Department of English

Bachelor Degree Project

English Linguistics

Spring 2021

Supervisor: Peter Sundkvist

World Englishes:

Attitude in the

Expanding Circle Towards East and

Southeast Asian Varieties of English

Laura Ortu

World Englishes: Attitude in the

Expanding Circle Towards East and Southeast Asian Varieties of English

Laura Ortu

Abstract

English has become an essential part in our lives. It is inevitable to formulate

an opinion when we meet a new person, and in particular we tend to focus our

attention on the way this person speaks. The present research aims to answer

the questions on how a European audience (Italian audience) perceives

different varieties of English to which it is exposed. Four different speakers

from four different Southeast and East Asian countries were selected and

recorded while reading a short text. These recordings were submitted to the

audience, which was asked to answer a set of questions about

comprehensibility and likability. Results show that the audience elected as

their favourite speaker the clearest accent to hear, thus suggesting that the

members of the audience might have been influenced by comprehensibility

and accentedness in the first place. Other variables, such as expressiveness,

were not significantly considered by the participants.

Keywords

Attitude, Asian accented English, comprehensibility, likability, accentedness.

Contents

1. Introduction ................................................................................... 2

1.1 Similar researches and gaps .............................................................. 3

1.2 The goal of the present research ........................................................ 3

1.3 Main features of the varieties of English analyzed................................. 4

1.1.1 Singaporean English ............................................................. 5

1.1.2 Japanese English .................................................................. 5

1.1.3 Vietnamese English ............................................................... 6

1.1.4 Chinese English .................................................................... 7

2.Methodology ................................................................................... 7

2.1 Matched Guise Technique ................................................................. 8

2.2 Speakers ........................................................................................ 8

2.3 Italian audience ............................................................................... 9

2.4 Set of questions .............................................................................. 9

3. Results .......................................................................................... 9

3.1 Part I ............................................................................................ 10

3.2. Part II .......................................................................................... 13

4. Discussion ..................................................................................... 16

4.1 Hypotheses .................................................................................... 16

4.2 The first thesis statement ................................................................ 17

4.3 Flows of the project ........................................................................ 18

5. Conclusion .................................................................................... 18

References ........................................................................................ 20

Appendix A ....................................................................................... 22

Appendix B ....................................................................................... 23

Appendix C ....................................................................................... 28

Appendix D ....................................................................................... 29

Appendix E ....................................................................................... 30

1

2

1. Introduction

English has become a fundamental tool in our lives. This language is used in several

fields, with specific vocabulary and various registers. However, it is important to

remember that there are numerous varieties of English. British English, American

English, Canadian English, Australian English: these are just some of the most

widespread varieties of World Englishes. It may also be interesting to consider how

these different accents are perceived: generally, British English sounds “snobbish”,

American English sounds “cool” (Kaur, 2014, p. 6), and so on. However, English has

different varieties outside the “Inner Circle”, or “the ‘owners’ of this language” as

Canagarajah (2013), while citing Kachru (1986) defines them (p. 58). It is important to

remember that the ownership of English is not only in the hands of the Inner Circle

users. The definition of “variety” is not straightforward, because even within the Inner

Circle there are local varieties. As Canagarajah states, “[a]ttitudes of ownership and

purity may inhibit NES engagement with other languages and communities, but cannot

leave their competence or norms unscathed from the results of such contact” (p. 60).

The “Expanding Circle”, which is formed by people who “didn’t come under British

colonization, but are using English as a foreign language for contact with other similar

countries” (Canagarajah, p. 58), is a solid reality that has to be considered while

studying features of English. Therefore, it is appealing to consider how these accents are

perceived.

As stated before, English is spoken in other communities whose speakers are not

considered “native”. In order to be economically competitive, some East Asian

countries have reached a high level of English and in some of them it has even become

a second language. Singaporean English, Chinese English, Vietnamese English or

Japanese English are just four new varieties of English that captured the interest of

phoneticians around the world. However, it is important to remember that some

phoneticians would accept Singaporean English as a variety of English, whereas some

others would refuse, for instance, Vietnamese English. This can be connected to the fact

that Vietnamese English is a “recent” variety of English; however, as stated in

Canagarajah, these “definitions of what is or isn’t a ‘variety’ shouldn’t prevent us from

acknowledging significant communicative practices” (p. 60). One of the main aspects

considered is accentedness, i.e. “a judgment of accent strength” (Lochland, 2020, p. 23).

Therefore, scholars have decided to discover how these accents were perceived inside

the countries where they were spoken. Tan and Tan (2008), for instance, determined

that Singaporean English is valued by its speakers, albeit not considered “fashionable”

(p. 476). Moreover, Tokumoto and Shibata (2011) affirm that the Japanese group

analyzed preferred the British or American accents (more intelligent and confident),

rather than the Japanese accented English (ranked as funny and modest) (p. 393).

3

1.1 Similar researches and gaps

Several studies have been conducted on how East Asian English accentedness is

considered by other East Asian speakers of English. In their aforementioned article, Tan

and Tan (2008) highlighted how Singaporean English and Singaporean English accent

were perceived in Singapore. A man from Singapore who used to live in the U.S.A. read

two texts – one written in Standard English and the other one written in Singlish (i.e.

broader form of Singaporean English) – and two more speakers from Indonesia and

Hong Kong were asked to do the same. The recordings were played in several schools

in Singapore and the pupils answered some attitudinal questions. Tan and Tan’s

research underlined that a text written in Standard English and read with a Singaporean

English accent was not recognized as proper Singlish, even though it was considered as

intelligent. A further research is presented by Kalra and Thanavisuth (2018), where

different accents from five different East Asian speakers of English are submitted to

Thai students and then analyzed. Japanese English was perceived as “unfriendly, boring

and lazy” (p. 288), whereas Chinese English was classified as “friendly and smart” (p.

289), even though the comprehension was high in both cases. In a similar research

project, led by Sangnok and Jaturapitakkul (2019), undergraduate Thai students were

asked to express their preference towards Singaporean, Filipino and Indian English. The

students ranked Singaporean English accent as a more “difficult accent to understand”

compared to the Indian one (p. 35). Kang, Vo and Moran (2016) present results on how

Vietnamese accented English is evaluated by native and non-native speakers of English

(American, Vietnamese and Arabic students), in terms of accentedness and

comprehensibility. Participants found Vietnamese accented English difficult to

understand, and Vietnamese L1 speakers were more severe while judging the

pronunciation of this variety of English, because of their familiarity with it (p. 11).

