Intonation in the Perception of Brummie

Abstract

The Birmingham accent, also known as Brummie, enjoys a very bad reputation in Great Britain. It was suggested that its intonation is responsible for the stigma in the first place (How to speak Brummie, http://www.bbc.co.uk/dna/h2g2/A496352, 2001). Since Birmingham intonation differs from standard British intonation, and because intonation is indeed pragmatically meaningful for the perception of speech, two experiments at Adam Mickiewicz University were carried out to verify the hypothesis. The first one was to answer if Brummie is indeed disfavoured when compared to other dialects. The subjects listened to three Brummie speakers and three speakers of different accents of English, and rated the perceived attractiveness, friendliness and intelligence of the recordings on a 5-point Likert scale. The accent, as expected, was deemed the least attractive and intelligent. The second, core part of the experiment investigated to what extent intonation is responsible for this bad perception of Brummie speech. This time, the subjects were to listen to two versions of the speech sample of a given accent. The first version contained only intonation (the speech was unintelligible), and the second one included only segmentals (intonation was removed from the signal). The modifications were made using the methodology of Van Bezooijen and Gooskens (J Lang Soc Psychol 18(1):31–48, 1999) in PRAAT. The results of the two versions of each recording were then compared. On the whole, it turned out that Brummie intonation is indeed seen more negatively than RP intonation.

Full text

Intonation in the Perception of Brummie
Kamil Malarski
Abstract The Birmingham accent, also known as Brummie, enjoys a very bad
reputation in Great Britain. It was suggested that its intonation is responsible for
the stigma in the first place (How to speak Brummie, http://www.bbc.co.uk/dna/
h2g2/A496352, 2001). Since Birmingham intonation differs from standard British
intonation, and because intonation is indeed pragmatically meaningful for the
perception of speech, two experiments at Adam Mickiewicz University were
carried out to verify the hypothesis. The first one was to answer if Brummie is
indeed disfavoured when compared to other dialects. The subjects listened to three
Brummie speakers and three speakers of different accents of English, and rated the
perceived attractiveness, friendliness and intelligence of the recordings on a 5-
point Likert scale. The accent, as expected, was deemed the least attractive and
intelligent. The second, core part of the experiment investigated to what extent
intonation is responsible for this bad perception of Brummie speech. This time,
the subjects were to listen to two versions of the speech sample of a given accent.
The first version contained only intonation (the speech was unintelligible), and the
second one included only segmentals (intonation was removed from the signal).
The modifications were made using the methodology of Van Bezooijen and
Gooskens (J Lang Soc Psychol 18(1):31–48, 1999) in PRAAT. The results of the
two versions of each recording were then compared. On the whole, it turned out
that Brummie intonation is indeed seen more negatively than RP intonation.
K. Malarski (&)
Adam Mickiewicz University, Poznan
´, Poland
e-mail: kamil_go@yahoo.com
E. Waniek-Klimczak and L. R. Shockey (eds.), Teaching and Researching English
Accents in Native and Non-native Speakers, Second Language Learning and Teaching,
DOI: 10.1007/978-3-642-24019-5_15, !Springer-Verlag Berlin Heidelberg 2013
207

1 Introduction
Brummie is the accent spoken in the city of Birmingham in the area of the West
Midlands. Despite its being widely discussed in the media, Brummie has not
received too much attention from linguists. This was noted by Foulkes and
Docherty (1999) and several years later by Clark (2008). There are a few sources
(Wells 1982; Clark 2008) describing its segmental phonology but the information
about its suprasegmental features is really scarce. The following paper is an
attempt to fill this gap and to understand popular opinions on this accent. This
article is largely based on Malarski (2010).
2 Brummie’s Stigma
The West Midlands accent is possibly the most stigmatized variety in Great
Britain. It scores worst in various accent evaluation studies that considers criteria
like intelligence, prestige or attractiveness (Giles 1970; Hiraga 2005; Coupland
and Bishop 2007). Among other things, it has been described by people as ‘‘lazy’’,
‘‘ugly’’ and ‘‘uneducated’’ (Thorne 2005). This has serious social implications for
the Brummie users. As Dixon et al. (2003) show, subjects are much more likely to
be thought guilty when accused of a crime if they speak in Brummie rather than in
Standard British English. Giles et al. (1975), too, show that people usually see the
Birmingham accent speakers as less trustworthy than the RP speakers.
Of course, there is no one simple answer to why certain varieties of English are
perceived as worse than others. The bias against the Birmingham accent could be
because it has traditionally been an industrial city with a large proportion of
inhabitants coming from lower socioeconomic classes. Apart from the obvious
connection with the people who speak them, accents are also evaluated on the
basis of their phonetic traits (Van Bezooijen and Gooskens 1999, 47). It was
suggested in an online article (http://www.bbc.co.uk/dna/h2g2/A496352) that
Brummie is seen negatively entirely because of its intonation. The article was
written by a layperson rather than a linguist but strongly inspired the following
study.
3 Intonation in Brummie
Linguists have emphasised that intonation is pragmatically and semantically
meaningful for the listeners (Grabe et al. 2003, 379) and conveys emotion (Hlebec
2008). It has even been suggested that prosody, which includes intonation, is more
telling than segments in evaluating the speech of others (Anderson-Hsieh et al. 1992).
208 K. Malarski

