“Swedish People’s Attitudes to Different Accents of English”

STOCKHOLM UNIVERSITY
Department of English
“Swedish People’s Attitudes to Different
Accents of English”
Franz Myrman
Special Project PK
Linguistics
HT 2004
Supervisor: Peter Sundkvist
Abstract
The aim of this study is to investigate Swedish people’s attitudes to
different accents of English. The focus is on Swedish female
informants’ attitudes to RP, non-standard English English, standard
Irish English and non-standard Irish English. Furthermore, this
study aims to give a wider perspective on how important
experience of specific accents is, in order to have systematic
attitudes to them.
The results show that experience is very important. They also
show that Swedish people have experience of the English language
and some of its accents, although this experience is relatively
limited. Above all, this study shows that Swedish people do have
systematic attitudes to English accents, but not as systematic as
native speakers of English have.
Key words:
Attitude, English language, accent, variety, personality attributes,
RP, non-standard English English, standard Irish English, nonstandard Irish English, EEAW, Osgood rating scale.
Acknowledgements:
This essay would never have come to existence without the help of
a great number of people. I should want to thank the four speakers
who gave “little pieces of their persons”. Such a thing indeed takes
courage and confidence. I should also like to thank all the
informants who have offered time and energy and especially let me
know their attitudes to some accents of the English language.
Additionally, I would like to thank Dr James Green at University
of Otago, New Zealand, and Dr Kirk Sullivan at Umeå University,
who have both provided me with information about the EEAWstudy. Last but not least, I also want to thank my supervisor Dr
Peter Sundkvist for his feedback and support in the course of the
entire process.
Table of contents
Page number
1. Introduction
1
2. Aims and questions at issue
2
3. Background information
3.1 Previous studies
3.1.1 Previous methods
3.1.2 Previous results
3.2 Hypotheses
3
3
3
4
5
4. Some definitions
5
5. Some information about the participants
5.1 The accents
5.1.1 Accent 1: RP
5.1.2 Accent 2: Non-standard English English
5.1.3 Accent 3: Standard Irish English
5.1.4 Accent 4: Non-standard Irish English
5.2 The informants
6
6
6
6
7
7
8
6. Method
6.1 Advantages and disadvantages with the method
8
9
7. Results
7.1 Total results
7.1.1 The Osgood rating scale
7.1.1.1 RP vs. non-standard English English
7.1.1.2 Standard Irish English vs. non-standard Irish English
7.1.1.3 RP vs. standard Irish English
7.1.1.4 Non-standard English English vs. non-standard Irish English
7.1.1.5 Relative position order - a general illustration
7.1.2 Social class
7.1.3 Identification of accent
11
11
11
12
12
13
13
13
14
15
8. Discussion
16
9. Conclusion
17
10. References
19
11. Appendices
20
1. Introduction
The way in which we speak has an effect on other people. Accordingly, it is not just
what you say, but also how you say it that creates meaning:
The reason why many American citizens would not
have wanted, and probably still don’t want, to buy a
new car from Richard Nixon is not because of the
things he said but because of the way he said them
(Scherer 1979:88, cited Harrison 1974, cited Kraus
1962).
Attitudes to varieties of language seem to be very much dependent on experience.
According to Holmes, attitudes are not due to inherent features of language: “When
people listen to accents or languages they have never heard before, their assessments
are totally random. […]. In other words there is no universal consensus about which
languages sound most beautiful and which most ugly” (Holmes 1992:346). However,
it should be mentioned that inherent features still might affect people’s attitudes. Thus
far, we do not know to what extent that is.
It is generally understood that Swedish people are familiar with the English
language. Nevertheless, they are rarely as familiar with the English language as native
speakers of English are. Most Swedish people speak some unspecified variety of
English. Moreover, drawing on personal experiences, many Swedish people can
distinguish at least some varieties of English from other varieties of English.
Therefore, if Swedish people were to be exposed to different varieties of English,
there might be seen that they have different attitudes to the different varieties.
Possibly, these attitudes would be in line with the attitudes that native speakers of
English have. However, the Swedish people’s attitudes might be different, that is due
to lack of experience of the English language.
Swedish people have had different amounts of exposure to different varieties of
English. What is more, they are expected to have heard the different varieties in
different kinds of situations. For instance, they have normally been taught RP in
school and they have possibly heard RP and some non-standard English English
varieties on television. Swedish people could be expected to have heard RP in news
broadcasts, in documentaries and comparable programs, whereas non-standard
English English varieties in TV-programs such as English soap operas. Irish varieties,
1
however, should be rather unfamiliar to those Swedish people who have not visited
Ireland or who have not a particular interest in the country. Nevertheless, they could
also have been exposed to some Irish varieties on television and in other media. These
different exposures, differing in both quantity and quality, are highly probable to
affect Swedish people’s attitudes. The question is whether this will be seen as a result
of this study.
2. Aims and questions at issue
The main aim of this essay is to investigate whether non-natives, still with some
knowledge of the speakers’ accents, have attitudes to the accents that are along the
lines of native informants’ attitudes. The accents studied here are RP, non-standard
English English, standard Irish English and non-standard Irish English. The
representatives for each of the accents are presented further in section 5.1. Please note
that we focus on accents and not dialects. Therefore, we should here consider nonstandard English English an accent and not compare it with Standard English, which
is a widely recognised dialect of English. RP, however, is spoken by many Standard
English speakers. The definitions of the terms dialect and accent are clarified in
section 4.
More specifically, this essay aims to present what personality attributes Swedish
people ascribe to the different accents and whether these personality attributes differ
especially concerning the four relations below:
-
RP vs. non-standard English English
-
Standard Irish English vs. non-standard Irish English
-
RP vs. standard Irish English
-
Non-standard English English vs. non-standard Irish English
In addition, this essay aims to give a wider perspective on how important experience
of varieties of language is, in order to have systematic attitudes to the varieties.
One reason for choosing English English and Irish English is availability. Another
reason is personal interest in the two countries and their language varieties. What is
advantageous with English English and Irish English is also that we get one standard
accent (RP), which should be fairly recognisable to Swedish people, and another
2
standard accent (the Irish), which should not be very familiar. None of the nonstandard accents is very likely to be widely recognised by Swedish people. A question
is whether this will be seen as a result of this study.
3. Background information
3.1 Previous studies
In 1927, the American linguist Edward Sapir published the article “Speech as a
personality trait”. This article is “not explicitly founded on any experimentation; it
could be seen as a kind of programme tract for much of the work that has since been
carried out, not least within the field of sociolinguistics” (Mobärg 1989:5). Additional
works were then produced during the next few decades. It was, however, not until the
1960s that attitudinal studies really started to gain ground. Wallace E. Lambert and his
colleagues published several very influential articles during the 1960s and 1970s
(Mobärg 1989:14).
3.1.1 Previous methods
Two different methods have mainly been used for measuring attitude to accents
(Gallois and Callan 1981:348). One is the method called the “matched-guise
technique”. This technique was what mainly made Lambert’s studies special as
compared to previous studies (Mobärg 1989:14). In such an experiment, perfectly
bilingual or bidialectal speakers record the same passage in the two varieties they
command. Their voices are then often mixed with “filler voices”, which are there so
that the informants find it hard to understand that they hear the same voices twice.