Moreover, Arabic and American speakers of English ranked the Vietnamese accent as

“more accented than their own Vietnamese listeners” (p. 12).

There are several researches on how East Asian varieties of English are perceived by

Asian or native speakers of English. However, it was not possible to find projects on

attitudes from a European audience (with English as their L2) towards the

aforementioned varieties of Asian English.

1.2 The goal of the present research

The aim of this research is to discover how a European audience reacts to both

accentedness and comprehensibility related to four different East Asian English

varieties (Singaporean, Japanese, Vietnamese, and Chinese). Given the fact that both

Europe and Asia are in the “Expanding Circle” – excluding English speaking countries

and former colonies – it would be interesting to test Europeans’ reaction to their Asian

peers. A further reason is related to the fact that, as stated in the previous paragraph, it

was not possible to find previous researches on this matter. This project will develop a

similar study to the one proposed by Tan and Tan (2008), although there will be some

4

modifications. The readers will come from four different East Asian countries and the

audience will be exclusively formed by European speakers of English, namely Italians.

This research aims to answer the following questions:

1) How are East Asian varieties of English perceived by an Italian audience?

2) Which variety of East Asian accented English is the most likable?

3) Which variety of English is the most understandable?

It will be interesting to see if the audience will be influenced by stereotypes. Trudgill

(2000) states that “we rely on stereotypes when we first meet and interact with people

[…] and use the way they speak to build up a picture of what sort of person we think

they are (p. 195).

It is complicated to predict some results due to a lack of previous similar researches in a

European context. However, since there are analogous projects with Asian or native

speakers audience, it may be possible to formulate several hypotheses:

a possible outcome could be connected to comprehensibility: the audience may

agree on the fact that a comprehensible accent could be elected as their favourite

one;

another result can be influenced by the accent and the expressiveness of the

speakers. The audience can be deceived by the interpretation of the text from the

speaker while reading, and rate their favourite accent based on this aspect;

a further hypothesis can be connected to the chosen set of adjectives. The

participants could rank with higher votes positive adjectives or lower votes

negative adjectives based on comprehensibility. This aspect was also analyzed in

the aforementioned article by Kalra and Thanavisuth (2018), where

comprehensibility did not influence the ranking process of adjectives by the

participants.

In order to test these hypotheses, a modified Indirect Methodology will be applied

(matched-guise or MG), which “involves presenting study participants with recordings

of one bilingual or bi-dialect speaker reading passages that are identical in all aspects

except for the variable under investigation” (in this case, different East Asian English

accents) (Cavallaro et al., 2014, p. 379). However, minor changes have been brought to

this technique, as explained in section 2.1.

1.3 Main features of the varieties of English analyzed

In this section will be briefly described the main features of the East Asian varieties of

English examined. The four East and Southeast Asian varieties of English chosen are:

Singaporean English, Japanese English, Vietnamese English and Chinese English.

These varieties were chosen because of their position in Kachru’s circles, which is

similar to the Italian one (except for Singapore).

5

In her article, Kobayashi (2011), using Kachru’s model of three circles of English,

considers Singapore as an “Outer Circle” country, where rich families from East Asia

(mostly from China and Japan) send their children to study English (p. 236). Moreover,

she includes both China and Japan in the “Expanding Circle”, with China’s example

that “might soon challenge Kachru’s three-categorisation and the integral notion of

‘English speakers’” (p. 237). Kirkpatrick (2020) agrees with Kobayashi on the position

inside the circles of China, Japan, and Singapore; he also inserts Vietnam in the

“Expanding Circle”, since it was “some form of colony but not of English-speaking

empires” (p. 553). Their main phonetic and phonological features of these Asian

accented varieties of English are briefly described in the following paragraphs. These

characteristics are compared to Received Pronunciation (RP), in order to better explain

thanks to a more familiar variety of English.

1.1.1 Singaporean English

English is one of the four official languages spoken in Singapore, together with Malay,

Mandarin and Tamil. As defined in Leimgruber (2011), Singaporean English, or

Standard Singapore English (SSE), has the “vernacular, Colloquial Singapore English

(CSE), often called ‘Singlish’ by speakers, government language planners, and, indeed,

linguists” (p. 47). The main features of Singlish, according to Leimgruber (2011, pp.

48–49) are:

1. long vowels merge into a single phoneme. This variation does not affect minimal

pairs;

2. vowels [ɛ] and [æ] collapse into one phoneme /ɛ/;

3. CSE has only six vowels: /i/, /ɛ/, /a/, /ə/, /ɔ/ and /u/ (according to Bao, 1998, as

cited in Leimgruber);

4. the diphthongs listed by Bao (1998), as cited in Leimgruber, are /ɔi/, /ai/, /au/, /iə/

and /uə/;

5. the consonants are identical to RP ones. The only exceptions are /θ/ and /ð/,

replaced by /t/ and /d/ respectively. /θ/ is uttered as a [t] or [θ] when it occurs in the

beginning or middle of a word, whereas it is pronounced as [f] in word-final

position ([θ] can occur as well).

1.1.2 Japanese English

English was not considered important in Japan until the Meiji Restoration in 1868,

when “new enlightened leaders came to realize that English would be essential for the

country’s modernization and development” (Honna, 2020, p. 253). English was then

taught in secondary schools and its teaching was strengthened after World War II.

However, the relationship between Japan and English did not seem auspicious. Several

English words were substituted with Japanese words because the concepts expressed by

English words were “not easy to express in Japanese” (Honna, p. 255). English is now

the official language used in many Japanese companies. As stated in Yamaguchi and

6

Chiew (2019, pp. 44–48), and Hazan et al. (2005, p. 364), vowels and consonants

features commonly associated with a Japanese accent of English include the following:

1. /æ/, /ə/, /ʌ/, /ɜ/ and /ɑ/ are frequently substituted by a centralized vowel [ɐ];

2. /ɔ/ and /ʌ/ are uttered as a more close and centralized vowel. The first one as [o]

and the second one as [a] or [o];

3. Japanese speakers of English can pronounce dental fricatives /θ/ and /ð/,

although there are minimal variations in uttering them as [s] or [z];

4. In Japanese there are no lateral approximants. Therefore, /l/ is replaced by the

alveolar flap [ɾ];

5. Labiodentals /v/ and /f/ and bilabials /b/ and /p/ are difficult to pronounce for

Japanese speakers of English, because these sounds do not appear in the Japanese

consonant inventory.