Brummie intonation differs from the Standard British English intonation, pri-
marily in that Brummie and other urban British varieties like Scouse, Geordie,
Belfast and Glasgow use rising tones for statements (Wells 1982, 91; Cruttenden
1994, 138–139). These rises, in Brummie, are mostly realized by ‘‘rise-plateaus’’
and ‘‘rise-plateau-slumps’’ (Cruttenden 1994, 139). Cruttenden (1994, 139)
explains his own term ‘‘rise-plateau’’ as ‘‘the jump-up on the unaccented syllable
following the nucleus and the maintenance of this level on succeeding unaccented
syllables’’. The ‘‘rise-plateau-slump’’ is similar with the exception that the last two
syllables in an intonational unit can drop in pitch.
Ladd (1996, 125) interprets Birmingham declaratives as demonstrating a stylised
low rise intonation and asserts that they are the same tones as rise-plateau-slumps.
Brummie questions, on the other hand, are interpreted as rise-falls (Ladd 1996,
125). These intonational contours, that is low rises and rise-falls, appear in standard
British intonation. There, however, they convey negative attitudes. In RP, low rises
communicate ‘‘resentfulness’’, ‘‘deprecation’’, ‘‘reproving criticism’’ (O’Connor
and Arnold 1967, 169), ‘‘non-finality’’ and ‘‘suggestion’’ (Hirst 1999, 63). Rise-
falls, in questions, convey ‘‘challenging’’, ‘‘antagonistic’’ and ‘‘disclaiming
responsibility’’ attitudes (O’Connor and Arnold 1967, 147). Thus, it seems that
there are reasonable grounds to claim that Brummie rising tones can elicit unin-
tended negative interpretations.
4 Experiment 1
The experiment comprises of two parts. In the first experiment (Experiment 1), the
subjects were asked to evaluate three Brummie speakers and three speakers of
different varieties of English according to three criteria. Its aim was to discover
whether the Birmingham accent is as disfavoured by Polish students as it is among
British people. The first part was meant as an introduction to the second and more
important part of the experiment.
4.1 Data
Six speech samples appeared in Experiment 1. There were three Brummie
speakers (all males), one RP speaker (a male), one rural West Yorkshire speaker (a
female) and one Liverpool speaker (a female). Two Brummie recordings came
from the British Library Archival Sound Recordings available online (1998). One
Brummie speaker was recorded from the Internet. The remaining samples came
from the International Dialects of English Archive website (1997). All six voices
were recorded using PRAAT 5.1.18. For better loudness and quality, they were
normalised and dynamically compressed in Audacity 1.2.6. The samples lasted
from 9 to 18 s.
Intonation in the Perception of Brummie 209

