Intelligibility And Comprehensibility Of Communication English Language Essay

4864 words (19 pages) Essay

1st Jan 1970 English Language Reference this

Disclaimer: This work has been submitted by a university student. This is not an example of the work produced by our Essay Writing Service. You can view samples of our professional work here.

Any opinions, findings, conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of UKEssays.com.

In order to assess intelligibility and comprehensibility of both speaker and listeners factors in the communication between Cantonese and French as Non-Bilingual English Speakers, the speech of four picture descriptions and a read passage were recorded to analyze the phonological variances at the segmental level in two experiments on the ground of markedness of L1 transfer¿½CMarkedness Differential Hypothesis (MDH) in second language acquisition. For Experiment 1, it should be noticed that sufficient input of certain foreign-accented speech will contribute to better perception. For Experiment 1& 2, the results suggest that the accuracy of vowel affects intelligibility and comprehensibility the most and it turns out that the mispronunciation of word-final consonant and onset or coda clusters is not as detrimental to intelligibility as expected.

Introduction

English as an international language is prevailing as a communicative tool among non English speaking countries. According to ¿½¿½A quarter of Chinese study English: official¿½¿½ (2006), more than 300 million Chinese, nearly a quarter of the country’s population, have studied English either as a major course or as an elective subject. Jenkins (2000) claimed that a variety of social contexts, and different English accents might threaten intelligibility especially in communication with Non-Bilingual English Speakers (NBES). According to Skutnabb-Kangas & McCarty (1984), bilingual speakers are able to function in two or more languages, acquire the

communicative and cognitive competence, and establish both L1 and L2 identities in

monolingual or multilingual communities at the same level as native speakers. Since most Chinese learners in mainland China are non-bilingual English speakers, a salient problem emerges since Chinese learners are inhibited to communicate with foreigners and have difficulties understanding different non-native English accents.

Native vs. Foreign and Native vs. Intelligibility

For the last few decades, the discussion on whether there is a ¿½¿½native¿½¿½ accent that L2 learners might actually acquire to get rid of their ¿½¿½foreign¿½¿½ accents for good has drawn the attention of teachers and researchers. Bongaerts (1999; cf. Hansen 2008, p.200) believed that most people who learn English after early childhood are unlikely to speak with a native or native-like accent, except for those who are highly motivated to learn L2 pronunciation, according to Moyer (2004; cf. Hansen 2008, p.200), or those whom Ioup, Boustagui, EI Tigi, & Moselle (1994; cf. Hansen 2008, p.200) considered to have a special talent in L2 learning. Foreign-accented English is common, yet it does not hinder millions of second language users around the world from successful communication. It indicates that intelligibility and comprehensibility rather than accent-free pronunciation is the ultimate goal that L2 learners and teachers should aim at. Smith and Nelson¿½¿½s definition of intelligibility and comprehensibility, referred to the degree of ¿½¿½word and utterance recognition¿½¿½ and ¿½¿½word and utterance meaning¿½¿½ respectively, are further described by Bamgbose (1998; cf. Jenkins, 2000, p.70) as ¿½¿½a complex of factors comprising recognizing an expression, knowing its meaning, and knowing what that meaning signifies in the sociocultural context¿½¿½. He

also notices the importance of speaker and listener factors to speech interpretation.

Hence, intelligibility in production and comprehensibility in perception are inseparable in communication.

Research Objective, Theoretical Framework, and Significance

Previous research in L2 phonology has focused more on the speech accommodation and other social factors such as gender, social identity, and extent of L1/L2 use etc, which investigate the social communication between bilingual speakers

Get Help With Your Essay

If you need assistance with writing your essay, our professional essay writing service is here to help!

Find out more

(e.g. immigrants or students abroad) and native English speakers. However, there are few attempts on the NBES communication. Therefore, the research demonstrated whether it is really difficult for Chinese learners to understand French English even if people are speaking the same language; if ¿½¿½Chinglish pronunciation¿½¿½ renders any unintelligibility to interfere inter-cultural communication.

As Jenkins (2000) noted the L1 transfer in NBES communication is important with ¿½¿½speakers of different first languages varying their pronunciation under the influence of their L1 phonology¿½¿½, the research aimed to investigate to what extent L1 transfer affects the L2 speech. Also the phonological variants in the perspective of L1 transfer based on the theory in linguistic markedness proposed by Zobl (1983; cf. Larsen-Freeman, 2000, p.101) were analyzed. They noted that ¿½¿½linguistically unmarked features of the L1 will tend to transfer, but that linguistically marked L1 features will not¿½¿½. Larsen-Freeman (2000) mentioned, for those languages which have voiceless stops but do not have voiced one, voiceless is unmarked while voiced is marked due to their additional phonological feature. The Markedness Differential

Hypothesis (MDH) raised by Eckman (1997; cf. Larsen-Freeman, 2000, p.102)

provided L1 transfer with a systematic framework. He claimed that

(a)

Those areas of the L2 which differ from the L1, and are more marked than

the L1 will be difficult.