The advantage with this method is of course, at least theoretically, that you will hear
two different language varieties, but with the same voice features.
A second technique, and the most common used, has been to employ several
speakers from each accent group. This method was used in the EEAW-project
(Evaluating English Accents Worldwide) (University of Otago 1999) and it will also
be used in this study with one exception; we will here only use one speaker of each
accent.
3
Many researchers have used what is called an Osgood rating scale (see e.g. Chiba
et al. 1995) and it is based on bipolar terms, for instance unreliable-reliable. It was
used in the EEAW-study and it will also be used here.
3.1.2 Previous results
When accents are familiar to informants they may stand for different values. For
instance, it has been shown that an accent regarded as a standard accent of a given
language is valued differently from a non-standard accent of the same language.
Various studies have found that nonstandard or
foreign-accented speakers are rated less favorably
on personality attributes (Callan et al 1983:407).
RP users of ‘proper’ English are, for instance,
commonly credited with greater levels of
intelligence, authority and self confidence, whereas
speakers with rural accents are conversely assumed
to be more friendly, more sympathetic and more
good-natured, as well as less authoritative
(Mugglestone 1995:59).
Janet Holmes also takes the listeners’ age into consideration:
The local accent (as compared to RP) is likely to be
less highly rated than the standard accent, especially
by older listeners, and especially on the status
related features such as confidence, ambition,
intelligence, leadership skills, high education and
high status job. Local accents generally gain higher
ratings on solidarity related features such as
sincerity, friendliness, reliability, and sense of
humour, and from young people (Holmes
1992:351).
Nearly all studies performed in the field have made use of native speakers, as well as
native informants of the specific varieties. Hence, there are not many results with for
instance native speakers and non-native informants.
As far as Sweden is concerned, there seems only to have been one study of
Swedish people’s attitudes to different accents of English. This is the EEAW-study.
The EEAW-study focused on widely recognised varieties of English. These are
4
English English, North American English, Australian English and New Zealand
English. EEAW employed one male and one female speaker for each of the varieties
and it used 60 male and 54 female informants from Umeå University. Other variables
that could be considered are speaker’s and informant’s age, level of education,
occupation and the informant’s recognition of the speaker’s accent etc.
3.2 Hypotheses
Since Swedish people are familiar with the English language and expectedly some of
the accents in this study, the Swedish informants were expected to have attitudes
comparable to those of native informants. However, their attitudes were likely to be
less systematic; this is due to the informants’ exposures to the different accents
differing in both quantity and quality and also differing to natives’ exposures to the
accents.
An earlier assumption was that the RP speaker would gain higher ratings on
personality attributes such as ‘high level of education’, ‘intelligent’ and
‘authoritativeness’, whereas the non-standard English English accent would be
regarded as more ‘friendly’, ‘warm’ and ‘reliable’, in comparison. The relative
attitudes to the Irish accents were not expected to be as systematic, just like the
relative attitudes between RP vs. standard Irish English, and non-standard English
English vs. non-standard Irish English.
4. Some definitions
Technical terms are important tools when we want to be sure that we all refer to
exactly the same features out in the world. However, these terms can be troublesome,
and although we strive in order to make them as clearly defined as possible, we may
still think differently about them. In view of that, let us turn to some definitions of
dialect, accent and variety. Although the term accent is of most importance in this
essay, it may help to have all these definitions at hand.
First of all, we must be able to distinguish dialect from accent, where the former
concerns vocabulary, grammar, as well as pronunciation; the latter is only to do with
pronunciation. If we do not want to talk specifically about one of these terms linguists
commonly use the superordinate term variety: “kind of language” (Trudgill 2000:5).
This essay aims to deal with accents only. Again, accents are uniquely to do with
5
pronunciation, not vocabulary and grammar. However, we should still be aware that a
certain pronunciation may be a variety-dependent realisation of a grammatical feature.
5. Some information about the participants
5.1 The accents
The four accents will now be presented in connection with a brief description of each
of the four speakers who were employed to represent them. The triangle next to each
of the accents refers to the colour-matching triangle on the map, in Fig 1 below.
Subsequently, the triangle on the map corresponds to the regional origin of the
specific speaker.
5.1.1 Accent 1: RP
Received Pronunciation or RP is also commonly, yet more popularly, referred to as
Queen’s English, Oxford English or BBC English. RP is “the accent which developed
largely in the residential, fee-paying English ‘Public Schools’ favoured by the
aristocracy and the upper-middle-classes, at least for their sons, and which was until
quite recently required of all BBC announcers” (Trudgill 2000:7). It is also the
English accent most commonly taught to foreign learners, for instance in Sweden.
The representative for RP in this study is a 56 year old lecturer at Stockholm
University. He is originally from southwest England, where he was educated at
private schools. This is the area that is marked in Fig 1. Nevertheless, RP is not
geographically related to any specific area; it has speakers mainly in England, but also
in Australia, New Zealand and South Africa (Trudgill 2000:7-8). About 3 % of the
British population speak RP.
5.1.2 Accent 2: Non-standard English English
The representative for non-standard English English in this study is a 37 year old
student at Stockholm University. He is originally from Blackpool in northern England
and although his accent is not as strong anymore, it still differs very much from RP.
For instance, it has quite a different intonation pattern and it uses the /Υ/-vowel in
words of the STRUT class (Wells 1982:131): ‘something’, ‘utterly’. This is one
widely recognised feature of northern English varieties.
6
5.1.3 Accent 3: Standard Irish English
The representative for standard Irish English in this study was born in Donadea
Demesne, Donadea, Naas, Co. Kildare. He is from County Kildare, which is in the
eastern province of Ireland (Leinster), just west of Dublin County. He is 29 years of
age and he is currently employed at the Irish Embassy in Stockholm, Sweden.
5.1.4 Accent 4: Non-standard Irish English
The representative of non-standard Irish English is from Cork, south Ireland. He is
currently a 40 year old student at Stockholm University. His accent was not expected
to be recognised among the informants, perhaps apart from that it was of Irish origin.
Many of the informants thought this accent sounded rather odd.
The British Isles
Figure 1.
Above is a map to represent the British Isles. The four speakers are marked by the
coloured triangles, based on regional origin.
7
5.2 The informants
The informants are 31 Swedish females between the ages of 19 and 56. However, 28
of them are between 19 and 34, and 19 informants are between 22 and 26. The mean
age is 27.3 and 24 is both the median age and the mode age. More information on the
informants’ ages is presented in Appendix F.
A great majority, 29/31 of the informants, are currently studying or have studied at
institutions of higher education. Some are language students and some have studied
economy, psychology or mathematics among other disciplines. None of the
informants have studied at the English Department at Stockholm University and none
of them expressed recognition of any of the specific speakers.
6. Method
First of all, it should be acknowledged that the method used in this study is the
method that was also used in the EEAW-study, especially as we here similarly
employ the Osgood rating scale and the EEAW-presentation of results.