4.2 Procedure
The experiment was in the form an online questionnaire carried out through the
Moodle e-learning platform at the departmental website of the School of English at
Adam Mickiewicz University in Poznan
´. The participants were to evaluate the
recordings on the 5-point Likert scale. The criteria were niceness, intelligence and
friendliness, features which are usually low-scoring for Brummie. On the scale, 1
meant ugly/unintelligent/unfriendly whereas 5 meant nice/intelligent/friendly. 3
was neutral. The listeners were also given the chance to comment in their own
words on how a speaker sounded. Spoken passages which could tell something
about the origins of the speakers, or their occupation, or in any way influence the
judgments were eliminated. The questionnaire can be found in Appendix 1.
4.3 Subjects
Thirty-one Polish students of English at the School of English at Adam Mick-
iewicz University in Poznan
´, 26 women and five men, took part in the experiment.
They were judged to be very fluent in English and ranged in age from 21 to 26.
Twelve were familiar with the American pronunciation model, and 19 had learnt
British pronunciation. Although they had had formal training in phonetics and
phonology and linguistics, they were not expected to identify the dialects, which
were presented in this study, apart from RP (Standard Southern British) (see, e.g.,
Weckwerth et al. 2005 for accent recognition among Polish students of English).
4.4 Hypothesis
It was argued in the preceding sections that certain intonational features of
Brummie have the potential to influence people’s negative reactions towards the
accent. Therefore, Birmingham English is expected to score low in this accent
evaluation study.
5 Results
The complete results are presented in Table 1of Appendix 2. The scores were
rounded to the nearest 0.1. The accents were classified according to the overall
result, which was the sum of the scores from each of the three criteria. Addi-
tionally, at the bottom of the table, there appears the mean score for all three
Brummie speakers.
210 K. Malarski

The Birmingham accent, in average, was judged to be less desirable than other
accents which confirms the hypothesis for this part of the experiment. Two
Birmingham speakers were evaluated as least acceptable. However, one Brummie
user scored a little better and came before the Liverpool speaker. Birmingham
English was judged to be an unintelligent and rather ugly accent. Intelligence was
ranked near the bottom, and it is the only mean Brummie score that is below 3.
There is one especially cogent finding for the role of intonation in perceiving
the Birmingham accent: one subject commented, ‘‘the last segment in brother
scared me’’. The word ‘‘brother’’ was the last word in the utterance and it was said
on a rising intonation. On the spectrogram, it resembles a rise-plateau (see Fig. 1).
What the listener meant was probably not the last segment (which was a schwa)
but the last syllable, which had a characteristic rising intonation.
6 Experiment 2
The second part of the experiment, inspired by the work of Van Bezooijen and
Gooskens (1999), forms the core of this paper. It was aimed at determining exactly
to what extent Brummie intonation is judged negatively.
6.1 Data
In this part of the experiment, there were twelve recordings to be listened to by the
participants. These were six speech samples, each played in two versions. The first
Fig. 1 The sentence-final word ‘‘brother’’ pronounced by Brummie 1 speaker
Intonation in the Perception of Brummie 211

version included only intonation, the second version the segmental information
with a flat intonation. To attain an intonational version of a speech sample a
lowpass filter at 350 Hz was used. Thus, the speech was unintelligible, and only
intonation was heard. In the segmental version, pitch was flattened and normalised
at 109 Hz, i.e., an average pitch level for all speakers. All modifications were
performed in PRAAT using the methodology of Van Bezooijen and Gooskens
(1999). The recordings, just like in the first part of the experiment, were com-
pressed and normalised for better quality. They lasted, with one exception, from 10
to 30 s, which was considered long enough to carry a recognisable intonation
pattern. Even in the shortest speech sample, which lasted less than 2 s, the two
sentences were included. The samples came from the same speakers which were
featured in Experiment 1.
6.2 Procedure
The speakers appeared in a different order than they did in Experiment 1. Also, the
samples were different. Two versions of each speech sample were played one by
one, the intonational version first and the flat-intonation version second. The lis-
teners were supposed to rate each of them on exactly the same Likert scale as in
the first part of the experiment.
6.3 Subjects
The same subjects, who participated in Experiment 1, took part in Experiment 2.
6.4 Hypothesis
My hypotheses were that the Birmingham intonation would be less favoured than
RP intonation and that the intonational versions of Brummie speech would be less
favoured than the ones with flat intonation.
7 Results
The complete results are presented in Tables 2and 3in Appendix 2. The scores,
again, were rounded to one decimal place.
The first hypothesis is easily supported. RP intonation was seen as friendlier,
more intelligent and nicer than Birmingham intonation. All three Brummie
212 K. Malarski

