An analysis of vietnamese Efl students’ pronunciation of english affricates and nasals

International Journal of English Linguistics; Vol. 8, No. 2; 2018 
ISSN 1923-869X E-ISSN 1923-8703 
Published by Canadian Center of Science and Education 
298 
An Analysis of Vietnamese EFL Students’ Pronunciation of English 
Affricates and Nasals 
Do Phat Loi1 
1 Faculty of Foreign Languages, Van Hien University, Ho Chi Minh City, Vietnam 
Correspondence: Do Phat Loi, Faculty of Foreign Languages, Van Hien University, Ho Chi Minh City, Vietnam. 
E-mail: loidp@vhu.edu.vn 
Received: October 25, 2017 Accepted: December 29, 2017 Online Published: January 19, 2018 
doi:10.5539/ijel.v8n2p298 URL:  
Abstract 
English learners in Vietnam may have many difficulties when pronouncing English, mainly Nasals and Affricates 
since English has been their second most used language. There are many factors which lead to this problem like 
their mother tongue, their study environment, and a lack of knowledge about English phonology and phonetics. 
This assignment analyzes the problems in English’s Affricates and Nasals pronunciations experienced by 
Vietnamese EFL students. Data were collected and analyzed from many Vietnamese students including tips and 
techniques to improve each different pronunciation of Affricates and Nasals. This research will hopefully widen 
knowledge for those who want to improve their pronunciation of Affricates and Nasals and establish more 
information which could be able to be developed in the English phonology and phonetics systems in Vietnam. 
1. Introduction 
1.1 The Background 
English is one of the most commonly used languages all over the world as it is used in many different contexts. 
In Vietnam, Project 2020 by the Vietnamese government has launched in-service training programs for ELT 
teachers at elementary, middle and high schools (Bui Phu Hung, 2016). Also, several schools in Ho Chi Minh 
City, Vietnam, have introduced content-based instructional programs (Bui Phu Hung & Tran Thi Hai, 2016). 
However, Vietnamese students’ pronunciation errors are still an issue of concern (Graceffo, 2010). Pronunciation 
errors may be caused by the teacher’s application of inappropriate teaching methods, interference of L1 (Shak, 
Lee, & Stephen, 2016). 
Pronunciation plays a significant role in real-life communication as it reflects the speakers’ identities and the 
communities which they belong to (Seidlhofer, 2001). Good pronunciation also helps students get academic 
achievements and graduates get accepted for vacancies and promotions at work. However, Shak, Lee and 
Stephen (2016) identified many problems in Vietnamese students’ pronunication of English. 
This study is aimed to analyze Vietnamese EFL students’ errors in pronunciation of affricates and nasals in 
English and suggests methods of teaching and learning pronunciation for Vietnamese students at low levels of 
proficiency as Vietnamese language does not have consonant clusters or a glide from one consonant to another as 
in English affricates (Huynh Trang Nguyen & Dutta, 2017) and Vietnamese EFL students do not pronounce 
English final consonants clearly enough (Duong Thi Nu, 2009). 
1.2 Research Questions 
1). Which of the English affricates and nasals do Vietnamese EFL students make most pronunciation errors in? 
2). Do Vietnamese students make more errors in pronoucing affricates or nasals? 
2. Literature Review 
2.1 Affricates 
2.1.1 Definition 
An affricate is a consonant that begins with a stop and ends with a fricative. These two sounds are generally 
homorganic (Roach, 2001). In other words, the places of articulation of these two sounds should be close. In 
English, there are two affricates, known as /tʃ/ and /dʒ/ in which the first consonant is an alveolar plosive, 
ijel.ccsenet.org International Journal of English Linguistics Vol. 8, No. 2; 2018 
299 
produced with a plosion, and the second is a post-alveolar fricative, produced with friction (Roach, 2001). Clark, 
Yallop & Fletcher, 1995, p. 65) believes that “There is almost always some degree of air turbulence (and hence 
friction) at the release of a stop. This is normally of such short duration that it counts as part of the release burst 
of the stop itself. However, when the release is strongly frictional and is extended in duration, it can be identified 
as a separate fricative phase of the articulation. A single complex segment of this kind, in which the articulators 
release an occlusion through a controlled fricative phase, is known as an affricate or affricative.” (Clark, Yallop, 
& Fletcher, 1995, p. 65) 
Pisoni & Remez (2005, p. 195) identify that the English affricates are an acoustic segment of oral alveolar stops 
and post-alveolar fricatives which are incorporated perceptually into a single phonological entity. This acoustic 
sequence is produced by the oral alveolar stop followed by the post-alveolar fricative. As a consequence, the 
English affricates contain acoustic cues for both stops and fricatives. That means the acoustic cues for the 
English affricates /tʃ/, /dʒ/ are more complicated than those of the English fricatives /ʃ/, /ʒ/. 