1.1.3 Vietnamese English

Low (2020) states that English arrived in Vietnam during the Vietnam War. It was

introduced by the Americans and was later banned after the end of the war as the

language of the enemy (p. 141). Nowadays, English is a compulsory subject in schools,

even though outside school context an insignificant percentage of the population can

speak fluent English (p. 141). Sundkvist and Nguyen (2020) describe the main features

of this variety of English as follows (pp. 690–691):

1. Large set of monophthongs /i, e, ɛ, a, ɔ, o, u, ɤ, ɯ/, and three diphthongs /iə, ɯə,

uə/. There is no contrast between tense and lax vowels;

2. In word-initial position orthographic <b, d> are pronounced as voiced implosives

/ɓ, ɗ/;

3. Voiced <th> is normally pronounced as a dental stop or an alveolar fricative; a

voiceless <th> is pronounced as a dental fricative [θ] or a dental stop;

4. /dʒ/ is replaced by /dz/ or /ʒ/ in initial position, and in other positions as /g/ or /z/;

5. A glottal stop is inserted before vowels, word-initially and sometimes word-

internally;

6. Final stops /p, t, k/ are pronounced with glottalization [Ɂp˺, Ɂt˺, Ɂk˺];

7. English codas are usually modified with consonant omission, consonant

replacement, or vowel insertion;

8. Single consonants of all manners are often omitted after diphthongs and less

commonly after other vowels;

9. Voiced stops are often substituted with their counterparts (i.e. /d/ is substituted with

/t/, /g/ with /k/, etcetera). Fricatives are commonly replaced by a stop (e.g. /s, z/

with /t/, /ʃ/ with /k/);

10. /l/ is often replaced by /n/ in Hanoi speakers, and he contrary was also found. It is

less common that /l/ is substituted with /r/;

11. In final position, /st/ is reduced to either /t/ or /s/.

12. Three-consonant clusters, such as /kst/ (next), was reduced to /k/, /kt/, or /t/.

7

1.1.4 Chinese English

Honna (2020) states that English arrived in China in 1637, when British ships arrived in

Macau and Canton. After 100 years, “Chinese Pidgin English” developed and it was

used as a lingua franca for trades. Nowadays, English is recognized as an important

language and it is used in international communication. Moreover, English is a

compulsory subject in schools and it is taught from the first grade onward (pp. 248–

250). Deterding (2006a) lists the main features of pronunciation as follows (pp. 178–

193):

1. Chinese English has an extra final vowel (or epenthetic vowel), especially when the

word ends with a plosive. Usually, this vowel is a schwa /ə/. This peculiarity

occurs because Standard Chinese does not allow final plosives;

2. The pronunciation of English by Chinese speakers tends to have a syllable–

based rhythm;

3. Vowels are usually nasalized in specific contexts (i.e. when they occur in a

final position before a final nasal consonant);

4. Voiceless dental fricative /θ/ is pronounced in two ways. If this phoneme occurs in

the beginning of the word, it is uttered as [θ]; in the middle or at the end of a word

it is pronounced as [s];

5. Voiced dental fricative /ð/ is substituted by both [d] or [z]. However, [d] is more

common and occurs especially in medial positions in a word;

6. Voiceless glottal fricative /h/ is pronounced as [x], due to a transference from

Standard Chinese.

7. Voiced postalveolar fricative /ʒ/ is usually pronounced as /ɹ/. Many Chinese

speakers have difficulties articulating this sound, which is realized as a retroflex

approximant similar to English /r/ even though the lips are not rounded;

8. Voiced fricatives /v/ and /z/ are substituted respectively by [w] and [dz];

9. Initial lateral approximant /l/ is substituted by [n] (central China only). Sometimes,

/l/ is uttered as a [ə] or [ʊ] (vocalized /l/);

2.Methodology

The method applied in this research project is a listening test. Twenty-nine Italian

speakers of English were selected as audience and they were asked to listen to four

recordings from four East and Southeast Asian male speakers, and answer a set of

questions. The speakers read the short text “The boy who Cried Wolf” (Appendix A),

which was taken from Deterding (2006b, p. 193). The participants were asked to answer

the questions based on their impressions about accentedness and comprehensibility. The

set of adjectives utilized are, as Cavallaro and Chin (2009) defines them, a “battery of

semantic scales” (p. 144), such as “friendly”, “reserved”, “trustworthy”, and so on. The

aforementioned list of adjectives was inspired by the one applied in Kalra and

Thanavisuth’s research project (2018, p. 287).

8

2.1 Matched Guise Technique

Cavallaro and Chin (2009) define the Matched Guise technique (MG) as a “more

indirect methodological approach to investigating language attitudes [which] is the

measurement of more subjective reactions to variations in languages” (p. 144). This

technique was first introduced by Lambert, Hodgson, Gardner and Fillenbaum (1960),

and it was presented as an “experimental investigation of the extent to which peoples'

judgments about a speaker are determined by his voice” (p. 44). This technique involves

speakers reading the same paragraph, even though one of them reads the text with

diverse accents. This approach aims to obtain a reliable outcome on how the audience

perceives the accents, without knowing that some of the voices they are hearing actually

belong to the same speaker. However, the validity of MG was harshly criticized by

several scholars. Cavallaro and Chin (2009) state that

MG test uses one speaker to produce the desired number of speech samples.

While it is normally possible to find a speaker who can successfully mimic a

small number of different accents, this is not possible in the case of larger

studies when larger numbers of dialects are being tested. The belief is that it is

difficult (or impossible) to find one speaker who can speak or mimic all the

dialectal varieties being investigated (p. 145).

To avoid this problem, the present research applied a variation of the MG technique

known as the “verbal guise test”, in which “different speakers are used for the different

varieties tested” (Cavallaro and Chin, 2009, p. 145).