Four accents of English were recorded (see section 5.1 for accent details). They all
read the same text. This text has been used several times in studies like this one and it
originates from Wells (1982). The text is presented in Appendix E. Subsequently, 31
Swedish female informants were asked to state their attitudes to the accents, by
listening to the four accents and filling out a questionnaire. The full questionnaire is
presented in Appendix D. First, they would listen to one accent and then they would
get an extra minute to fill out the following:
This speaker gave the impression of being:
Reliable
Ambitious
Humorous
Authoritative
Competent
Cheerful
Friendly
Dominant
Intelligent
Assertive
Controlling
Warm
Hardworking
Not at
all
1
1
1
1
1
1
1
1
1
1
1
1
1
Very
2
2
2
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
3
3
3
8
4
4
4
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
5
5
5
6
6
6
6
6
6
6
6
6
6
6
6
6
The person's speech was:
Pleasant
Attractive
Powerful
Strong
Educated
Not at
all
1
1
1
1
1
Very
2
2
2
2
2
3
3
3
3
3
4
4
4
4
4
5
5
5
5
5
6
6
6
6
6
This scale, used by many researchers, is the previously mentioned Osgood rating
scale. For instance, on the first personality attribute ‘reliable’, they would circle ‘1’ if
they thought the speaker was not at all reliable and ‘6’ if they thought the speaker
sounded very reliable and so forth. Some of the attributes are perhaps not clearly seen
as personality attributes. However, this should just be born in mind since we will not
make any such distinction. Additionally, they were to state what social class they
thought the speaker belonged to and they were also asked to speculate freely about the
speaker’s regional origin. After they had listened to the first accent they were asked to
listen to and do the same thing with the three remaining accents, one after the other.
The questionnaire was produced in English but with the closest Swedish equivalent
stated in direct relation to the English word.
Half of the informants (16) listened to the accents in the following order: accent 4,
3, 1, and 2. The other half (15) listened to the accents in the order: 1, 3, 2, and 4. This
would reveal if the order of the accents would have any major impact on the results.
Henceforth, this will be referred to as the order of presentation (OOP).
The results will first be presented in section 7 and then discussed in section 8.
Social class and recognition of accent will be discussed in direct relation to the total
results.
6.1 Advantages and disadvantages with the method
The reason for only recording males and have only females as informants was simply
due to time pressure. It would have been too many variables to analyse with both
males and females recorded as well as male and female informants. The result of this
is that we have not got any results with female voices and no results with male
informants. Subsequently, we will not be able to lump any other gender-based groups
together, in order to present representative results from the Swedish population as a
9
whole. Instead, however, we have allowed ourselves to give a deeper analysis of the
more specific results that can now be achieved within the same period of time.
A limitation of this study might be that of the speakers’ individual voice features
and reading techniques. The speakers all have different voice features and reading
techniques, not only different accents. Age is one factor that could evidently
contribute to different voice features. By using the matched-guise technique this
problem could have been solved. However, as Gallois and Callan argue (1981:349),
the matched-guise technique might employ “feigned accents which may really only
represent the speaker’s stereotypes”. The EEAW-project also favours the method used
in this study: “[i]t seemed […] impossible to find speakers of each gender who could
give convincing renderings of all four accents without falling into the trap of
projecting a stereotype rather than the genuine article”.
The Osgood rating scale might seem problematic and also the fact that only the
means of the results are analysed. The scale makes use of an ordinal level of
measurement, which means that “…it is not possible to claim that there are truly equal
intervals between the points on the scale…” (Butler 1985:12). The means of the
results do not reveal if half of the informants gave the rate ‘1’ and the other half gave
the rate ‘5’; the mean would consequently be 3.0. This was also the procedure used in
the EEAW-study and despite the implications mentioned it still gives valuable results.
The speakers’ styles are likely to be quite formal in reading and may not reflect
normal conversational speech. Normal conversational speech will consequently be left
outside the scope of this investigation.
The fact that the four speakers all read the same text would make it hard for them
to reveal any obvious differences in grammar and/or vocabulary, which was not an
aim of this investigation.
The informants were to answer an open question about the origins of the four
accents. This would allow them to speak freely about this question and not just state
whether they thought it was for instance an English English or Irish English accent
that they heard. This method was not used in the EEAW project. Nevertheless, it
seemed more suitable here since the geographical variance was not as great. Please
note that the informants were not told anything about the accents beforehand, except
that they were accents of the English language.
Having the questionnaire in English and Swedish would result in the different
words not losing too many connotations and/or semantic features, as could be the case
10
with a purely Swedish questionnaire. Naturally, if the questionnaire was produced in
English only, this could also lead to misunderstandings. None of the informants
expressed any difficulties understanding the questionnaire. Of course, however, this is
not evidence that no one had this kind of problem.
7. Results
7.1 Total results
The total results will be now presented and commented on. The results will
subsequently be discussed and analysed in section 8 and 9. We will start with the
ratings on personality attributes, where the Osgood rating scale was used. After that,
social class and identification of accent will be presented and commented on. More
detailed results are presented in Appendix A.
7.1.1 The Osgood rating scale
Generally and as can be seen from this study, Swedish people seem to prefer standard
varieties to non-standard varieties of the English language. This study, however, also
shows that Swedish people have different attitudes to different personality attributes
and for different accents. Below, the total results are seen in two linear diagrams. This
way of presenting the results was also used in the EEAW-study. The informants are
divided into two groups depending on OOP. The OOP of the accents for group A is
accent 4, 3, 1, 2 (Fig 2). The OOP for group B is accent 1, 3, 2, 4 (Fig 3).
Figure 2.
1. RP
2. Non-standard English English
3. Standard Irish English
4. Non-standard Irish English
m
or
ki
ng
sa
nt
At
v
tr a
oi
ce
ct
iv
e
Po
vo
w
ic
er
fu e
lv
St
oi
ro
ce
n
Ed
g
uc voi
c
at
ed e
vo
ic
e
W
ar
dw
Pl
ea
H
ar
R
el
ia
Am ble
bi
tio
u
H
um s
o
Au
ro
us
th
or
ita
ti
C
om ve
pe
te
nt
C
he
er
fu
l
Fr
ie
nd
l
D
om y
in
an
In
t
te
l li
ge
As nt
se
rti
C
on ve
tro
lli
ng
6
5,5
5
4,5
4
3,5
3
2,5
2
1,5
1
Means of personality attributes for OOP-group A (accent 4, 3, 1, 2).
11
Figure 3.
1. RP
2. Non-standard English English
3. Standard Irish English
4. Non-standard Irish English
m
A
R
el
ia
bl
e
bi
t
H iou
um s
o
A
ut rou
ho
s
ri
C tati
o m ve
pe
te
C nt
he
er
Fr ful
ie
nd
D
o m ly
in
In an
t
te
llig
e
A
ss nt
e
C rtiv
on
e
tro
ll i
ng
H Wa
ar
rm
P dw o
le
r
as
ki
A an ng
ttr
tv
ac
oi
ti
c
P ve e
ow vo
ic
er
fu e
S l vo
tr
i
E ong ce
du
v
ca
o
te ice
d
vo
ic
e
6
5,5
5
4,5
4
3,5
3
2,5
2
1,5
1
Means of personality attributes for OOP-group B (accent 1, 3, 2, 4).
7.1.1.1 RP vs. non-standard English English
RP is rated higher than non-standard English English on most personality attributes
and for both OOP-groups. In OOP-group A (Fig 2), however, the ratings are much
closer to each other on attributes such as ‘humorous’, ‘cheerful’, ‘friendliness’ and
‘dominant’. ‘Warm’ is even rated a little higher for non-standard English English.
