1969. 

Nevertheless, our results show a persistent hierarchy of accent evaluations, one that penalises non-standard working-class and ethnic accents and upholds the belief that national standard varieties are the most prestigious.