/tʃ/ is a voiceless consonant in which /t/ glides from alveolar to post-alveolar position, both /t/ and /ʃ/ are both 
voiceless consonants so /tʃ/ is a voiceless consonants. /tʃ/ can be spelled in the initial position, medial position 
and final position, usually spelled “ch”. Nonetheless, /dʒ/ is a voiced consonant which can be spelled in initial 
position, medial position and final position as “j”. The following table can illustrate this point. 
Table 1. English affricates 
Affricates Initial Medial Final 
/tʃ/ check /tʃɛk/ 
chat /tʃat/ 
teacher /ˈtitʃər/ 
kitchen /ˈkɪtʃən/ 
watch /ˈwɑtʃ/ 
catch /kætʃ/ 
/dʒ/ job /dʒɑb/ 
jam /dʒæm/ 
adjust /əˈdʒəst/ 
project /prəˈdʒɛkt/ 
fridge /frɪdʒ/ 
average /ˈævərɪdʒ/ 
Misra (2000, p. 78) identifies that /tʃ/ is a voiceless alveolar palatal affricate. This sound can be produced when 
the tip of the tongue touches the alveolar ridge or just behind it. For example, the sound /č/ in the word “church”. 
It may be noticed that the production of this sound is similar to that of /š/, except that the airflow is initially 
stopped and then released slowly. On contrary, /dʒ/ is a voiced alveolar palatal affricate. The sound /ǰ/ in the 
word “judge” is produced in a similar fashion as /č/, except that it is a voiced sound. The fricative /s/, /z/, /š/,/ž/ 
and the affricates /č/ and /ǰ/ are together called as sibilants, as they produce a typical hissing noise. 
2.1.2 Arguments on Affricates 
There have been arguments on analyzing affricates. First is the one-phoneme analysis, as explained in the first 
part, affricates /tʃ/ and /dʒ/ are phonetically composed of a plosive followed by a fricative, and it is possible to 
treat each pair /tʃ/ and /dʒ/ as a single consonant phoneme, so we called this the one-phoneme analysis. Second is 
two-phoneme analysis, affricates are also be treated as composed of two phoneme each, respectively, /t/ followed 
by /ʃ/ and /d/ followed by /ʒ/ , all of which are established as independent phonemes in English. 
Table 2. Phonemic analysis of affricates 
 Church judge 
One-phoneme analysis t-ʃ-ər-t-ʃ d-ʒ-ʌ-d-ʒ 
Two-phoneme analysis tʃ-ər-tʃ dʒ-ʌ-dʒ 
As in the two-phoneme analysis, it is possible to say that an affricate is composed of two phonemes each -/t/ plus 
/ʃ/ or /d/ plus /ʒ/ respectively. If we adopt the two-phoneme analysis, the words “church” and “judge” would be 
composed of five phonemes each instead of the three-phoneme that result from the one-phoneme analysis. 
However, most linguists prefer the two-phoneme analysis as it has one main advantage that if there are no 
separate /tʃ/ and /dʒ/ phonemes, then our total set of English consonants is smaller. Many phonologists have 
claimed that one should prefer the analysis which is the most “economical’’ in the number of phonemes it results 
in (Roach, 2001, p. 122). 