(b)The relative degree of difficulty of the area of the L2 which are more

marked than the L1 will correspond to the relative degree of markedness.

(c)

Those area of the L2 which are different from the L1, but are not more

marked than the L1, will not be difficult.

(Eckman,1997; cf. Larsen-Freeman, 2000, p.102)

In order to clarify the voice/voiceless distinction in three word positions, Table

1 shows the possible combination of voice/voiceless in initial, medial, and finial position in different languages. Compare English with French (Corsican), French is less marked. Thus, French might have difficulties in producing English words with the voiced/voiceless consonants in the medial and final position.

Languages with voiced & voiceless contrast English, Arabic, Swedish German, Japanese, Catalan Corsican, Sardinian Not exists

Initial position

¿½¿½ ¿½¿½ ¿½¿½ ¿½¿½/¿½¿½

Medial position

¿½¿½ ¿½¿½ ¿½¿½ ¿½¿½/¿½¿½

Final position

¿½¿½

¿½¿½

¿½¿½

¿½¿½/¿½¿½

Table 1: Three possible word positions in words with a voiced/voiceless distinction established by Dinnsen and Eckman (1975, cf. Larsen-Freeman, 2000, p.102)

On the other hand, Yavas(2008) claimed that obstruents are more difficult than sonorants for Cantonese learners. Since some obstruents and sonorants are permitted as codas in Cantonese, they encountered difficulty in learning single codas of English.

The significance of intelligibility and comprehensibility in intercultural communication means a lot for Chinese learners. Instead of struggling with their ¿½¿½accents¿½¿½ and inhibited to speak out confidently, they will focus on the phonological core in order to realize intelligibility, comprehensibility, and acceptability in their speech. Ultimately, it might enhance their competence of phonological adjustment and cultural awareness.

Literature Review

Jenkins (2000) argued that Smith and Nelson¿½¿½s definition of intelligibility identifies perception processing at the micro-level e.g. recognizing words and utterances in the context but they probably underestimate the influence of pronunciation on miscommunication or even communication breakdown. Pronunciation, as Jenkins (2000) implied, is the crucial factor in unintelligibility in communication between non bilingual English speakers, due to the pattern underlying NBES interaction. They rely much more on identifying the acoustic signal (bottom-up processing) than making use of contextual cues (top-down processing) with limited top-down skills and schemata to perceive the foreign accented speech. Intelligibility and comprehensibility of the NBES communication are thus closely related to inter-speaker segmental variation e.g. L1 transfer and intra-speaker(different L1 speakers communicating with the same L2) segmental variation regarding the degree of speech accommodation (or the strategy of convergence). Beebe and Giles (1984) pointed out that making one¿½¿½s speech respectively more or less like the interlocutor aims to

promote communicative efficiency. In her earlier pilot study (Jenkins, 2000) claimed

same-L1 English interlocutors were more likely to maintain mutual perception than different-L1English interlocutors because of foreign accent and unrelated L1 background. Notwithstanding, she reported that the desire of being understood motivates different-L1 English interlocutors to speak more slowly and articulate more clearly. Intelligibility can be achieved by transfer replacement that different-L1 English speakers would replace their L1 transfer and attempt to converge on target-language forms or the form of interlocutor¿½¿½s IL, especially in information exchange tasks. Jenkins was also concerned about whether it would affect either perception of the interlocutor or production of the speaker and whether it would be considered a high risk for intelligibility problems or a big leap for convergence on acceptable pronunciation.

The contrastive analysis was based on a study of Jenkins (2000) in which six English learners, two Japanese, three Swiss-German, and one Swiss-French (not engaged in same-L1 pairs) at the upper-intermediate level were grouped in same-L1 pairs and different-L1 pairs to complete different tasks (e.g. information exchange task) as preparation for the Advanced English Speaking exam. They completed the questionnaires and participated in interviews as well. The transcriptions annotated the phonological deviants from target-language form to examine for evidence of convergence by replacing L1 transfer errors.

Jenkins (2000) noted two issues during the interpretation of phonological variation and convergence. First of all, the phonological competence of the same-L1 pair was expected to improve because they were recorded six weeks after the

different-L1 pair. However, it was reported that phonological deviations in the same-

L1 pair outweighed those in the different-L1 pair. For example, Japanese substituted /r/ for /./ e.g. flower [.fra..] and beard was mispronounced [b.¿½¿½.]. Both were accountabilities for potential unintelligibility. On the other hand, the Japanese subjects had less non-target form than the Swiss-German subjects because they tended to speak more slowly, pause more frequently or for longer periods, and speak less to avoid losing face or avoid the embarrassment of not being understood. Thereby they consciously or subconsciously replaced L1 transfer deviants or suppressed their L1 transfer errors, which also accounted for convergence reflected in the different-L1 pair.

A relatively comprehensive picture of NBES interaction was provided. A speaker would make considerable effort to replace L1 transfer when he or she communicated with different-L1 English interlocutors.