This is all seen where the lines in the diagram are either coming closer to each other,
or as with ‘warm’, crossing each other.
In OOP-group B (Fig 3) the same five attributes are either rated a little lower for
RP compared to non-standard English English, or get a very similar rating. However,
in this OOP-group attributes such as ‘assertive’, ‘competent’, ‘controlling’,
‘hardworking’, ‘pleasant voice’, ‘attractive voice’, ‘powerful voice’, ‘strong voice’
are also rated quite similarly between the accents.
7.1.1.2 Standard Irish English vs. non-standard Irish English
OOP-group A rate standard Irish English higher on most attributes. Nevertheless,
attributes such as ‘humorous’, ‘cheerful’, ‘dominant’, ‘controlling’ and ‘assertive’ are
rated more similarly between the accents and ‘hardworking’ is rated higher for nonstandard Irish English.
OOP-group B rate standard Irish English higher on all attributes. As can be seen in
the linear diagram, however, the ratings are more similar between the accents on
attributes such as ‘cheerful’, ‘friendly’ and ‘hardworking’, especially.
12
7.1.1.3 RP vs. standard Irish English
OOP-group A rate RP higher than standard Irish English on most attributes. However,
attributes such as ‘reliable’, ‘ambitious’, ‘competent’, ‘friendly’, ‘dominant’,
‘intelligent’, ‘pleasant voice’ and ‘attractive voice’ are rated more similarly between
the accents. Additionally, standard Irish English is rated a little higher on the attribute
‘warm’.
OOP-group B rate RP higher than standard Irish English on most attributes.
However, attributes such as ‘humorous’ and ‘hardworking’ are rated more similarly
between the accents. Standard Irish English is rated a little higher on the attributes
‘reliable’ and ‘attractive voice’.
7.1.1.4 Non-standard English English vs. non-standard Irish English
OOP-group A rate non-standard English English higher than non-standard Irish
English on most attributes. However, attributes such as ‘reliable’, ‘dominant’,
‘intelligent’, ‘assertive’, are rated similarly between the accents. Additionally, nonstandard Irish English is rated quite distinctively higher on the attribute ‘hardworking’
and a little higher on ‘controlling’.
OOP-group B rate non-standard English English higher than non-standard Irish
English on all attributes except one, ‘reliable’. ‘Reliable’ is rated very similarly
between the accents. Attributes such as ‘friendly’ and ‘hardworking’ are also rated
quite similarly between the accents.
7.1.1.5 Relative position order - a general illustration
Below is a figure to represent the relative position order between the four accents, on
each of the personality attributes. The function of this diagram is just to be a very
general illustration of the results. It is important to realise that this is a comparison
between all four accents and just because one accent is positioned first on a particular
personality attribute, it is not certain that it is evaluated much higher than the accent
which is positioned as number three or four. In addition, this diagram shows an
average rating on both OOP-groups.
13
Figure 4.
2. Non-standard English English
4. Non-standard Irish English
A
R
el
ia
bl
m e
bi
H tiou
um s
o
A rou
ut
ho s
C rita
om t
p ive
C ete
he n
e t
Fr rful
ie
n
D dly
om
i
In n an
te t
lli
A g en
ss
er t
C tive
on
tr
W olli
a r ng
m
H
ar
dw
P or
le k
as in
g
A an
ttr t
ac
P t iv
ow e
e
S rfu
tr o l
n
E g
du
ca
te
d
1. RP
3. Standard Irish English
Relative position (1-4)
1
2
3
4
It can clearly be seen that RP is rated higher than non-standard English English; this is
on most personality attributes. The same result is seen between the two Irish accents.
RP is rated higher than standard Irish English on nearly all personality attributes, just
like non-standard English English is rated higher than non-standard Irish English on
nearly all personality attributes.
7.1.2 Social class
The informants were asked to state which social class they thought each of the
speakers belonged to. The means can be seen in Fig 5 below.
Figure 5.
1 – lower class. 2 – lower-middle class. 3 – middle class. 4 – upper-middle class. 5 – upper class.
OOP: 4312
OOP: 1324
5
4,5
4
3,5
3
2,5
2
1,5
1
1. RP
2. Non-standard
English English
3. Standard Irish
English
Estimate of social class.
14
4. Non-standard
Irish English
RP and non-standard Irish English show the most dissimilar results, whereas the other
two accents both generally get ratings around ‘middle class’. More detailed
information concerning the results is presented in Appendix B.
7.1.3 Identification of accent
The informants were asked to speculate freely about the regional provenance of each
accent. In Fig 6 below are the correct identification percentages for each accent. There
are very similar results between the two OOP-groups. Consequently, they are here
lumped together in a single representation. More detailed results are presented in
Appendix C.
Figure 6.
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
1. RP
2. Non-standard
English English
3. Standard Irish
English
4. Non-standard
Irish English
Correct identifications of the respective accents.
As for RP answers like ‘Great Britain’, ‘London’, ‘England’ and ‘south England’
were regarded as correct. An answer like ‘northern England’ was, however, seen as
incorrect. Although ‘London’ is not a perfect answer, it still shows that the informant
recognises the accent as being one of standard kind; the accent of the capital is very
often seen as similar to the standard accent of the specific country.
As for non-standard English English, ‘Great Britain’, ‘England’ and ‘northern
England’ were answers regarded as correct. Answers like southern England or
Scotland were seen as incorrect. We could have based correct answers on other
criteria. However, the results would not have been very different; it is rather clear that
RP is the most recognised accent, followed by non-standard English English. The
only answer to the two Irish accents which could be seen as correct was Ireland/Irish.
None of the informants was able to specify any of these accents further.
15
8. Discussion
Swedish people are taught English from very early ages and the most widely taught
variety is recognised as skolengelska, ‘school English’. This refers to a variety with a
pronunciation close to RP. Tape recordings for educational purposes are produced in
RP and teachers very often favour English English accents, which is probably a result
of them having been taught RP at young ages. Naturally, Swedish people come to
associate RP with education and all that comes with it. Apart from English English
varieties, Swedish people are also exposed a lot to American varieties. American
English was, however, not a variety investigated in this study. Instead, the informants
listened to varieties they have rarely or perhaps never been exposed to, except for RP
which in this study had a 74 % identification score. Consequently, the total results are
not as systematic as results with natives.
If we look at RP and non-standard English English, many personality attributes
associated with a non-standard accent among natives are not rated as far away from
RP on the scale, as for instance ‘high education’ and ‘authoritative’ are. ‘Humorous’,
‘cheerful’, ‘friendliness’ and ‘warm’ are four of these attributes.
When we look at standard Irish English and non-standard Irish English, ‘cheerful’
and ‘hardworking’ are two attributes that are rated more along the lines of natives’
results than other attributes are. Again, ‘high education’ and ‘authoritative’ are rated
distinctively higher for the standard accent.
RP is rated distinctively higher on ‘high education’ and ‘authoritative’ than the
standard Irish accent. Similar ratings between these accents are seen as for ‘reliable’
and ‘attractive voice’.
No distinct analysis can be made comparing non-standard English English and
non-standard Irish English. However, ‘reliable’ and ‘hardworking’ are two attributes
that do not seem to be much more associated with the non-standard English English
than the non-standard Irish English.