Phoneticians argue that the phonetic quality of the /t/ and /ʃ/ as in (“hutch” /hətʃ /and “watch apes” watʃ eps/) is 
different from realizations of /t/ and /ʃ/ in other contexts as in (“hush” /həʃ /or what shapes /wət ʃeps/). This 
argument is weak since there is no evidence for the existence of these phonetic differences. 
ijel.ccsenet.org International Journal of English Linguistics Vol. 8, No. 2; 2018 
300 
2.1.3 Distribution of Affricates in English Words 
The phonemes /tʃ/ and /dʒ/ are widely distributed like other consonants in all positions (initial, medial, final) 
while other combinations of plosives and fricative do not. However, several consonants like /r/, /w/, /j/, /h/, /ŋ/, 
/ʒ/ are generally accepted as phonemes despite not being free in distribution. If /tʃ/ and /dʒ/ are able to combine 
quite freely with other consonants to form consonant clusters, e.g., finally in “watched” / wɑtʃt/. However, such 
clusters do not exist in the initial position and are very limited in the final position. 
The intuitions of native speakers may suggest that /tʃ/ and /dʒ/ are each “one sound”. The problem is that 
discovering what native speakers feel about their own language is not easy. 
Furthermore, when it comes to assimilation, affricates have more drastic type of co-articulation. Instead of 
“sharing” part of a sound, the merged sounds are pronounced as an entirely different sound. Two examples of 
assimilation occur when /t/ or /d/ precede the “y sound” /y/. 
When the /t/ and /y/ assimilate, the sounds merge into the “ch sound” /ʧ/, which causes the phrase “don’t you” 
/dəʊnt juː/ to be pronounced as “donchou” /dəʊnʧu/. Similarly, the /d/ assimilates with the /y/ and is pronounced 
as a “j sound” /ʤ/, which causes the phrase “did you” /dɪdyu/ to be pronounced as “dijou” /dɪʤu/. 
2.2 Nasals 
2.2.1 Definition 
Hancock (2003) recognizes that the consonant sounds /m/, /n/ and /ŋ / are made by stopping the flow of air out of 
the mouth so that it goes through the nose instead. These three consonants are different because the air is stopped 
at different parts of the mouth. In particular, /m/, /n/ and /ŋ/ are bilabial, alveolar and velar in position of 
articulation respectively. Agreeing with Hancock (2003), Roach (2001, p. 58) further explains that the soft palate 
must be lowered for nasals to be produced. However, for all the other consonants and vowels, the soft palate is 
raised and air cannot pass through the nose. He adds air does not pass through the mouth; it is prevented by a 
complete closure in the mouth at some point. The three types of closure are: bilabial (lips), alveolar (the tongue 
blade against the alveolar ridge) and velar (the back of tongue against the palate). This set of places produces 
three nasal consonants /m/, /n/, and /ŋ/, which correspond to the three places of articulation for the pairs of 
plosives /p/ and /b/, /t/ and /d/ and /k/ and /g/. 
Almost all nasals are continuants, where air comes out through the nose but not through the mouth, as it is 
blocked by the lips or tongue. Nasals are all voiced and never appear in pair as plosives, fricatives and affricates. 
2.2.2 Characteristics of Nasals 
Roach (2001, p. 58) asserts that /ŋ/ consonant is at velar and this consonant is different from /m/ and /n/, which it 
causes a lot of trouble for those who is not a native speaker as this consonant does not exist in some languages. 
In addition, non-native English speakers’ pronunciation of this consonant may be found unintelligible. While 
large number of languages make use of phoneme /ŋ/, many others lack this sound. However, few of the sounds 
commonly found among the phonemic inventories of the world's languages exhibit a more clearly definable 
distribution than that exhibited by /ŋ/. This distribution has two unrelated aspects. One is the striking areal 
distribution of the presence vs. absence of phonemic /ŋ/ among the languages of the world. The other striking 
aspect of phonemic /ŋ/ is its phonotactic distribution. That means it may not appear in all positions in the word, 
but rather confines itself to initial, medial, or final position, or some combination thereof. In the case of 
restriction of ŋ to non-initial position, this, too, has a relatively pronounced areal skewing among the world’s 
languages. 