Critique

The design of Jenkins¿½¿½s research was almost impeccable because she has successfully elicited the participants to describe certain things with pictures during the information exchange task. Even though there were a few sentences with grammatical flaws but overall they were comprehensible in the oral assessment. As Munro (2008;cf. Hansen, 2008, p.208) suggested, controlled production tasks might only tap into the sub-skills and underestimate the speaker¿½¿½s communicative competence: e.g. reading materials may lead to mispronunciations because of orthographic interference, dysfluencies, and unnatural prosody . Thus, sometimes tasks that elicit extemporaneous speech were preferable.

On the other hand, Jenkins (2000) reported that though Japanese took on the Swiss-French speakers speech characteristics during the six-month program, the traditional sense of convergence rarely occurs during the information exchange task. The result indicated that even though mutual perception is built, but the influence of L1 transfer is more salient than convergence.

Experiment 1: Information Exchange Task

Methodology

Inspired by Jenkins¿½¿½s (2000) study, an experiment is going to be implemented to find out to what extent French-accented English renders difficulties for Cantonese speakers and whether Cantonese-accented English is hard for French speakers to comprehend. The first experiment requires four participants in UIC, two Year 4 Chinese students whose L1 is Cantonese and two French speakers, to accomplish the information exchange tasks. What mainly differs from Jenkins¿½¿½s study is: (1)the two Chinese have not converse with the French before; (2) each of the Chinese students listened to the previously-recorded English speech from the French and then retold the content in Cantonese, while the French listened to the recorded English speech from the Cantonese and retold the descriptions in English, which means they had no face to face communication. The issue of face loss was avoided in order to elicit longer speech and convergence was not involved in order to elicit their L2 variance. In addition, two Cantonese (C1 & C2) who major in TESL (Teaching English as a Second Language) have passed the TEM-4 (Test for English Major) but do not know

French. The other two participants (F1 & F2) are French professors in UIC. C1 and F1

are in Dyad . while C2 and F2 are in Dyad .. .

In order to assess the intelligibility, first, the speeches of C1, C2, F1, and F2 were recorded individually. Second, the full texts were written down and the phonemes that deviated from Received Pronunciation or American English were transcribed and annotated in IPA forms. Third, according to Munro¿½¿½s (2008; cf. Hansen, 2008, p.206) approach for assessing L2 speech, besides the researcher¿½¿½s transcription, the responses from participants were taken into consideration. Thus, after the participants in the same dyad listened to the previously-recorded speech, to which they were only allowed to listen to once, any comment relevant to the interlocutors¿½¿½ pronunciation including those unintelligible phonological variances was noted down.

Results The full texts were written and the phonological deviations were marked by IPA forms.

According to what F1 and F2 reported, they had no difficulties in understanding the speech produced by C1 and C2. However, C2 had some problems in identifying certain words in the speech of F2 because of not being able to retell the whole picture (picture 2). The results indicated that the phonological deviations C1 and C2 did not lead to unintelligibility for F1 and F2. Description of F1: Picture 1 (See Appendix A) Ok, it¿½¿½s a western man, I will say, I don¿½¿½t know. Fifty-five, sixty, fifty-five years old. Eh, looks like in the restaurant, he¿½¿½s drinking a beer, and having I don¿½¿½t know, some French fries and salad [s¿½¿½¿½¿½l¿½¿½.]. He¿½¿½s eh, kind of red. Eh, the food doesn¿½¿½t look very tasty [d..st.]. Eh, it¿½¿½s a big beer, he already had more than half. And the restaurant doesn¿½¿½t look very nice. Eh, I don¿½¿½t know it could be many countries. I would say it¿½¿½s a European, yeah that¿½¿½s it. He has mustache. He has mustache, eh, glasses, he¿½¿½s eh, eh becoming bald, not many hairs left. Wearing a white jacket, sweater and a shirt underneath. Eh it¿½¿½s a smoking restaurant. There¿½¿½s an ashtray [..str.] on the table. Eh, I don¿½¿½t know this beer brand, Everest. What else.

C1 reported that she understood most of the speech but she did not catch the words ashtray and the brand of the beer.

Description of F2: Picture 2 (See Appendix A) So this [z.s] is a picture, eh of a flat [f.¿½¿½t]. Eh, obviously [….s..] it is going to be one Christmas, because there is a Christmas tree and a gift just under [..

..] it. And some Christmas, typical Christmas decoration [d.k.¿½¿½r….n]. Eh, the picture was taken at night. You can see from eh, outside the window. Em, there¿½¿½s a view from outside. Eh there is a view on the [d.] balcony [b¿½¿½.¿½¿½k.n.] and it¿½¿½s dark outside. You do have some light inside. On the left there is a chimney [..mn.], a chimney [..mn.] with a, a fire inside. And next to the chimney [..mn.] on the right you have a pile of books [b.s]. And behind [b..] it I am not quite sure it¿½¿½s a candle. Or it¿½¿½s not really clear to me what,

.what it is. In front of the chimney we can see a, a coffee table quite regular one. And, and on the right corner, just in front, eh, the window, we do see a Christmas tree as I say eh before. Coffee and those are coffee table with a small decoration [d.k.¿½¿½r…n] on it and a chair on the right. Red [ra.d] and grey. Just in front facing, facing the chimney there¿½¿½s a sofa [‘s..f¿½¿½], I think.