Conclusively, ‘high education’ and ‘authoritativeness’ are two attributes that the
Swedish informants always associated with a standard- and/or an English English
accent. In the light of the 74% identification score for RP and knowing that RP in
Sweden is associated with education, it does not come as a surprise that ‘high level of
education’ and ‘authoritativeness’ show the most systematic results in this study. The
fact that this includes the relation between the Irish accents is also very interesting.
16
Perhaps the informants look upon the standard Irish accent as more similar than nonstandard Irish English to other standard accents of English. Similarly, but not as
distinctively, ‘cheerful’ seems to be the personality attribute most consistently
associated more with a non-standard accent than other attributes are, although it does
not constitute a perfect example.
In order to see whether this kind of study was recommendable or not, two
different OOP:s were applied. This would hopefully reveal whether the OOP had had
any impact on the results. It turned out that the OOP has a large impact. When the
informants heard RP first, it got lower scores. This is a result with 16 out of the total
18 personality attributes. As for non-standard Irish English this is the case with 17
features out of 18, with the difference that it gets higher scores if listened to initially.
In other words, the accent listened to initially gets ratings closer to the middle of the
scale, since there is no accent to weigh against. Subsequently, if the same accent is at
another position in the OOP, it will be rated more towards the end of the scale where
it generally seems to belong. On account of this, many results of this study cannot be
analysed very deeply. However, the results discussed above have been analysed
owing to their consistency across the OOP-groups.
Clearly, Swedish people are not very familiar with all accents of England and
Ireland. Hence, their assessments are less systematic than results for native
informants. Still, however, Swedish people have some knowledge of English and they
do consistently associate ‘authoritative’ and ‘high education’ with someone who has a
standard accent. This is independent of OOP. Nevertheless, it should also be
mentioned that some of the recognised features could be from other varieties of
language; that is Swedish and German among other varieties. It appears, however,
impossible to say to what extent that is.
9. Conclusion
The main aim of this essay was to reveal patterns as regard Swedish people’s attitudes
to four accents of the English language. It is clear that different accents are related to
different values. Nevertheless, experience of the particular accents is of great
importance and should be taken into consideration when analysing results from a
study like this. Previous studies reveal that native speakers of English associate
standard English varieties with high education and authoritativeness. Similarly, they
17
associate cheerfulness and reliability with non-standard accents. Swedish people,
however, give the impression of associating all of these four personality attributes
rather with standard accents than with non-standard accents. However, as for nonstandard accents, the two latter attributes seem to get ratings that are closer to its
standard accent, if compared to the two former attributes.
As a consequence of the informants’ previous exposures to the accents differing in
both quantity and quality, they do have less systematic attitudes to the accents in this
study than natives would probably have. This is all in line with Holmes’ belief that
unfamiliar varieties do not appear to reveal any patterns in informants’ values.
However, the accents in this study are not totally unfamiliar to the informants.
Evaluating varieties is not a straightforward process. Many things can have an
impact on a person’s evaluation and sometimes people do not know what to answer,
although they recognise the variety they are being exposed to. Nevertheless, several
studies within the field show similar results and some results of this study correspond
to results of previous studies. Therefore, if we make use of this study in conjunction
with other studies, we should be able to attain a wider picture of how varieties are
evaluated; this is dependent on the characteristics of the speakers and the informants
and especially on the informants’ experiences of the varieties they are evaluating.
18
10. References
Butler, Christopher. (1985). Statistics in Linguistics. Oxford: Basil Blackwell Ltd.
Callan, Victor. J et al. (1983). ‘Evaluative Reactions to Accented English: Ethnicity,
Sex, Role and Context’. Journal of Cross-Cultural Psychology: 407. Beverly
Hills, California: Sage.
Chiba, Reiko et al. (1995). ‘Japanese Attitudes Toward English Accents’. World
Englishes: 77-86. Oxford: Pergamon.
Crystal, David. (1997). The Cambridge Encyclopedia of Language (2 nd edition).
Cambridge: Cambridge University Press.
P
P
Deumert, Andrea., Leap, William. L., Mesthrie, Rajend and Swann, Joan. (2000).
Introducing Sociolinguistics. Edinburgh: Edinburgh University Press.
Gallois, Cynthia and Callan, Victor. (1981) ‘Personality Impressions Elicited by
Accented English speech’. Journal of Cross-Culural Psychology: 347-49. Beverly
Hills, California: Sage.
Holmes, Janet. (1992). An Introduction to Sociolinguistics. New York: Longman
Group UK Limited.
Hudson, Richard Anthony. (1980). Sociolinguistics. Cambridge: Cambridge
University Press.
T
T
Mobärg, Mats. (1989). English Standard Pronunciations: A Study of Attitudes.
Gothenburg: Acta Universitatis Gothoburgensis.
Mugglestone, Lynda. (1995). Talking Proper: The Rise of Accent as Social Symbol.
New York: Oxford University Press Inc.
Scherer, Klaus R. (1979). Language and social psychology: Voice and speech
Correlates of Perceived Social Influence in Simulated Juries. Great Britain: Basil
Blackwell Publisher Limited.
Trudgill, Peter. (2000). Sociolinguistics (4 th edition). London: Penguin books.
P
P
University of Otago. (1999). Evaluating English Accents Wordwide [Online].
Available: http://www.otago.ac.nz/anthropology/Linguistic/Accents.html. [Access
date: 2005, February 25].
Wells, John Christopher. (1982). Accents of English 1: An introduction. Cambridge:
Cambridge University Press.
19
11. Appendices
Appendix A
Below are diagrams for all four accents for all personality attributes and for both OOP-groups.
The different ratings for each accent, personality attribute and OOP-group are listed to the left
and next to it a line is drawn for each of the informants’ answers on that particular rating.