Table 3. Distribution of nasals in a word 
 Position 
Consonants 
Initial Medial Final 
/m/ mother /ˈmʌðə/ 
more /ˈmɔːri/ 
man /mæn/ 
family /ˈfæməli/ 
human /ˈhjuːmən/ 
image /ˈɪmɪdʒ/
flame /fleɪm/ 
swim /swɪm/ 
thumb /θʌm/ 
/n/ nurse /nɜːs/ 
knife /naɪf/ 
night /naɪt/ 
planet /ˈplænɪt/ 
pineapple /ˈpaɪnæpl/ 
pony /ˈpəʊni/
clown /klaʊn/ 
done /dʌn/ 
phone /fəʊn/ 
/ŋ/ longer /ˈlɒŋɡə(r)/ 
singer /ˈsɪŋə(r)/ 
anger /ˈæŋɡə(r)/ 
jungle /ˈdʒʌŋɡl/ 
thinking /ˈθɪŋkɪŋ/
king /kɪŋ/ 
king /sprɪŋ/ 
tounge /tʌŋ/ 
ijel.ccsenet.org International Journal of English Linguistics Vol. 8, No. 2; 2018 
301 
Table 3 shows that /m/ and /n/ appear quite widely in a word, but /ŋ/ does not appear in the initial position. 
However, Vietnamese language has a consonant also called velar nasal occurs in all initial, medial and final 
positions (Kirby, 2011). 
2.3 Previous Studies 
There have been many studies on EFL students’ pronunciation of English, but the studies by Dikilitaş & 
Geylanioğlu (2012), Duong Thi Nu (2009), Hassan (2014), Mirza (2015), Ohata (2004) and Zhang (2009) as 
they have been conducted in contexts where English is used as a foreign language. 
Dikilitaş & Geylanioğlu (2012) conducted a study on 24 Turkish EFL adult students’ pronunciation of English 
consonants and vowels. In this study, each of the participants was given 10 words for each sound. Their 
pronunciations were recorded and transcribed IPA alphabet and compared to the phonetic transcription in 
dictionaries. The Turkish participants were found having difficulties in pronuncing the target sounds, including 
/ŋ/. It is recommended from the study that students should be trained through concetualization theory to assist 
learners in forming an idea or principle of the target pronunciation lessons. 
Duong Thi Nu (2009) studies the pronunciation mistakes of English fricatives and affricates made by Vietnamese 
students. The scope included both the manners and places of articulation of these consonants. She also suggested 
ways to improve EFL students’ pronunciation of English: good teacher instruction, comparisons and contrasts of 
English and Vietnamese phonetic features and more controlled practice and language use. 
Hanssan (2014) investigated Sudanese problems with pronunciation of English words. 50 Sudanese students at 
University of Sudan of Science and Technology were involved as participants in the study. Their spoken English 
was observed, recorded and analyzed. These participants were also interviewed. A conclusion of the study is that 
interference of L1, differences between two languages, inconsistency of sounds and spelling were the main 
causes of English pronunciation problems. 
Mirza (2015) analyzed Lebanese EFL and EFL students’ pronunciation mistakes. 22 participants were included 
in the study. Tests and checklists were applied. The findings revealed that the participants made more mistakes 
with fricatives and affricates. 
Ohata (2004) examines phonological difference between English and Japanese. The study also discussed 
problematic areas of pronunciation for japanese learners of English. The author also figured out implications for 
L2 pronunciation teaching based on the contrastive analysis of the two languages. 
Zhang (2009) analyzed common problems in Chinese EFL learners’ pronunciation. The factors that led to these 
problems included L1 interference, learners’ age, attitudes and their inadequate understanding of phonological 
and phonetic system of English. 
3. Research Methods 
3.1 The Pilot Study 
5 students were voluntarily included in the pilot study and they were not included in the main study. They were 
given a text and were required to read the text loud. Their readings were recorded and then compared to available 
recordings, made by native-English speakers, accompanying the textbook. After the pilot study, the text was cut 
off to 200 words as the participants claimed that a long text would take them much time. 
3.2 Participants 
32 Vietnamese EFL students from different institutions of higher education located in Ho Chi Minh City were 
included in the study. Their age range was 18-23 and they were from different cities and provinces in Vietnam. 
They had all learned English at basic levels of education. 