C2 did not catch several words such as flat, chimney at the beginning. She guessed the word flat from the context and she noticed the word chimney after the interlocutor repeated the word and began to converge on the target pronunciation. She could not retell the position (under) of the gift, the object (books) next to the chimney, and the color of the chair.

Description of C1:Picture 3 (See Appendix A)

Em, first, when I see the picture it gives me a sense of peace. I think this is a romantic

wedding which was held in a big garden. I think it¿½¿½s a garden inside the house of the

holdster, and many guests [g.s] sitting around the round and chatting with each other,

and beside the guest [g.s], there is a band [b.n] and they really enjoy the music. At

the corner there are many gifts put aside. I think they are sented by the guests [g.s].

Description of C2: Picture 4 (See Appendix A)

Eh, this picture, describe that people are having [h.v.n], are having some kind of

activity in the open air [.a]. And then, and then it¿½¿½s a sunny day and so far away, there

seems to be a beach. And then there are many trees. Tree is very tall and green, and

surrounding the people, surrounding the ground and there is a grass land. There are

many people seems like they are having picnic [p.n.]. They are not having picnic

[p.n.k] for the picnic, seems like they are waiting for some act, activity which is gone

to happen because there is a very big scream in the middle of the crowd. And but the

activity haven¿½¿½t start yet, so the people is enjoying their dinner. Close to me there are

three people. And two fat guys is having their dinner and there are many bikes [ba.s]

lying [.a.n] on the ground. Maybe these people bike [ba.] to this place.

F2 had no problem in retelling the speech except she could not catch the word bikes in the last sentence.

Discussion

French participants were expected to have some difficulties in understanding the Cantonese, but it turned out that they understood their English speech well. The possible explanations might be: (1) sufficient previous input of Chinese-accented speech was helpful when they listened to the speech of the Cantonese. According to the study of Clarke (2000), the Chinese and Spanish participants who were provided with sufficient access to different voices of the same foreign accent of English and required to find the characteristics of the same accent of different voices would adjust to the foreign accent. (2) the speech produced by the Cantonese was more intelligible and comprehensible not only due to the higher accuracy of the pronunciation, but also because of the error type of their pronunciation. Bent, Bradlow, & Smith (2001; cf. Ocke-Schwen & Munro, 2007, p.349) reported that the mispronunciation of the vowels rendered more difficulties in perception than inaccuracy of consonants (e.g. omitting of certain onset or coda clusters) did. Though the Cantonese habitually omitted the consonant in the coda clusters, the French had no difficulties in identifying the words. Here L1 transfer was not a hinder for perception. Contrasted with the unintelligibility in consonants, the French mispronounced the vowels was more confusing.

Experiment 2: Reading Short Passage

Methodology

The compare and contrast between the transcription of two dyad and what they reported afterward, Experiment 2 was conducted to find out the exact phonological variances that hindered the perception. F2 was chosen to read the short passage because (1) C2 reported that the speech of F2 was more accented and she relied on guessing the word meanings from the context while C1 could understand the speech of F1 well. (2) Both C1 and C2 were fully comprehended by F1 and F2. They were familiar to Cantonese-accented speech. Thus, F2 was required to read a short passage created based on the inventory of minimal word pairs (See Appendix C). The recorded speech of F2 was later played to C2.

Find out how UKEssays.com can help you!

Our academic experts are ready and waiting to assist with any writing project you may have. From simple essay plans, through to full dissertations, you can guarantee we have a service perfectly matched to your needs.

View our services

An inventory of vocabulary, based on English Pronunciation Practice written by Arnold and Gimson (1968), was devised to elicit potential segmental variance. As Yavas (2008)suggested, the missing interdental /¿½¿½/ and /e/ and phonemes in French give rise to the mispronunciation such as /s/ and /z/. The affricates /./ and /./ are rendered as /./ and /./. According to Markedness Differential Hypothesis (a), the missing phonemes in French indicated that the French might have difficulty in producing the English phonemes that are more marked (Eckman,1997; cf. Larsen-Freeman, 2000, p.102). Also, Yavas (2008) noted that voiceless stops such as /p, t, k/ are unaspirated in French, especially in the word final position, which is the same as in Cantonese English. In that case, it was supposed that the Cantonese would be able to perceive the phonological variances. Whereas with minor phonological

difference, /t, d/ are dental rather than alveolar, /t/ is not voiceless in French (Yavas,

2008). It is possible that French will produce /d/ instead of /t/ in English. In the information exchange task, F2 attempted the word chimney three times which indicated that /./ would be difficult for French speakers. However, F2 articulated [zis] for this but [d.] for the the. Therefore, /e/ was left out in the inventory. On the other hand, Bent, Bradlow, & Smith (2001; cf. Ocke-Schwen & Munro, 2007, p.349) found out that ¿½¿½errors in word-initial position were more detrimental to intelligibility than errors in other positions¿½¿½, which was taken into consideration when devising the inventory.