1. RP: OOP: 4, 3, 1, 2 (16 informants).
1. RP: OOP: 1, 3, 2, 4 (15 informants).
Reliable 4.25
1
2
3
lll
4
lll
5
lll
6
ll
Ambitious 4.125
1
2
l
3
lll
4
lll
5
lll
6
ll
Humorous 3.5
1
2
lll
3
lll
4
lll
5
lll
6
Authoritative 4.625
1
2
3
ll
4
lll
5
lll
6
lll
Competent 4.25
1
2
3
lll
4
lll
5
lll
6
Cheerful 3.6875
1
l
2
lll
3
lll
4
lll
5
ll
6
lll
Friendly 4.0625
1
2
3
lll
4
lll
5
lll
6
ll
Dominant 3.4375
1
l
2
3
lll
4
lll
5
l
6
Intelligent 4.125
1
2
3
lll
4
lll
5
lll
6
l
Reliable 3.3333
1
2
ll
3
lllllll
4
5
lllll
6
l
Ambitious (one missing) 3.7857
1
2
lll
3
lll
4
ll
5
llllll
6
Humorous (one missing) 2.7143
1
l
2
llllll
3
lll
4
llll
5
6
Authoritative (one missing) 3.7857
1
l
2
l
3
lll
4
lllll
5
lll
6
l
Competent (one missing) 3.7857
1
2
3
lllllll
4
lll
5
llll
6
Cheerful (one missing) 3.0
1
ll
2
llll
3
lll
4
ll
5
lll
6
Friendly (one missing) 3.7143
1
2
lll
3
ll
4
lllll
5
llll
6
Dominant (one missing) 3.1429
1
l
2
lll
3
lllll
4
lll
5
ll
6
Intelligent (one missing) 4.0
1
2
3
lll
4
llllllll
5
lll
6
l
lll
l
llll
ll
ll
ll
lll
lll
llll
l
lll
ll
llll
llll
l
llll
l
20
Assertive 3.875
1
l
2
l
3
lll
4
llll
5
llll
6
Controlling 3.6875
1
l
2
l
3
llll
4
llll
5
ll
6
ll
Warm 3.0
1
ll
2
lll
3
llll
4
5
ll
6
l
Hardworking 3.5625
1
2
ll
3
llll
4
llll
5
l
6
l
Pleasant 3.9375
1
l
2
l
3
lll
4
llll
5
llll
6
l
Attractive 3.4375
1
l
2
ll
3
llll
4
llll
5
ll
6
l
Powerful 3.625
1
2
ll
3
llll
4
llll
5
ll
6
l
Strong 3.6875
1
2
l
3
llll
4
llll
5
ll
6
l
Educated 4.6875
1
2
3
ll
4
llll
5
llll
6
llll
Assertive 3.2667
1
l
2
ll
3
llll
4
llllllll
5
6
Controlling (two missing) 3.5385
1
2
l
3
llllll
4
llll
5
ll
6
Warm (one missing) 3.7143
1
2
ll
3
llll
4
lllll
5
ll
6
l
Hardworking (one missing) 3.3571
1
2
lll
3
llll
4
llllll
5
l
6
Pleasant 4.0
1
2
ll
3
llll
4
ll
5
llllll
6
l
Attractive (one missing) 2.9286
1
l
2
llll
3
llllll
4
l
5
ll
6
Powerful 3.0667
1
2
llllll
3
llll
4
lll
5
ll
6
Strong 3.0667
1
l
2
ll
3
lllllllll
4
l
5
ll
6
Educated 3.9333
1
2
ll
3
ll
4
lllllll
5
lll
6
l
l
ll
ll
llll
ll
ll
l
l
ll
ll
l
lll
l
l
l
21
2. Non-standard English English.
OOP: 4, 3, 1, 2 (16 informants).
2. Non-standard English English.
OOP: 1, 3, 2, 4 (15 informants).
Reliable 3.4375
1
2
lll
3
lll
4
lll
5
l
6
l
Ambitious 3.125
1
2
lll
3
lll
4
ll
5
ll
6
Humorous 3.125
1
l
2
lll
3
lll
4
lll
5
l
6
l
Authoritative 2.875
1
l
2
lll
3
lll
4
l
5
l
6
l
Competent 3.3125
1
2
lll
3
lll
4
ll
5
6
ll
Cheerful 3.4375
1
2
lll
3
lll
4
lll
5
ll
6
l
Friendly 3.9375
1
2
3
lll
4
lll
5
lll
6
l
Dominant 3.1875
1
l
2
lll
3
lll
4
lll
5
l
6
l
Intelligent 3.1875
1
2
lll
3
lll
4
lll
5
ll
6
Assertive 3.375
1
l
2
lll
3
lll
4
ll
5
6
lll
Reliable 2.8666
1
2
lllll
3
lllllll
4
lll
5
6
Ambitious (one missing) 3.3571
1
2
lll
3
llll
4
llllll
5
l
6
Humorous 3.2667
1
l
2
lll
3
lll
4
lllllll
5
l
6
Authoritative 3.1333
1
2
lll
3
llllllll
4
lll
5
l
6
Competent 3.5333
1
2
ll
3
lllll
4
llllll
5
ll
6
Cheerful 3.6
1
2
llll
3
l
4
lllllll
5
lll
6
Friendly 3.7333
1
2
ll
3
llll
4
lllll
5
llll
6
Dominant (one missing) 3.2143
1
2
lllll
3
llll
4
ll
5
lll
6
Intelligent (one missing) 3.4286
1
2
ll
3
lllllll
4
ll
5
lll
6
Assertive 3.4
1
2
llll
3
llll
4
llll
5
lll
6
lll
ll
l
lllll
l
lll
lll
lll
llllll
ll
ll
lll
lll
llll
l
llll
llll
22
Controlling 2.75
1
ll
2
ll
3
ll
4
5
l
6
l
Warm 3.375
1
2
ll
3
ll
4
ll
5
l
6
l
Hardworking 3.0
1
l
2
ll
3
ll
4
ll
5
l
6
Pleasant 3.5625
1
l
2
ll
3
ll
4
ll
5
ll
6
l
Attractive 2.9375
1
2
ll
3
ll
4
ll
5
l
6
Powerful 2.9375
1
ll
2
ll
3
ll
4
ll
5
ll
6
l
Strong 3.1875
1
l
2
ll
3
ll
4
ll
5
ll
6
l
Educated 3.0625
1
l
2
ll
3
ll
4
5
6
ll
Controlling (one missing) 3.2143
1
2
llll
3
llll
4
lllll
5
l
6
Warm 3.4667
1
l
2
ll
3
llll
4
llllll
5
l
6
l
Hardworking 3.2
1
l
2
llll
3
llll
4
lll
5
lll
6
Pleasant 3.6667
1
l
2
l
3
llll
4
llllll
5
ll
6
l
Attractive 3.0667
1
2
lllll
3
llllll
4
ll
5
ll
6
Powerful (one missing) 3.0
1
2
lllllll
3
lll
4
l
5
lll
6
Strong 3.0667
1
2
llllll
3
lll
4
lllll
5
l
6
Educated 3.0667
1
2
lllll
3
lllll
4
llll
5
l
6
l
l
llllll
ll
lll
lll
l
llllll
l
l
ll
l
ll
llllll
lll
llll
l
lll
ll
l
l
llllllll
23
3. Standard Irish English
OOP: 4, 3, 1, 2 (16 informants).
3. Standard Irish English.
OOP: 1, 3, 2, 4 (15 informants).