2.2 Speakers

Four East and Southeast Asian speakers of English from four different countries in the

aforementioned area were selected (Singapore, Japan, Vietnam and China). Three of the

speakers were enlisted because of their friendship with the author of this paper; the last

speaker was recruited by the supervisor of this project. All of them were asked to read

the aforementioned brief text, “The boy who Cried Wolf”. This text was chosen because

some of the main features of the RP British English within the text were compared to

Singaporean (p. 189) and Standard Chinese (p. 190) English pronunciation. Therefore,

the author of this paper wanted to extend the aforementioned text to other Asian

accented English and see whether they share some similarities in the pronunciation or

not. All speakers are male, between 20 and 60 years of age, and all of them are able to

speak English as a Foreign Language (EFL). The four speakers are fluent in English,

since they speak this language at work and at university on a daily basis. One of the

speakers is an undergraduate student, one is a neuroscientist, and another one is a paper

processor. However, for one speaker this piece of information is not available. The

speakers were asked to familiarize with the text before recording themselves. Moreover,

they were asked to read with their native accents.

9

2.3 Italian audience

29 Italian speakers of EFL were selected in order to participate to this project. The

majority of them have been studying English for at least ten years. Only a restricted

percentage was exposed to different accents than British English and American English

(mostly European accented English varieties or Indian accented English). The choice of

selecting only Italian speakers of English was mostly due to the current Covid-19

situation. This decision represented the only solution possible to interview a sufficient

number of people while observing the Covid-19 restrictions established by the Italian

government. Therefore, the collection of the questionnaires requested a significant

period of time. A further reason why only Italian speakers of English were selected is

that the aim was to understand how they perceived an unknown variety of the language

they study or studied, since the majority were exposed only to British English or

American English. This is an important feature of the audience, since their unfamiliarity

with these accents may help the participants avoiding prejudices. Moreover, Italians

were chosen because they have difficulties speaking foreign languages: according to the

Italian National Institute of Statistics (ISTAT), in 2015 a 48,1% of the population was

able to speak English (2017, p. 9). It could be interesting to understand how these 29

participants perceive a non-familiar variety of English.

The majority of the members from the audience is composed by undergraduate students.

Two of them have a degree in English. Five of them are undergraduate students of

English and the rest studied English in school and have different backgrounds

(engineering students, psychology students, law students, philosophy students,

employees). The age range of the interviewees is between 20s and 40s. The audience

was asked to complete the questionnaire after listening to the recordings; moreover,

they were instructed to focus on the likability and comprehensibility of the

aforementioned varieties.

2.4 Set of questions

The questionnaire (Appendix B) is divided into two parts. In the first part the

participants were requested to express how they perceived the accent submitted through

a set of semantic scales (e.g. “posh”, “rural”, “reserved”, and so on). Furthermore, they

were asked to guess the nationality of the speaker. In the second part were presented

three open–ended questions and the audience answered with their personal opinion

about likability and comprehensibility, after rehearing the recordings a second time. The

audience read the questions in part I, before completing the questionnaire. The same

procedure was followed for part II. Part I and part II were separated by a blank page, in

order to prevent the audience from reading the questions in the second part before it was

requested.

3. Results

10

In the following sections are presented the main result of this project. It is important to

clarify that the definition “speaker 1” always refers to the Singaporean speaker;

“speaker 2” identifies with the Japanese speaker; “speaker 3” indicates the speaker from

Vietnam and “speaker 4” is used to refer to the Chinese speaker.

3.1 Part I

In the next tables, which follow the same procedure, the median value was calculated

for each speaker. The 15th

of the 29 participants was considered and this gave the

median value for “low”, “medium” and “high”. These three groups include the ranking

scale from 1 to 5 the audience utilized to express their preferences. “Low” includes the

votes from 1 to 2; “medium” represents 3; “high” 4 and 5. In the columns, the “N”

represents the total number of participants, i.e. 29, and the number between round

brackets indicates the number of participants, but in percentage.

Table 1: comprehensibility (median)

Speaker 1 Speaker 2 Speaker 3 Speaker 4

N - (%) N - (%) N - (%) N - (%)

Low (1-2) 11 (38%) 0 (0%) 1 (3%) 9 (31 %)

Medium (3) 10 (34%) 4 (14%) 2 (6%) 7 (24%)

High (4-5) 8 (28%) 25 (86%) 26 (91%) 13 (45%)

Median Medium High High Medium

Total N=29 (100%)

Table 1 presents a median value of the comprehensibility for each speaker. The

audience ranked with highly comprehensible the accents from speakers 2 (86%) and 3

(91%). On the contrary, the first speaker obtained “low” from the 38% of the audience

and speaker 4 was ranked in the same way by the 31% of the participants.

Table 2: median adjectives speaker 1

Adjectives Low Medium High Median

% % %

Friendly 38% 38% 24% Medium

Reserved 17% 31% 52% High

Trustworthy 14% 28% 59% High

Unreliable 76% 21% 3% Low

Posh 59% 14% 27% Low

Rural 72% 10% 17% Low

Cool 45% 21% 34% Medium

Well Educated 10% 31% 59% High

Poor Formal

Education 72% 17% 10% Low

11

Table 2 presents the percentage of the median value for each adjective for speaker 1.

The median resulted high for “reserved”, “trustworthy”, and “well educated”. The

“medium” value was attributed to “friendly” and “cool”; “unreliable”, “posh”, “rural,

and “poor formal education” were ranked as “low”.

Table 3: median adjectives speaker 2

Adjectives Low Medium High Median

% % %

Friendly 38% 17% 45% Medium

Reserved 79% 17% 3% Low

Trustworthy 21% 31% 48% Medium

Unreliable 90% 7% 3% Low

Posh 79% 7% 14% Low

Rural 48% 21% 31% Medium

Cool 41% 21% 37% Medium

Well Educated 28% 24% 48% Medium

Poor Formal Education 48% 31% 21% Medium

The same procedure in table 2 was applied to table 3 for speaker 2. In this case, there

are no “high” median. “Reserved”, “unreliable” and “posh” were ranked as “low”, and

the rest of the set was considered as “medium”.

Table 4: median adjectives speaker 3

Adjectives Low Medium High Median

% % %

Friendly 28% 34% 38% Medium

Reserved 55% 34% 10% Low

Trustworthy 21% 48% 31% Medium

Unreliable 86% 3% 10% Low

Posh 59% 17% 24% Low

Rural 59% 31% 10% Low

Cool 28% 41% 31% Medium

Well Educated 17% 31% 52% High

Poor Formal Education 69% 14% 17% Low

Table 4 collects the median values of the adjectives for speaker 3. “High” was attributed

only to “well educated”; “medium” to “friendly”, “trustworthy” and “cool”. The rest

was considered as “low”.