In terms of vowel, the phonemes that differ between Received Pronunciation and American English were not considered. Minimal pair words with the phonemes /., ./, /., i/, and /., u/ were devised due to the missing phonemes of /./, /./, and / ./ in French (Lodge, 2009) .

Inventory for Task 2: the minimal word pairs with potential variances for French

1.

/s/ & /¿½¿½/

sink think sin thin seem theme sank thank some thumb

boots booth face faith

2.

/./ & /./

chip ship cheer sheer cheat sheet cheap sheep

watch wash march marsh arch/ash

3. /./ & /./

ten tan mess mass pet pat bed bad bet bat

4. /e/ & /./

main men pain pen sailor seller

5. /./ & /i/

bit beat bit seat bin bean sick seek live leave ship sheep

6.

/./& /u/

books boots pull pool full fool

7.

/t/ &/d/

tear deer time dime

8.

/k/ & /g/

pick pig coat goat

9.

/p/ & /b/

pill bill pen Ben

Results

C2 noticed that she could not rely on the context as she did in the last experiment, which increased the difficulties in terms of identifying the phonological variances. Therefore, she could not catch the gist of the whole passage. The outcomes indicated that the intelligibility was mainly determined by the accuracy of the phonological core e.g. the accuracy of producing vowel and word-initial consonant. On the other hand, the intelligibility of speech might be influenced by the type of task. The speed of reading was faster than the speed of describing picture.

He knows where the pain is and he has to write down his feelings. He is writing a book about a sailor who puts all his time on watching and raising a sheep. One day, he is told [t..d] its¿½¿½ wool [hwu] is not better than an ordinary coat. He is so angry that he throws the beans [b.ns] and pills [p..s] away. His faith [feit] and money are lost but he doesn¿½¿½t want to give his only chip away. The only way to make him feel better is to go for a pool [p.l] and hear the fool cheer from the mass of people in the bar.

From the transciption, L1 transfer frequently occured when F2 attempted those

minial pair words.

Discussion

The nonexistent phonemes in French caused obstacles in producing certain English phonemes. Thus, it led to unintelligibility for NBES. F1 reported that there was reasons that the French usually do not produce the sounds such as /¿½¿½/. They consider putting the tongue outside the mouth as rather impolite. Other than L1 transfer, the habit formation of speech might explain why they did not produce those sounds. In terms of the type of task, Rogers (1997; cf. Ocke-Schwen & Munro, 2007, p.353) found ¿½¿½a significant positive correlation between percent of minimal pairs corrected identified and intelligibility of sentences read by Chinese¿½¿½ based on the dictation of native speakers. However, no correlation was found between minimal pairs identification and intelligibility of a read passage. The results of the study implied that higher accuracy of minimal pair identification of a reading text might not contribute to the intelligibility of a read passage. Therefore, factors such as suprasegmental including intonation and stress might affect intelligibility of a read passage.

Conclusion

In order to probe the phonological factors at the segmental level that might affect intelligibility and comprehensibility in the communication between Cantonese and French as Non-Bilingual English Speakers, two experiments were conducted. Experiment 1 focus on the speech production while Experiment 2 emphasized the speech perception. On the ground of Markedness Differential Hypothesis and the feature of French phonemes contrasted with English phonemes, the inventory of minimal pairs was devised to find out the exact phonological core. Based on the analyzed transcriptions and what the Cantonese participants reported, the accuracy of vowel affected intelligibility the most, then followed with the accuracy of word-initial consonant. The least detrimental to intelligibility were the mispronunciation of word-final consonant and onset or coda clusters.

First of all, to some extent, the research findings in terms of phonological core will contribute to both speech production and perception in the Non-bilingual English communication. Also, it will be revealing for the students who major in English and do not perceive the importance of improving their listening skills especially about how to adjust to listening to foreign-accented English other than Received Pronunciation or American English. Finally, the research findings in terms of phonological core will be helpful for English teachers who are in favor of error correction in the context of second language classes. They can highlight the phonological core rather than just repeating modeling and drilling words.

The recommended research in the future might be to investigate how the input would affect the perception of different foreign-accented English. The research might provide more participants with treatment and might seek the correlation between intelligibility and the natualness of suprasegmentals.

Appendix A

Pictures for Task 1

Appendix B Devised test For Task 2 Reading short passage (Words in the blankets are potential variances generated from the inventory of

minimal word pair) He knows where the pain (pen) is and he has to write down his feelings. He is writing a book about a sailor (seller) who puts all his time (dime) on watching (washing) and raising a sheep (ship). One day, he is told its¿½¿½ wool is not better than an ordinary coat (goat). He is so angry that he throws the beans (bins) and pills (bills) away. His faith (face) and money are lost but he doesn¿½¿½t want to give his only chip (sheep) away. The only way to make him feel better is to go for a pool (pull) and hear the fool (full) cheer from the mass (mess) of people in the bar.