Reliable 4.3125
1
2
3
lll
4
llllll
5
llllll
6
l
Ambitious 4.125
1
2
l
3
lll
4
lllll
5
lllllll
6
Humorous 2.9375
1
l
2
lllll
3
llllllll
4
ll
5
6
Authoritative 3.5625
1
2
l
3
lllllll
4
llllll
5
ll
6
Competent 4.25
1
2
3
lll
4
llllll
5
lllllll
6
Cheerful 3.0625
1
2
lllll
3
lllllll
4
ll
5
ll
6
Friendly (one missing) 4.2667
1
2
3
lll
4
lllll
5
lllllll
6
Dominant 3.3125
1
l
2
lll
3
lllll
4
lllll
5
l
6
l
Intelligent (one missing) 4.2667
1
2
3
lll
4
llllll
5
lllll
6
l
Assertive (one missing) 3.3333
1
2
lll
3
lllllll
4
ll
5
lll
6
Reliable 3.8667
1
2
ll
3
llll
4
lllll
5
ll
6
ll
Ambitious (one missing) 3.0
1
l
2
ll
3
llllllll
4
ll
5
l
6
Humorous (one missing) 2.7143
1
lll
2
llll
3
lll
4
ll
5
ll
6
Authoritative 3.0667
1
ll
2
l
3
lllllll
4
llll
5
l
6
Competent (one missing) 3.2143
1
2
llll
3
llllll
4
ll
5
l
6
l
Cheerful 2.5333
1
lll
2
lllll
3
llll
4
ll
5
1
6
Friendly 3.4
1
2
lll
3
lllll
4
lllll
5
ll
6
Dominant (one missing) 2.7143
1
l
2
llllll
3
llll
4
ll
5
l
6
Intelligent (one missing) 3.3571
1
2
ll
3
llllllll
4
ll
5
l
6
l
Assertive (one missing) 2.8571
1
2
lllllll
3
llll
4
ll
5
6
l
24
Controlling (two missing) 2.9231
1
ll
2
lll
3
llll
4
lll
5
6
l
Warm (one missing) 3.0
1
ll
2
lll
3
lll
4
lllll
5
l
6
Hardworking (two missing) 3.3846
1
l
2
ll
3
llll
4
llll
5
l
6
l
Pleasant (one missing) 3.2143
1
l
2
ll
3
lllll
4
lllll
5
l
6
Attractive (two missing) 3.3077
1
ll
2
lll
3
l
4
llll
5
ll
6
l
Powerful (two missing) 2.1538
1
ll
2
lllllll
3
llll
4
5
6
Strong (two missing) 2.5385
1
2
llllllll
3
llll
4
5
l
6
Educated (one missing) 3.0
1
2
lllll
3
lllllll
4
5
l
6
l
Controlling 3.3125
1
2
lll
3
llllllll
4
ll
5
lll
6
Warm (one missing) 3.375
1
l
2
l
3
llll
4
llllll
5
lll
6
Hardworking 3.3125
1
l
2
3
llllllllll
4
lll
5
ll
6
Pleasant 4.0625
1
l
2
3
lllll
4
ll
5
lllllll
6
l
Attractive 3.5
1
l
2
ll
3
lllll
4
llll
5
llll
6
Powerful 3.0625
1
2
lllll
3
lllll
4
llllll
5
6
Strong 3.1875
1
l
2
lll
3
llllll
4
llll
5
ll
6
Educated 4.0625
1
2
3
llllll
4
llll
5
lllll
6
l
25
4. Non-standard Irish English.
OOP: 4, 3, 1, 2 (16 informants).
4. Non-standard Irish English.
OOP: 1, 3, 2, 4 (15 informants).
Reliable 3.375
1
ll
2
l
3
lllll
4
lllll
5
lll
6
Ambitious 2.75
1
ll
2
llll
3
lllllll
4
ll
5
l
6
Humorous 2.8125
1
ll
2
llllll
3
llll
4
l
5
lll
6
Authoritative 2.4375
1
ll
2
llllllll
3
lll
4
lll
5
6
Competent (one missing) 3.0667
1
2
lll
3
llllllll
4
llll
5
6
Cheerful 2.75
1
llll
2
lllll
3
ll
4
l
5
llll
6
Friendly (one missing) 3.3333
1
2
lllll
3
lll
4
llll
5
lll
6
Dominant (one missing) 3.0667
1
ll
2
ll
3
lllll
4
lllll
5
l
6
Intelligent (one missing) 3.0667
1
l
2
lll
3
llllll
4
llll
5
l
6
Assertive 3.375
1
l
2
lll
3
llll
4
lllll
5
lll
6
Reliable 2.9333
1
2
llllll
3
lllllll
4
5
l
6
l
Ambitious (one missing) 2.7143
1
ll
2
lllll
3
llll
4
ll
5
6
l
Humorous (one missing) 2.2857
1
llllll
2
ll
3
lll
4
ll
5
l
6
Authoritative (one missing) 2.4286
1
ll
2
llllll
3
llll
4
ll
5
6
Competent (one missing) 2.6429
1
l
2
lllllll
3
lll
4
ll
5
l
6
Cheerful 2.2667
1
lllll
2
lllll
3
lll
4
l
5
6
l
Friendly 3.2
1
2
llll
3
lllll
4
lllll
5
l
6
Dominant (one missing) 2.0
1
lll
2
lllllllll
3
l
4
l
5
6
Intelligent 2.7333
1
ll
2
lll
3
llllllll
4
l
5
l
6
Assertive 2.3333
1
l
2
lllllllll
3
llll
4
l
5
6
26
Controlling (one missing) 2.2857
1
l
2
lllllllll
3
lll
4
l
5
6
Warm 2.6667
1
ll
2
llllll
3
llll
4
ll
5
6
l
Hardworking (one missing) 3.0714
1
l
2
lllll
3
ll
4
llll
5
ll
6
Pleasant 2.6667
1
lll
2
lll
3
llllll
4
ll
5
l
6
Attractive 2.0
1
lllllll
2
llll
3
ll
4
l
5
l
6
Powerful (one missing) 1.7143
1
llllll
2
lllllll
3
4
l
5
6
Strong 2.0667
1
lllll
2
lllllll
3
l
4
l
5
l
6
Educated 2.2667
1
ll
2
llllllllll
3
l
4
l
5
l
6
Controlling (two missing) 3.1429
1
ll
2
llll
3
l
4
llll
5
lll
6
Warm 2.625
1
llll
2
lllll
3
lll
4
l
5
lll
6
Hardworking (one missing) 4.0667
1
2
ll
3
llll
4
ll
5
lllll
6
ll
Pleasant 2.8125
1
lllll
2
lll
3
ll
4
lll
5
ll
6
l
Attractive 2.25
1
llllll
2
llll
3
lll
4
ll
5
l
6
Powerful 1.875
1
llllll
2
lllllll
3
ll
4
l
5
6
Strong 2.375
1
llll
2
lllll
3
llllll
4
5
6
l
Educated 2.375
1
lll
2
llllll
3
lllll
4
ll
5
6
27
Appendix B
Below are the ratings on social class for each of the accents. Hence, the quantity refers to the
number of informants who gave that specific answer. The distinction between OOP-group A
and OOP-group B is indicated by distinctive colours. OOP for group A is 4, 3, 1, 2. OOP for
group B is 1, 3, 2, 4.
1. RP
1
Social class
Upper class
4
Upper-middle
class
9
5
OOP-group A
4
Middle class
Lower-middle
class
6
OOP-group B
1
Lower class
0
2
4
6
8
10
Quantity
Non-standard English English
Social class
Upper class
Upper-middle
class
OOP-group A
Middle class
OOP-group B
Lower-middle
class
Lower class
0
2
4
6
8
10
Quantity
Social class
3. Standard Irish English
Upper class
1
Upper-middle
class
1
5
OOP-group A
7
7
Middle class
Lower-middle
class
4
Lower class
OOP-group B
5
1
0
2
4
6
Quantity
28
8
10
Social class
4. Non-standard Irish English
Upper class
1
Upper-middle
class
1
OOP-group A
6
Middle class
OOP-group B
Lower-middle
class
7
3
Lower class
0
2
8
4
4
6
8
10
Quantity
Appendix C
Below are all the informants’ estimates on the origins of the accents. The informants’
individual answers are separated by commas.
OOP: 4312.
1. RP.
England – London, södra England, London, Storbritannien, London, någon tjusig trakt i England - Oxford,
landsbygdssnobb, Oxford skolengelska, England – London, vet ej, Storbritannien, inget svar, England, England,
ingen aning, England.