12

Table 5: median adjectives speaker 4

Adjectives Low Medium High Median

% % %

Friendly 41% 35% 24% Medium

Reserved 17% 17% 66% High

Trustworthy 21% 52% 27% Medium

Unreliable 72% 21% 7% Low

Posh 59% 24% 17% Low

Rural 66% 10% 24% Low

Cool 62% 21% 17% Low

Well Educated 24% 31% 45% Medium

Poor Formal Education 55% 21% 24% Low

Table 5 presents the median value for the adjectives for speaker 4. In this case, only the

adjective “reserved” was considered as “high”; “friendly”, “trustworthy” and “well

educated” obtained “medium” and the rest was ranked as “low”.

Table 6: median adjectives four speakers

Adjectives Speaker 1 Speaker 2 Speaker 3 Speaker 4

Friendly Medium Medium Medium Medium

Reserved High Low Low High

Trustworthy High Medium Medium Medium

Unreliable Low Low Low Low

Posh Low Low Low Low

Rural Low Medium Low Low

Cool Medium Medium Medium Low

Well Educated High Medium High Medium

Poor Formal

Education Low Medium Low Low

Table 6 collects all the median for all the speakers. The most reserved were judged

speakers 1 and 4; speaker 1 was also considered as the most trustworthy. Speakers 1 and

3 were resulted the most “well educated”, since they obtained “high” in this quality; on

the contrary, speakers 2 and 4 were given “medium” from the audience. All the speakers

were judged as “low” in “unreliable” and “posh”, and “medium” in “friendly”. Speaker

2 was the only one to obtain “medium” in “rural” and “poor formal education”, whereas

13

the others got “low”. Speaker 4 was the only one who was ranked “low” in “cool” (the

others were judged as “medium”).

3.2. Part II

Figure 1: most pleasant accent

Figure 1 presents the data regarding the most pleasant accent to listen to. 45% of the

audience agreed on electing the Vietnamese speaker’s accent (number 3) as the most

pleasant one to listen to. The second most pleasant one is the Japanese speaker’s

(number 2) with a 31% of preferences from the audience. The accents belonging to the

Singaporean speaker (number 1) and the Chinese speaker (number 4) were ranked as the

least pleasant. Speaker 1 was judged as the most pleasant accent to listen to by the 14%

of the audience, whereas speaker 4 obtained only 10% of the preferences.

Table 7: motivations for most pleasant accent

Motivation Speaker 1 Speaker 2 Speaker 3 Speaker 4

No problems understanding it 3% 7% 3% 0%

Clear, easy, expressive 7% 24% 40% 10%

Down to Earth, approachable 3% 0% 0% 0%

Peaceful and harmonious 0% 0% 0% 3%

Total for each speaker 13% 31% 43% 13%

Total 100%

Table 7 collects the audience’s explanations regarding their decisions on the most

pleasant accent to hear. Speaker 3 (Vietnam) collected a 43% of preferences from the

audience, which agreed on describing it as the “clearest, easiest and most expressive

accent” (40%); moreover, the audience had “no problems understanding it” (3%). For

the same reason, speaker 2 (who was considered the second most pleasant accent)

14

obtained 24% of votes. Both speaker 1 and 4 obtained 13% of votes as the most pleasant

accent. One member of the audience considered speaker 1’s accent as “down to Earth

and approachable”. Another participant labelled the accent of speaker 4 as “peaceful

and harmonious”. Moreover, 3% of the audience considered speaker 1 easy to

understand and 7% depicted speaker 2 as such.

Figure 2: favourite accent

Figure 2 reports the percentages of the favourite accent among the four speakers. 41%

of the audience considered speaker 3 from Vietnam as their favourite speaker, whereas

the 38% preferred speaker 2 from Japan. Speaker 1 from Singapore and speaker 4 from

China were ranked as least favourite from 14% and 7% of the audience, respectively.

Table 8: motivations for favourite accents

Motivation Speaker 1 Speaker 2 Speaker 3 Speaker 4

Most fluent speaker 7% 0% 0% 0%

Clear, easy, expressive 7% 28% 31% 0%

Fascinating and/or interesting

(linguistically) 0% 0% 10% 7%

Calming, friendly, funny 0% 10% 0% 0%

Total for each speaker 14% 38% 41% 7%

Total 100%

Table 8 shows the motivations behind the ranking of the favourite accent from the

audience. 31% suggested that speaker 3 was their favourite because he possesses the

“clearest, easiest and most expressive accent”. With the same explanation, 28% of the

participants labelled speaker 2 with the same definition. 10% of the audience considers

speakers 2 accent as “calming, friendly and funny”; another 10% for both speakers 3

and 4 states that these accents are fascinating or interesting from a linguistic point of

view. 7% preferred speaker 1 accent because it sounded as the “most fluent” and

15

another 7% affirms that this accent is the “clearest, easiest, most expressive accent”

among the four.

Figure 3: least favourite accent

In figure 3 is possible to read that nearly half of the participants (48%) ranked speaker 1

as their least favourite accent to hear. The second least favourite belongs to speaker 4

(35%) and the third most voted as least favourite is speaker 2 (17%). Surprisingly, not a

single member in the audience considered the Vietnamese speaker as their least

favourite speaker (0%).

Table 9: least favourite accent motivations

Motivation Speaker 1 Speaker 2 Speaker 3 Speaker 4

Not easy to understand 28% 7% 0% 21%

Too marked accent 0% 3% 0% 7%

Harsh, intimidating 7% 7% 0% 0%

His accent annoyed/bored me 7% 0% 0% 7%

He does not seem fluent in

English 3% 0% 0% 0%

I do not know why 3% 0% 0% 0%

Total for each speaker 48% 17% 0% 35%

Total 100%

In table 9 are collected the motivations the members of the audience provided for their

“least favourite accent” ranking. Since speaker 3 accent was the only one which did not

receive a vote as least favourite, there are no percentages for this speaker in the chart.

48% of the audience judged the accent from speaker 1 as their least favourite, because

“not easily understandable” (28%). A further 7% depicted the accents from both

speakers 1 and 2 as “harsh, intimidating”. The accent belonging to speaker 4 was judged

as “too marked” (7%) and “annoying/boring” (7%). One participant judged speaker 1 as

“not fluent in English”; moreover, another member of the audience was not able to

16

define the reason behind the decision and stated “I do not know”. Another participant

affirmed that speaker 2 possessed a “marked accent” and another 3% did not like his

voice.