In order to assess intelligibility and comprehensibility of both speaker and listeners factors in the communication between Cantonese and French as Non-Bilingual English Speakers, the speech of four picture descriptions and a read passage were recorded to analyze the phonological variances at the segmental level in two experiments on the ground of markedness of L1 transfer¿½CMarkedness Differential Hypothesis (MDH) in second language acquisition. For Experiment 1, it should be noticed that sufficient input of certain foreign-accented speech will contribute to better perception. For Experiment 1& 2, the results suggest that the accuracy of vowel affects intelligibility and comprehensibility the most and it turns out that the mispronunciation of word-final consonant and onset or coda clusters is not as detrimental to intelligibility as expected.

Introduction

English as an international language is prevailing as a communicative tool among non English speaking countries. According to ¿½¿½A quarter of Chinese study English: official¿½¿½ (2006), more than 300 million Chinese, nearly a quarter of the country’s population, have studied English either as a major course or as an elective subject. Jenkins (2000) claimed that a variety of social contexts, and different English accents might threaten intelligibility especially in communication with Non-Bilingual English Speakers (NBES). According to Skutnabb-Kangas & McCarty (1984), bilingual speakers are able to function in two or more languages, acquire the

communicative and cognitive competence, and establish both L1 and L2 identities in

monolingual or multilingual communities at the same level as native speakers. Since most Chinese learners in mainland China are non-bilingual English speakers, a salient problem emerges since Chinese learners are inhibited to communicate with foreigners and have difficulties understanding different non-native English accents.

Native vs. Foreign and Native vs. Intelligibility

For the last few decades, the discussion on whether there is a ¿½¿½native¿½¿½ accent that L2 learners might actually acquire to get rid of their ¿½¿½foreign¿½¿½ accents for good has drawn the attention of teachers and researchers. Bongaerts (1999; cf. Hansen 2008, p.200) believed that most people who learn English after early childhood are unlikely to speak with a native or native-like accent, except for those who are highly motivated to learn L2 pronunciation, according to Moyer (2004; cf. Hansen 2008, p.200), or those whom Ioup, Boustagui, EI Tigi, & Moselle (1994; cf. Hansen 2008, p.200) considered to have a special talent in L2 learning. Foreign-accented English is common, yet it does not hinder millions of second language users around the world from successful communication. It indicates that intelligibility and comprehensibility rather than accent-free pronunciation is the ultimate goal that L2 learners and teachers should aim at. Smith and Nelson¿½¿½s definition of intelligibility and comprehensibility, referred to the degree of ¿½¿½word and utterance recognition¿½¿½ and ¿½¿½word and utterance meaning¿½¿½ respectively, are further described by Bamgbose (1998; cf. Jenkins, 2000, p.70) as ¿½¿½a complex of factors comprising recognizing an expression, knowing its meaning, and knowing what that meaning signifies in the sociocultural context¿½¿½. He

also notices the importance of speaker and listener factors to speech interpretation.

Hence, intelligibility in production and comprehensibility in perception are inseparable in communication.

Research Objective, Theoretical Framework, and Significance

Previous research in L2 phonology has focused more on the speech accommodation and other social factors such as gender, social identity, and extent of L1/L2 use etc, which investigate the social communication between bilingual speakers

(e.g. immigrants or students abroad) and native English speakers. However, there are few attempts on the NBES communication. Therefore, the research demonstrated whether it is really difficult for Chinese learners to understand French English even if people are speaking the same language; if ¿½¿½Chinglish pronunciation¿½¿½ renders any unintelligibility to interfere inter-cultural communication.

As Jenkins (2000) noted the L1 transfer in NBES communication is important with ¿½¿½speakers of different first languages varying their pronunciation under the influence of their L1 phonology¿½¿½, the research aimed to investigate to what extent L1 transfer affects the L2 speech. Also the phonological variants in the perspective of L1 transfer based on the theory in linguistic markedness proposed by Zobl (1983; cf. Larsen-Freeman, 2000, p.101) were analyzed. They noted that ¿½¿½linguistically unmarked features of the L1 will tend to transfer, but that linguistically marked L1 features will not¿½¿½. Larsen-Freeman (2000) mentioned, for those languages which have voiceless stops but do not have voiced one, voiceless is unmarked while voiced is marked due to their additional phonological feature. The Markedness Differential

Hypothesis (MDH) raised by Eckman (1997; cf. Larsen-Freeman, 2000, p.102)

provided L1 transfer with a systematic framework. He claimed that

(a)

Those areas of the L2 which differ from the L1, and are more marked than

the L1 will be difficult.

(b)The relative degree of difficulty of the area of the L2 which are more

marked than the L1 will correspond to the relative degree of markedness.

(c)

Those area of the L2 which are different from the L1, but are not more

marked than the L1, will not be difficult.

(Eckman,1997; cf. Larsen-Freeman, 2000, p.102)

In order to clarify the voice/voiceless distinction in three word positions, Table

1 shows the possible combination of voice/voiceless in initial, medial, and finial position in different languages. Compare English with French (Corsican), French is less marked. Thus, French might have difficulties in producing English words with the voiced/voiceless consonants in the medial and final position.