speakers came at the bottom of the table which means that their intonation was
considered less attractive than the intonation of other varieties. To either support or
reject the second hypothesis, however, the results for each Birmingham speaker
have to be investigated separately.
Results for Speaker Brummie 1 suggest that the utterances with normal into-
nation were more acceptable than those with flat intonation. Here, the hypothesis
that the rising tones at the end of declarative sentences evoke unpleasant feelings
in the listeners is not applicable, for the speaker used a rise sentence-finally. The
pattern (see Fig. 2) resembles a rise-plateau where the relatively high pitch of the
rise is maintained until the end of the second sentence.
For the second Birmingham speaker (Brummie 2), on the contrary, the
hypothesis is confirmed as his utterances with normal intonation was perceived as
uglier and less friendly than his utterances with flat intonation. This speaker,
however, produced a falling tone sentence-finally. He did produce some rising
tones in statements but in non-final positions which is very common in standard
speech. Therefore, although on the surface it appears that the hypothesis was
confirmed, this speaker did not use contours common in Brummie.
The third Brummie user (Brummie 3) scored exactly the same overall for both
types of utterance. His rank for utterances with normal intonation was the lowest of
all speakers (see Table 2). These negative judgments were not caused by the use
of a rising tone sentence-finally because the speaker used a falling tone at the end of
the sentence. Here, the second hypothesis is rejected, as well.
8 Conclusion
It has been suggested by lay observers that the low status of Birmingham English
may be due to the intonation used in this accent. Linguistic explanations also point
to intonation as very important in understanding and evaluating the speech of
others. Birmingham intonation differs from standard Southern British intonation.
Experiment 1 shows that rising tones in final positions as used in Brummie can be
unpleasant for the subjects. In Experiment 2, Birmingham intonation was arguably
viewed by the listeners as uglier, less friendly and less intelligent than the standard
intonation. However, results were somewhat ambiguous in this respect.
As noted above, literature on the Birmingham accent is scant. This study is a
small contribution, but there are numerous possible directions for further research.
The Brummie pitch range, which can have a great influence on how speech is
perceived (see, e.g., Grabe et al. 2003, 379), is an obvious example.
Acknowledgments I would like to thank dr Jarosław Weckwerth for very helpful comments on
the earlier draft of this article.
Intonation in the Perception of Brummie 213

Appendix 1
Questionnaire on accents of English
Please take 10 min to fill in this questionnaire
Please answer these questions about yourself first
1. Are you male or female?
2. How old are you? [optional]
3. A student of which year are you currently?
4. How long have you been learning English?
5. Which pronunciation model have you been taught?
6. Have you been to an English speaking country? How many times? How long
did you stay there? Where exactly have you been?
Listen to the following speech samples and rate them on the scales provided.
Preferably, listen to these through headphones. You can also leave some
comments.
7. Listen to the recording.
Does the speaker sound
1-Ugly, 5-Nice
1-Unfriendly, 5-Friendly
1-Unintelligent, 5-Intelligent
8. Comment on how the speaker sounds if you like.
[The same instructions for the remaining five recordings].
Fig. 2 The rising intonation by Brummie 1 speaker
214 K. Malarski

Now, listen to these recordings. They have been modified in some ways. Rate
them on the same scales. Preferably, listen to these through headphones. You can
also leave some comments.
19. Listen to the recording.
Does the speaker sound
1-Ugly, 5-Nice
1-Unfriendly, 5-Friendly
1-Unintelligent, 5-Intelligent
20. Comment on how the speaker sounds, if you like.
[The same instructions for the remaining eleven recordings].
43. If you have any comments on the questionnaire write them down.
Thank you very much for devoting your time and filling in this questionnaire.
Appendix 2
Table 1 Students’ reactions to Brummie and other accents according to the 5-point Likert scale
Accent Niceness Friendliness Intelligence Overall
RP 4.4 4.2 4.0 12.6
Rural West Yorkshire 3.8 3.9 3.5 11.2
Brummie 2 3.6 3.8 3.3 10.7
Liverpool 3.5 3.1 3.4 10.0
Brummie 1 3.3 3.5 3.0 9.8
Brummie 3 2.3 3.0 2.5 7.8
Mean Brummie 3.1 3.4 2.9 9.4
Table 2 Students’ attitudes towards intonation of the six varieties of British English
Accent Niceness Friendliness Intelligence Overall
Rural West Yorkshire intonation 2.5 3.1 2.8 8.4
RP intonation 2.5 2.8 2.9 8.2
Liverpool intonation 2.5 2.6 2.7 7.8
Brummie 2 intonation 2.3 2.6 2.8 7.7
Brummie 1 intonation 2.2 2.3 2.5 7.0
Brummie 3 intonation 2.1 2.3 2.5 6.9
Intonation in the Perception of Brummie 215