It is important to state that the motivations collected were expressed in open-ended

questions, and that the labels in the charts result from the collection of all the similar

answers within the closest definition. For example, several members from the audience

stated that an accent was clear and easy to understand, while some others stated only

that the accent was expressive and/or clear to understand. The decision was made to

unify these two types of answers under the label “clear, easy, most expressive accent”

(as in tables 7, 8, and 9).

4. Discussion

4.1 Hypotheses

The outcomes of the questionnaires partially confirm the hypotheses formulated in

subdivision 1.2. The first hypothesis developed concerned the comprehensibility of the

accents from the four speakers. It was affirmed that the more comprehensible the

accent, the higher the possibility for it to be chosen as the audience’s favourite. This

hypothesis was confirmed by the results of the questionnaires. As shown in table 1, the

most comprehensible speakers were ranked the Japanese (number 2) and the

Vietnamese (number 3), both of them with “high” as median value. As presented in

figures 1 and 2, speaker 3 was elected as the participants’ most pleasant accent to hear

(45%) and their favourite accent among the four (41%). The reason behind the decision

of electing this accent as the most pleasant was justified from the 40% with “clear, easy,

expressive” (table 7) and from the 31% with the same reasons for their favourite accent

(table 8). Moreover, 0% of the participants voted for the speaker as their least favourite

accent, as fig. 3 exemplifies. The same line of reasoning can be followed for the

opposite, i.e. the participants’ least favourite accent. 48% of the audience agreed on

electing the accent of speaker 1 as their least favourite (fig.3), with the main motivation

being “not easy understandable” for the 28% of them (table 9). This hypothesis helped

answering the thesis statements numbered 2 and 3 in section 1.2. The most likable

accent was rated the one belonging to speaker 3, which was also the easiest and clearest

accent to understand.

The second theory proposed in subdivision 1.2 concerned the expressiveness of the

speaker reading the text and the chance to prefer that accent among the others. This

hypothesis was only partially confirmed by the results obtained from the questionnaires.

Speaker 2 was the only one who interpreted the text, and it was a personal choice.

However, as reported in fig.1, only 31% of the participants voted this accent as the most

pleasant one, with the explanation that it was “clear, easy to understand and expressive”

(24%). Moreover, 38% of the speakers expressed their preference for this accent as their

favourite among the four, as reported in fig.2. Their motivation, taken from table 8, was

17

that this was a clear and easy accent to understand (28%). This demonstrates that the

audience preferred the accents only based on a comprehensibility scale, and paid nearly

no attention to the interpretation of the speaker while reading the text.

The third proposal stated in section 1.2 involved the ranking of the adjectives. It was

supposed that the participants would have attributed higher votes to positive adjectives

and lower votes to negative ones based on their preferences on comprehensibility. This

hypothesis is true for some speakers and false for some others. For example, the thesis

is true for the accent belonging to speaker 3 (voted as the audience’s favourite), and

false for the accent of speaker 1 (the least favourite of the participants), but it also

applies to the other two speakers. As shown in tables 2 (speaker 1) and 4 (speaker 3),

good rankings were attributed to positive accents and negative adjectives were judged

with low grades, even though speaker 1 was the least favourite accent among the four.

The least and most favourite accents have similar results. As shown in fig. 1, both

speaker 1 and 3 got “medium” in “friendly” and “cool”; they both obtained “high” in

“well educated” and “low” in “unreliable”, “posh”, “rural”, and “poor formal

education”. The only different results concern “reserved” and “trustworthy”: speaker 1

obtained “high” in both, whereas speaker 3 got “low” and “medium” respectively.

The failure of the second hypothesis, as stated before, could reside in the fact that the

participants based their vote exclusively on the comprehensibility of the accent and not

on the way the text was read by the speakers. Moreover, the partial confirmation of the

third hypothesis may be related to two different variables. The first one could be that the

audience was influenced by their integrity and did not want to rank negative adjectives

with high grades. The second one may be related to the fact that, even though the first

speaker was elected as the least favourite because it was hard to understand, the

members of the audience based their decision only on the accent and not on the speaker

himself. This last affirmation is reinforce by the fact that, even though speaker 1 was the

least favourite of the audience, he was the only one among the 4 speakers to get “high”

in trustworthy (fig.1).

4.2 The first thesis statement

However, whereas thesis statements 2 and 3 were easily answered thanks to the

questionnaire’s outcomes, the first thesis statement in section 1.2, “How are East Asian

varieties of English perceived by an Italian audience?”, has a more vague answer. After

the end of the project, a substantial percentage of the audience was shocked to discover

the real nationalities of the speakers after their guesses (Appendix C), stating that they

did not sound as if they were from that country. This is a peculiar explanation, given the

fact that only a restricted number of the participants affirmed to be in contact with Asian

accented English (Appendix D). However, it was not possible to find a valid answer to

the question aforementioned. It seems that the audience did not pay much attention to

the nationality of the speakers, and that their opinions were not influenced by the

geographical origin of the four speakers. The stereotypes were not included in the

18

ranking process, because the majority of the members of the audience were not familiar

with these accents.

4.3 Limitations of the project

Two main problems emerged at the end of the research project. The first one is related

to the fact that it was not possible to meet every participant in person. As mentioned

before in section 2.3, the pandemic complicated the situation and it was necessary to

follow the restrictions the Italian government has issued in order to stem the virus from

spreading. A considerable number of members from the audience, therefore, were

provided with a set of instructions (Appendix E), in order to explain how to properly

complete the questionnaire.

The second issue regards the geographical origin of the speakers, because the majority

of the participants was not aware of the composition of Southeast Asia and East Asia.

This lack of knowledge influenced the guessing of the country of origin of the speakers

which, however, was not a strictly fundamental piece of information for the final

outcome.

5. Conclusion

As a conclusion, this research project aimed to find an answer to the three statement

thesis cited in subsection 1.2 and provided a perspective on how varieties of English

from the Expanding Circle are perceived from members of the same community. In fact,

there were no researches on how Asian accented English was perceived from a

European audience (Italian audience, in this case).