Languages with voiced & voiceless contrast English, Arabic, Swedish German, Japanese, Catalan Corsican, Sardinian Not exists

Initial position

¿½¿½ ¿½¿½ ¿½¿½ ¿½¿½/¿½¿½

Medial position

¿½¿½ ¿½¿½ ¿½¿½ ¿½¿½/¿½¿½

Final position

¿½¿½

¿½¿½

¿½¿½

¿½¿½/¿½¿½

Table 1: Three possible word positions in words with a voiced/voiceless distinction established by Dinnsen and Eckman (1975, cf. Larsen-Freeman, 2000, p.102)

On the other hand, Yavas(2008) claimed that obstruents are more difficult than sonorants for Cantonese learners. Since some obstruents and sonorants are permitted as codas in Cantonese, they encountered difficulty in learning single codas of English.

The significance of intelligibility and comprehensibility in intercultural communication means a lot for Chinese learners. Instead of struggling with their ¿½¿½accents¿½¿½ and inhibited to speak out confidently, they will focus on the phonological core in order to realize intelligibility, comprehensibility, and acceptability in their speech. Ultimately, it might enhance their competence of phonological adjustment and cultural awareness.

Literature Review

Jenkins (2000) argued that Smith and Nelson¿½¿½s definition of intelligibility identifies perception processing at the micro-level e.g. recognizing words and utterances in the context but they probably underestimate the influence of pronunciation on miscommunication or even communication breakdown. Pronunciation, as Jenkins (2000) implied, is the crucial factor in unintelligibility in communication between non bilingual English speakers, due to the pattern underlying NBES interaction. They rely much more on identifying the acoustic signal (bottom-up processing) than making use of contextual cues (top-down processing) with limited top-down skills and schemata to perceive the foreign accented speech. Intelligibility and comprehensibility of the NBES communication are thus closely related to inter-speaker segmental variation e.g. L1 transfer and intra-speaker(different L1 speakers communicating with the same L2) segmental variation regarding the degree of speech accommodation (or the strategy of convergence). Beebe and Giles (1984) pointed out that making one¿½¿½s speech respectively more or less like the interlocutor aims to

promote communicative efficiency. In her earlier pilot study (Jenkins, 2000) claimed

same-L1 English interlocutors were more likely to maintain mutual perception than different-L1English interlocutors because of foreign accent and unrelated L1 background. Notwithstanding, she reported that the desire of being understood motivates different-L1 English interlocutors to speak more slowly and articulate more clearly. Intelligibility can be achieved by transfer replacement that different-L1 English speakers would replace their L1 transfer and attempt to converge on target-language forms or the form of interlocutor¿½¿½s IL, especially in information exchange tasks. Jenkins was also concerned about whether it would affect either perception of the interlocutor or production of the speaker and whether it would be considered a high risk for intelligibility problems or a big leap for convergence on acceptable pronunciation.

The contrastive analysis was based on a study of Jenkins (2000) in which six English learners, two Japanese, three Swiss-German, and one Swiss-French (not engaged in same-L1 pairs) at the upper-intermediate level were grouped in same-L1 pairs and different-L1 pairs to complete different tasks (e.g. information exchange task) as preparation for the Advanced English Speaking exam. They completed the questionnaires and participated in interviews as well. The transcriptions annotated the phonological deviants from target-language form to examine for evidence of convergence by replacing L1 transfer errors.

Jenkins (2000) noted two issues during the interpretation of phonological variation and convergence. First of all, the phonological competence of the same-L1 pair was expected to improve because they were recorded six weeks after the

different-L1 pair. However, it was reported that phonological deviations in the same-

L1 pair outweighed those in the different-L1 pair. For example, Japanese substituted /r/ for /./ e.g. flower [.fra..] and beard was mispronounced [b.¿½¿½.]. Both were accountabilities for potential unintelligibility. On the other hand, the Japanese subjects had less non-target form than the Swiss-German subjects because they tended to speak more slowly, pause more frequently or for longer periods, and speak less to avoid losing face or avoid the embarrassment of not being understood. Thereby they consciously or subconsciously replaced L1 transfer deviants or suppressed their L1 transfer errors, which also accounted for convergence reflected in the different-L1 pair.

A relatively comprehensive picture of NBES interaction was provided. A speaker would make considerable effort to replace L1 transfer when he or she communicated with different-L1 English interlocutors.

Critique

The design of Jenkins¿½¿½s research was almost impeccable because she has successfully elicited the participants to describe certain things with pictures during the information exchange task. Even though there were a few sentences with grammatical flaws but overall they were comprehensible in the oral assessment. As Munro (2008;cf. Hansen, 2008, p.208) suggested, controlled production tasks might only tap into the sub-skills and underestimate the speaker¿½¿½s communicative competence: e.g. reading materials may lead to mispronunciations because of orthographic interference, dysfluencies, and unnatural prosody . Thus, sometimes tasks that elicit extemporaneous speech were preferable.

On the other hand, Jenkins (2000) reported that though Japanese took on the Swiss-French speakers speech characteristics during the six-month program, the traditional sense of convergence rarely occurs during the information exchange task. The result indicated that even though mutual perception is built, but the influence of L1 transfer is more salient than convergence.