2. Non-standard English English
England, norra/mellersta England, centrala England, Australien, Skottland, nä vet inte England?, inget svar,
London, Storbritannien (Irland?) England, vet ej, inget svar, inget svar, England, England, inget svar, England.
3. Standard Irish English
England stor stad typ London, Irland, södra England dock ej kustregion, Irland, London, Skottland, inget svar,
Wales kanske, England – London, Irland, Irland, ingen aning, Irland, storstad i England södra delen kanske
London, London, England.
4. Non-standard Irish English
Wales Manchester nånstans i Storbritannien men ute i bygden, Irland, norra England, Sydafrika, England,
Irland, norra England, Skottland nånstans, norra England, norrut landsbygd Storbritannien, Storbritannien
London?, ingen aning, Irland, norra England storstad, södra England, Storbritannien.
OOP: 1324.
1. RP.
England, Sydafrika, England, Storbritannien förmodligen England, Storbritannien utanför London, inget svar,
Irland, Europa (stad?), stor stad i England, England, London, England – London, någon by ute på landsbygden,
södra England nära storstad, England.
29
2. Non-standard English English
England, Storbritannien, inget svar, Storbritannien, inget svar, Sverige, Australien, Europa, Europa, England,
södra England, England, någon större stad i England, USA, England.
3. Standard Irish English
Irland, Irland el Skottland, inget svar, lät lite amerikanskt, vet ej, inget svar, Irland, Europa, förort i England,
Sverige, Irland, norra England, någon stad i USA, Storbritannien, Irland.
4. Non-standard Irish English
Skottland, Irland eller Skottland, Sverige, inget svar, Irland, inget svar, Irland, Europa, någonstans på landet i
Europa, Irland, Skottland, norra England – Manchester, ingen aning, Australien, Irland.
OOP-group A
Accent
4. Non-standard Irish
English
3
3. Standard Irish
English
5
2. Non-standard
English English
8
1. RP
13
0
2
4
6
8
10
12
14
16
Number of correct answers (maximum: 16)
Correct identification of the respective accents for OOP-group A. OOP: 4, 3, 1, 2.
Accent
OOP-group B
4. Non-standard Irish
English
5
3. Standard Irish
English
5
2. Non-standard
English English
7
1. RP
10
0
2
4
6
8
10
12
14
Number of correct answers (maximum: 15)
Correct identification of the respective accents for OOP-group B. OOP: 1, 3, 2, 4.
30
Appendix D
Below is the questionnaire that was used. Note that the informants were provided with three
more pages for the three remaining accents. These three pages are, however, identical to the
second page in this appendix.
Evaluating English Accents: Questionnaire
I den här studien är vi intresserade av dina intryck av ett antal talare. Ofta får vi ett intryck
av hur talaren är som person genom att lyssna till dennes röst, till exempel när vi hör en
främmande människa på radio eller i telefon. Vi kommer spela upp fyra inspelade röster för
dig. Varje talare kommer säga samma sak. Det är alltså meningen att du ska fokusera på hur
talaren låter och inte på vad han säger. Efter varje talare kommer du ha en minut på dig att
fylla i första delen av uppgiften genom att ringa in siffran som bäst överensstämmer med ditt
intryck av talaren. Till exempel om du tycker att talaren är mycket pålitlig (reliable) ringar du
in 5 eller 6, och om du tycker att talaren inte är alls pålitlig ringar du in 1 eller 2. Tänk inte
för länge på varje fråga, utan försök att ge oss ditt första intryck. Det finns inga rätta eller
felaktiga svar och det första intrycket är oftast mest överensstämmande med ditt intryck.
Om något är oklart, var god säg till nu!
Vänliga hälsningar,
Franz Myrman, Stockholms Universitet
Innan vi börjar skulle vi vilja ha lite information om dig. Var god fyll i alla uppgifter. Tack
på förhand!
Kön : ____________
Ålder: ____________
Tid som du spenderat i länder med engelska som modersmål. Namnge även dessa länder:
_________________________________________________________________
_________________________________________________________________
_________________________________________________________________
_________________________________________________________________
31
Speaker Number 1
Part 1
This speaker gave the impression of being / talaren gav intrycket av att vara:
Not at
all
Reliable / pålitlig
1
2
3
4
5
Ambitious / ambitiös
1
2
3
4
5
Humorous / humoristisk
1
2
3
4
5
Authoritative / auktoritativ
1
2
3
4
5
Competent / kompetent
1
2
3
4
5
Cheerful / gladlynt
1
2
3
4
5
Friendly / vänlig
1
2
3
4
5
Dominant / dominant
1
2
3
4
5
Intelligent / intelligent
1
2
3
4
5
Assertive / bestämd
1
2
3
4
5
Controlling / kontrollerande
1
2
3
4
5
Warm / varm
1
2
3
4
5
Hardworking / hårt arbetande
1
2
3
4
5
Very
6
6
6
6
6
6
6
6
6
6
6
6
6
The person's speech was / personens tal var:
Not at all
Pleasant / trevligt
Attractive /attraktivt
Powerful / mäktigt
Strong / starkt
Educated / välutbildat
1
1
1
1
1
Very
2
2
2
2
2
3
3
3
3
3
4
4
4
4
4
5
5
5
5
5
6
6
6
6
6
Part 2:
Answer each question by circling the response that is closest to your impression /
Besvara varje fråga genom att ringa in svaret som bäst överensstämmer med ditt intryck.
1. In what social class would you categorize the speaker?
I vilket socialskikt skulle du kategorisera denna talare?
1
2
3
Lower class
Arbetarklass
2.
lower-middle class
Lägre medelklass
Middle class
Medelklass
4
5
upper-middle class
Övre medelklass
Upper class
Överklass
Where is this person from? Please feel free to speculate:
Var kommer personen ifrån? Spekulera gärna:
__________________________________________________________
__________________________________________________________
__________________________________________________________
__________________________________________________________
32
Appendix E
Below is the text that the speakers read.
One day last year, when I was driving back to work
after I’d had lunch, I had an amazing and
unforgettable experience. It must have been two
o’clock — or perhaps a quarter of an hour later, a
quarter past two. It was an incredible thing, really.
I was sitting there at the steering wheel of my new
car, waiting for the lights to change, when all of a
sudden the car started to shake this way and that,
rocking from side to side, throwing me backwards
and forwards, up and down. I felt as if I was riding
a bucking horse. Worse than that, some mysterious
spirit or hostile force seemed to be venting its vast
fury upon the earth. And the noise! — there was a
kind of deep groaning and horrible awesome
grinding which seemed to fill the air. And then, a
short while after, the whole paroxysm had stopped,
just as suddenly. Everything was calm and smooth
again, quiet and peaceful once more. I put my foot
down, just a gentle pressure on the accelerator, or
the gas pedal, as it’s known in America, and drove
off. Everything was utterly normal once more. So
then, was this some very local and momentary
earth tremor which had struck us? Or, I asked
myself, was it a supernatural visitation, some fiery
storm of diabolical wrath? Or was it, rather, merely
that I’d drunk a double vodka or two during my
lunch?
T
T
Appendix F
1
55
52
49
46
43
1
1
Age
40
37
34
1
31
1
2
28
2
25
22
1
19
1
1
0
3
3
6
2
4
Quantity
Ages of informants.
33
6
7
8