In the previous section the main findings were discussed, and it emerged that the first

hypothesis in section 1.2 was confirmed and that the audience tended to elect as their

favourite accent the most comprehensible one, which is the accent belonging to the third

speaker (Vietnam). Moreover, the second and third hypotheses were partially

confirmed. The results obtained for the former one highlight the fact that the

expressiveness of the readers while reading the text was not included in the ranking

process, but the participants only considered the comprehensibility of the speaker. The

outcomes from the latter underline that the audience did not base their preferences on

their least or most favourite of the accents. In fact, high votes and low votes were given

to both positive and negative adjectives respectively. This demonstrates that the

members of the audience considered the mere accents, and not the speaker as a person.

Furthermore, it was difficult to find a concise answer to the first thesis statement

regarding the audience attitude towards these Asian accented varieties of English. It was

clear that the audience was not influenced by stereotypes. However, it is also true that

there were no stereotypes available in this case, simply because the majority of the

participants was not accustomed with these varieties of English.

19

In order to test the questions on a larger sample, this project could be used as a starting

point to expand this kind of research to a broader audience. It would be possible, for

example, to include members from different European countries, or participants who

share similarities in their L1s or cultures (e.g. Italians and Spanish, or Swedish,

Norwegian, and Danish). Additionally, in order to acquire a more thorough answer as to

whether or not stereotypes influence the results, it may be possible to add some

questions regarding this aspect. The audience can be divided into two groups with two

different questionnaires. To one group could be submitted the questionnaire from this

research, and the other group could be provided with a questionnaire in which the

nationalities are stated. The final step would include the comparison of the outcomes

from the two groups. It would be interesting to study whether the stereotypes, which did

not play a relevant role in this experiment, would influence the participants in this

hypothetical project.

20

References

Canagarajah, S. (2013). English as translingual. Translingual practice: Global Englishes and

cosmopolitan relations (pp. 56–75). London and New York: Routledge.

Cavallaro, F., & Chin, N.B. (2009). Between status and solidarity in Singapore. World

Englishes, 28 (2), 143–159. https://doi-org.ezp.sub.su.se/10.1111/j.1467-971X.2009.01580.x

Cavallaro, F., Ching Ng, B., and Fifer Seilhamer, M. (2014). Singapore colloquial English:

Issues of prestige and identity. World Englishes, 33 (3), 378–397.

https://doi.org/10.1111/weng.12096

Deterding, D. (2006a). The pronunciation of English by speakers from China. English World–

Wide, 27 (2), 175 –198.

http://explore.bl.uk/primo_library/libweb/action/display.do?tabs=detailsTab&gathStatTab=tr

ue&ct=display&fn=search&doc=ETOCRN190886668&indx=1&recIds=ETOCRN19088666

8

Deterding, D. (2006b). The North Wind versus a Wolf: short texts for the description and

measurement of English pronunciation. Journal of the International Phonetic Association,

36 (2), 187–196. http://journals.cambridge.org/abstract_S0025100306002544

Hazan, V., Sennema, A., Iba, M., and Faulkner, A. (2005). Effect of audiovisual perceptual

training on the perception and production of consonants by Japanese learners of English.

Speech Communication, 47 (3), 360–378. https://doi.org/10.1016/j.specom.2005.04.007

Honna, N. (2020). East Asian Englishes. In Kachru, B.B., Kachru, Y. & Nelson, C.L. (eds.)

(2006). The handbook of world Englishes. (pp. 248–265). Malden, MA: Blackwell Pub.

Italian National Institute of Statistics (ISTAT) (2017). L’uso della lingua italiana, dei dialetti e

delle lingue straniere. Retrieved and translated from http://www.istat.it

Kalra, R. & Thanavisuth, C. (2018). Do you like my English? Thai students’ attitudes towards

five different Asian accents. Arab World English Journal (AWEJ), 9 (4), 281–294.

https://dx.doi.org/10.24093/awej/vol9no4.21

Kang, O., Vo, S.C.T., and Moran, M.K. (2016). Perceptual judgments of accented speech by

listeners from different first language backgrounds. The Electronic Journal for English as a

Second Language, 20 (1), 1–24.

https://eric.ed.gov/contentdelivery/servlet/ERICServlet?accno=EJ1103316

Kaur, P. (2014). Accent attitudes: Reactions to English as a lingua franca. Procedia - Social and

Behavioral Sciences, 134 (3–12).

https://doi-org.ezp.sub.su.se/10.1016/j.sbspro.2014.04.218

Kirkpatrick, A. (2020). Englishes in the Expanding Circle: Focus on Asia. Russian Journal of

Linguistics, 24 (3), 551–568. DOI: 10.22363/2687‐0088‐2020‐24‐3‐551‐568

Kobayashi, Y. (2011). Expanding-circle students learning ‘standard English’ in the outer-circle

Asia. Journal of Multilingual and Multicultural Development, 32 (3), 235–248.

https://doi.org/10.1080/01434632.2010.536239

Lambert, W.E., Hodgson, R.C., Gardner, R.C., and Fillenbaum, S. (1960). Evaluational

21

reactions to spoken languages. The Journal of Abnormal and Social Psychology, 60 (1), 44–

51.

Leimgruber, J.R.E. (2011). Singapore English. Language and Linguistics Compass, 5 (1), 47–

62. https://doi.org/10.1111/j.1749-818X.2010.00262.x

Lochland, P. (2020). The accentedness of English as an additional language (EAL):

A nonnative speaker’s perspective. Asian Journal of English Language Studies (AJELS), 8

(23–56). https://doaj.org/article/24e22d23886440db8fec81ab9b20807f

Low, E.L. (2020). English in Southeast Asia. In Kachru, B.B., Kachru, Y. & Nelson, C.L. (eds.)

(2006). The handbook of world Englishes. (pp. 135–158). Malden, MA: Blackwell Pub.

Sangnok, P. & Jaturapitakkul, N. (2019). Perceptions of Thai undergraduate students toward the

Asian English accents on listening comprehension. rEFLections, 26 (2), 24–50.

https://eric.ed.gov/contentdelivery/servlet/ERICServlet?accno=EJ1267677

Sundkvist, P. & Nguyen, X.N.C.M. (2020). English in Vietnam. In Bolton, K., Botha, W.,

& Kirkpatrick, A. (eds.) (2020). The Handbook of Asian Englishes. (pp. 683–703). John

Wiley & Sons, INC. https://doi.org/10.1002/9781118791882.ch30

Tan, P.K.W. & Tan, D.K.H. (2008). Attitudes towards non-standard English in Singapore.