References
Anderson-Hsieh, Janet, Ruth Johnson and Kenneth Koehler. 1992. The relationship between
native speaker judgements of nonnative pronunciation and deviance in segmentals, prosody,
and syllable structure. Language Learning 42: 529–555.
British Library Archival Sound Recordings. 1998. (http://sounds.bl.uk/SearchResults.
aspx?query=Birmingham&category=Accents-and-dialects&publicdomainonly=false) (date
of access: 5 Feb. 2010).
Clark, Urszula. 2008. The English West Midlands: phonology. In Varieties of English 1: The
British Isles, eds. Bernd Kortmann and Clive Upton, 145–177. Berlin: Mouton de Gruyter.
Coupland, Nikolas and Hywel Bishop. 2007. Ideologised values for British accents. Journal of
Sociolinguistics 11: 74–93.
Cruttenden, Alan. 1994. Intonation. Cambridge: Cambridge University Press.
Dixon, John A., Berenice Mahoney and Roger Cocks. Accents of guilt? Effects of regional
accent, race, and crime type on attributions of guilt. Journal of Language and Social
Psychology 21: 162–168.
Foulkes, Paul and Gerard J. Docherty. 1999. Urban voices: Accent studies in the British Isles.
London: Arnold.
Giles, Howard. 1970. Evaluative reactions to accents. Educational Review 22: 211–227.
Giles, Howard, Susan Baker and Guy Fielding. 1975. Communication length as a behavioural
index of accent prejudice. Linguistics 166: 73–81.
Grabe, Esther, Burton S. Rosner, José E. García-Albea and Xiaolin Zhou. 2003. Perception of
English intonation by English, Spanish, and Chinese listeners. Language and Speech 46:
375–401.
Hiraga, Yuko. 2005. British attitudes towards six varieties of English in the USA and Britain.
World Englishes 24: 289–308.
Hirst, Daniel. 1999. Intonation in British English. In Intonation Systems. A Survey of Twenty
Languages, eds. Daniel Hirst and Albert Di Cristo, 56–77. Cambridge: Cambridge University
Press.
Hlebec, Boris. 2008. General attitudinal meanings in RP intonation. Studia Anglica Posnaniensia
44: 275–295.
International Dialects of English Archive. 1997.
(http://web.ku.edu/*idea/europe/england/england.htm) (date of access: 5 Feb. 2010).
How to speak Brummie. 2001. (http://www.bbc.co.uk/dna/h2g2/A496352) (date of access: 25
Nov. 2009).
Ladd, Robert D. 1996. Intonational Phonology. Cambridge: Cambridge University Press.
Malarski, Kamil. 2010. The Role of intonation in the perception of Brummie. Unpublished B.A.
paper. Adam Mickiewicz University.
O’Connor, Joseph D. and Gordon F. Arnold. 1967. Intonation of Colloquial English. London:
Longmans.
Table 3 Students’ attitudes towards segments of the six varieties of British English
Accent Niceness Friendliness Intelligence Overall
RP segments 2.9 2.9 3.1 8.9
Brummie 2 segments 2.7 2.9 2.6 8.2
Brummie 3 segments 2.1 2.5 2.3 6.9
Liverpool segments 2.0 2.4 2.3 6.7
West Yorkshire segments 1.7 2.2 2.5 6.4
Brummie 1 segments 1.9 1.9 2.5 6.3
216 K. Malarski

Thorne, Steve. 2005. Accent pride and prejudice: Are speakers of stigmatized variants really less
loyal? Journal of Quantitative Linguistics 12: 151–166.
Van Bezooijen, Renée and Charlotte Gooskens. 1999. Identification of language varieties: The
contribution of different linguistic levels. Journal of Language and Social Psychology 18, 1:
31–48.
Weckwerth, Jarosław, Katarzyna Janicka and Małgorzata Kul. 2005. Polish students’ attitudes to
native English accents as models for ESL pronunciation. In Pronunciation models: A
changing scene?, eds. K. Dziubalska-Kołaczyk and J. Przedlacka, 251–292. Frankfurt/Main:
Peter Lang Verlag.
Wells, John C. 1982. Accents of English. New York: Cambridge University Press.
Intonation in the Perception of Brummie 217