3.3 Instruments 
References accompanied by recordings were read through by the researcher to be level-fit. Finally, an 
approximately 200-word elementary-level paragraph was chosen. Recorders were also used to record the 
participants’ readings of the text. 
3.4 Procedure 
Firstly, an approximately 200-word paragraph was chosen among many textbooks with words containing 
affricates and nasals. The chosen paragraph was not commonly taught in Ho Chi Minh City, Vietnam. The 
paragraph was presented in a printed form. The participants were asked to read out loud the paragraph for a few 
times so they could be familiar with it. Afterwards, their readings were recorded and converted into MP3 format. 
ijel.ccsenet.org International Journal of English Linguistics Vol. 8, No. 2; 2018 
302 
Secondly, the recorded readings were transcribed according to English phonetic symbols (Roach, 2001) by four 
Vietnamese EFL teachers one by another. These four teachers used the recording in accompaniment with the text 
as a reference for their checking. Then the recorded readings were checked through by a native-English teacher 
who had come to Vietnam 2 weeks before the implementation of this study. The participants’ pronunciations of 
the affricates and nasals were labelled accuracy levels: high, intermediate, and low to mean the levels of 
identification of the phonemes. 
Finally, the participants’ pronunciation mistakes were summed up and discussed by all the checkers for a mutual 
agreement on the labels of accuracy. 
4. Results and Discussions 
32 recordings made by 32 the participants were analyzed with a focus on affricates and nasals at three main 
levels: high, intermediate and low. Table 4 and Table 5 illustrate this analysis of the participants’ pronunciation 
of each word in the text. The statistics show the numbers and percentages of pronunciation levels of accuracy of 
the mentioned words. 
Table 4. The analysis of affricates 
 Accuracy level 
 Word 
High Intermediate Low 
approach /əˈprəʊtʃ/ 34,4% 
(11/32) 
40,6% 
(13/32) 
25% 
 (8/32) 
children /ˈtʃɪldrən/ 59,4% 
(19/32) 
37,5% 
 (12/32) 
3,1% (1/32) 
education /ɛdʒʊˈkeɪʃən/ 15,6% 
(5/32) 
3,1% 
(1/32) 
81,3% 
(26/32) 
educationally /ˌɛdʒʊˈkeɪʃənəli/ 15,6% 
(5/32) 
3,1% 
(1/32) 
81,3% 
(26/32) 
rich /rɪtʃ/ 56,25% 
(18/32) 
12,5% 
(4/32) 
31,25% 
(10/32) 
encourage /ɛnˈkʌrɪdʒ/ 31,3% 
(10/32) 
15,6% 
(5/32) 
53,1% 
(17/32) 
watching /wɒtʃɪŋ/ 68,8% 
(22/32) 
21,9% 
(7/32) 
9,3% 
(3/32) 
literature /lɪtərətʃə/ 25% 
(8/32) 
40,6% 
(13/32) 
34,4% 
(11/32) 
language /laŋɡwɪdʒ/ 34,4% 
(11/32) 
25% 
(8/32) 
40,6% 
(13/32) 
bird-watching /bəːd-wɒtʃɪŋ/ 56,25% 
(18/32) 
37,5% 
(12/32) 
6,25% 
(2/32) 
church /tʃəːtʃ/ 34,4% 
(11/32) 
43,8% 
(14/32) 
21,8% 
(7/32) 
Table 5. The analysis of nasals 
 Accuracy level 
High Intermediate Low 
nikki 96.9% 
(31/32)
0% 
(0/32)
3.1% 
(1/32) 
homeschooling /həʊmskuːlɪŋ/ 87.5% 
(28/32)
12.5% 
(4/32)
0% 
(0/32) 
an /ən/ 100% 
(32/32)
0% 
(0/32)
0% 
(0/32) 
many /mɛni/ 96.9% 
(31/32)
3.1% 
(1/32)
0% 
(0/32) 
traditional /trəˈdɪʃənəl/ 93.75% 
(30/32)
6.25% 
(2/32)
0% 
(0/32) 
academics /ˌækəˈdemɪk/ 100% 
(32/32)
0% 
(0