World Englishes, 27 (3-4), 465–479.

https://doi-org.ezp.sub.su.se/10.1111/j.1467-971X.2008.00578.x

Tokumoto, M., & Shibata, M. (2011). Asian varieties of English: Attitudes towards

pronunciation. World Englishes, 30 (3), 392–408.

http://dx.doi.org/10.1111/j.1467-971X.2011.01710.x

Trudgill, P. (2000). Language and humanity. In Trudgill, P. (ed.) (2000). Sociolinguistics: an

introduction to language and society. (4. ed.) (pp.183–203). London: Penguin.

Yamaguchi, T., & Chiew, P.S. (2020) Is there conflation? An acoustic analysis of vowels in

Japanese English. Asian Englishes, 22 (1), 35–51.

https://doi.org/10.1080/13488678.2018.1563665

22

Appendix A

The Boy who Cried Wolf (Deterding 2006b, 193)

There was once a poor shepherd boy who used to watch his flocks in the fields next to a

dark forest near the foot of a mountain. One hot afternoon, he thought up a good plan to

get some company for himself and also have a little fun. Raising his fist in the air, he

ran down to the village shouting ‘Wolf, Wolf.’ As soon as they heard him, the villagers

all rushed from their homes, full of concern for his safety, and two of his cousins even

stayed with him for a short while. This gave the boy so much pleasure that a few days

later he tried exactly the same trick again, and once more he was successful. However,

not long after, a wolf that had just escaped from the zoo was looking for a change from

its usual diet of chicken and duck. So, overcoming its fear of being shot, it actually did

come out from the forest and began to threaten the sheep. Racing down to the village,

the boy of course cried out even louder than before. Unfortunately, as all the villagers

were convinced that he was trying to fool them a third time, they told him, ‘Go away

and don’t bother us again.’ And so the wolf had a feast.

23

Appendix B

Questionnaire for an Italian audience on attitude towards Asian English accents

Four people from four different Asian countries will read a short text, “The Boy who Cried

Wolf”. They will read the passage one at a time. Please provide your answers to the questions

below.

PART I

Speaker 1

1. Which country do you think the speaker is from? __________________________

2. Did you understand the speaker well? Rate from 1 to 5. 1 5

3. How do you rate the following qualities when you hear this accent? Rate from 1 to 5.

Speaker 2

4. Which country do you think the speaker is from? __________________________

5. Did you understand the speaker well? Rate from 1 to 5. 1 5

1 2 3 4 5

Friendly

Reserved

Trustworthy

Unreliable

Posh

Rural

Cool

Well educated

Poor Formal Education

24

6. How do you rate the following qualities when you hear this accent? Rate from 1 to 5.

Speaker 3

7. Which country do you think the speaker is from? __________________________

8. Did you understand the speaker well? Rate from 1 to 5. 1 5

9. How do you rate the following qualities when you hear this accent? Rate from 1 to 5.

Speaker 4

1 2 3 4 5

Friendly

Reserved

Trustworthy

Unreliable

Posh

Rural

Cool

Well educated

Poor Formal Education

1 2 3 4 5

Friendly

Reserved

Trustworthy

Unreliable

Posh

Rural

Cool

Well educated

Poor Formal Education

25

10. Which country do you think the speaker is from? __________________________

11. Did you understand the speaker well? Rate from 1 to 5. 1 5

12. How do you rate the following qualities when you hear this accent? Rate from 1 to 5.

1 2 3 4 5

Friendly

Reserved

Trustworthy

Unreliable

Posh

Rural

Cool

Well educated

Poor Formal Education

26

27

PART II

The recordings will be played once more. Please pay attention to the different accents and

answer the questions below.

13. Which accent is most pleasant to listen to? Please choose between speaker 1, 2, 3 or 4 and

motivate your answer.

14. Which accent do you prefer among the four different ones you have been listening to? Why?

15. Which accent is your least favoured? Why?

Please answer the following questions with your personal data (for practical reasons only).

16. Age:

17: Sex:

18. Occupation:

19. How long have you been studying English?

20. Are you exposed to different English accents, including different varieties (i.e. Indian

English, Singaporean English etc...)?

This is the end of the survey. Thank you very much for your help!

28

Appendix C

Guess the country

Speaker 1 Speaker 2 Speaker 3 Speaker 4

Guess Japan 21% Russia 31% S. Korea 17% China 31%

Total 100%

29

Appendix D

Other accents exposed to (beside RP and GenAm)

Accent Percentage

None 45%

Indian 10%

Russian 3%

Japanese 3%

European-accented English 8%

Several different accent (Asia) 3%

Several different accent (world) 28%

Total 100%

30

Appendix E

World Englishes: Attitude in the Expanding Circle Towards Asian

Varieties of English

Laura Ortu

BA student

Department of English, Stockholm University

Email: lauraortu58@gmail.com

Supervisor: Dr. Peter Sundkvist

Research conducted as part of the BA degree project

Set of Instructions

Hi! First of all, I would like to thank you helping me in this project. Here below will follow a

brief set of instructions on how to complete this questionnaire. The whole procedure will take

roughly 20/25 minutes. However, you can take all the time that you need.

The questionnaire is divided into two parts. Please take a few minutes to read the whole part

I. Do not read the questions in part II until you get there. You will notice that the two parts

are divided from a blank page in order to prevent the participant from reading part II ahead of

schedule.

In part I you are asked to listen to the whole recording from each speaker and to answer all

three questions from section “Speaker 1” while listening to “track 1” from speaker 1, and

repeat the same procedure for each speaker.

Once you are finished with part I you can proceed with part II. Please take a few minutes to

read the whole part II. While completing this section you do not have to listen to the whole

recording again, but feel free to do so whether if you do not feel sure about your decision.

Answer the questions that follow and motivate your answers.

The last 5 questions of the set are asked directly to you. I cannot communicate this

sensitive content to others and please remind that you are free to withdraw from the

experiment whenever you want.

If you need to contact me, please send me an email to lauraortu58@gmail.com

Please return the questionnaire with your answers not later than January 31st, 2021. Thank

you for your availability and comprehension!

31

Stockholms universitet

106 91 Stockholm

Telefon: 08–16 20 00

www.su.se