Experiment 1: Information Exchange Task

Methodology

Inspired by Jenkins¿½¿½s (2000) study, an experiment is going to be implemented to find out to what extent French-accented English renders difficulties for Cantonese speakers and whether Cantonese-accented English is hard for French speakers to comprehend. The first experiment requires four participants in UIC, two Year 4 Chinese students whose L1 is Cantonese and two French speakers, to accomplish the information exchange tasks. What mainly differs from Jenkins¿½¿½s study is: (1)the two Chinese have not converse with the French before; (2) each of the Chinese students listened to the previously-recorded English speech from the French and then retold the content in Cantonese, while the French listened to the recorded English speech from the Cantonese and retold the descriptions in English, which means they had no face to face communication. The issue of face loss was avoided in order to elicit longer speech and convergence was not involved in order to elicit their L2 variance. In addition, two Cantonese (C1 & C2) who major in TESL (Teaching English as a Second Language) have passed the TEM-4 (Test for English Major) but do not know

French. The other two participants (F1 & F2) are French professors in UIC. C1 and F1

are in Dyad . while C2 and F2 are in Dyad .. .

In order to assess the intelligibility, first, the speeches of C1, C2, F1, and F2 were recorded individually. Second, the full texts were written down and the phonemes that deviated from Received Pronunciation or American English were transcribed and annotated in IPA forms. Third, according to Munro¿½¿½s (2008; cf. Hansen, 2008, p.206) approach for assessing L2 speech, besides the researcher¿½¿½s transcription, the responses from participants were taken into consideration. Thus, after the participants in the same dyad listened to the previously-recorded speech, to which they were only allowed to listen to once, any comment relevant to the interlocutors¿½¿½ pronunciation including those unintelligible phonological variances was noted down.

Results The full texts were written and the phonological deviations were marked by IPA forms.

According to what F1 and F2 reported, they had no difficulties in understanding the speech produced by C1 and C2. However, C2 had some problems in identifying certain words in the speech of F2 because of not being able to retell the whole picture (picture 2). The results indicated that the phonological deviations C1 and C2 did not lead to unintelligibility for F1 and F2. Description of F1: Picture 1 (See Appendix A) Ok, it¿½¿½s a western man, I will say, I don¿½¿½t know. Fifty-five, sixty, fifty-five years old. Eh, looks like in the restaurant, he¿½¿½s drinking a beer, and having I don¿½¿½t know, some French fries and salad [s¿½¿½¿½¿½l¿½¿½.]. He¿½¿½s eh, kind of red. Eh, the food doesn¿½¿½t look very tasty [d..st.]. Eh, it¿½¿½s a big beer, he already had more than half. And the restaurant doesn¿½¿½t look very nice. Eh, I don¿½¿½t know it could be many countries. I would say it¿½¿½s a European, yeah that¿½¿½s it. He has mustache. He has mustache, eh, glasses, he¿½¿½s eh, eh becoming bald, not many hairs left. Wearing a white jacket, sweater and a shirt underneath. Eh it¿½¿½s a smoking restaurant. There¿½¿½s an ashtray [..str.] on the table. Eh, I don¿½¿½t know this beer brand, Everest. What else.

C1 reported that she understood most of the speech but she did not catch the words ashtray and the brand of the beer.

Description of F2: Picture 2 (See Appendix A) So this [z.s] is a picture, eh of a flat [f.¿½¿½t]. Eh, obviously [….s..] it is going to be one Christmas, because there is a Christmas tree and a gift just under [..

..] it. And some Christmas, typical Christmas decoration [d.k.¿½¿½r….n]. Eh, the picture was taken at night. You can see from eh, outside the window. Em, there¿½¿½s a view from outside. Eh there is a view on the [d.] balcony [b¿½¿½.¿½¿½k.n.] and it¿½¿½s dark outside. You do have some light inside. On the left there is a chimney [..mn.], a chimney [..mn.] with a, a fire inside. And next to the chimney [..mn.] on the right you have a pile of books [b.s]. And behind [b..] it I am not quite sure it¿½¿½s a candle. Or it¿½¿½s not really clear to me what,

.what it is. In front of the chimney we can see a, a coffee table quite regular one. And, and on the right corner, just in front, eh, the window, we do see a Christmas tree as I say eh before. Coffee and those are coffee table with a small decoration [d.k.¿½¿½r…n] on it and a chair on the right. Red [ra.d] and grey. Just in front facing, facing the chimney there¿½¿½s a sofa [‘s..f¿½¿½], I think.

C2 did not catch several words such as flat, chimney at the beginning. She guessed the word flat from the context and she noticed the word chimney after the interlocutor repeated the word and began to converge on the target pronunciation. She could not retell the position (under) of the gift, the object (books) next to the chimney, and the color of the chair.

Description of C1:Picture 3 (See Appendix A)

Em, first, when I see the picture it gives me a sense of peace. I think this is a romantic

wedding which was held in a big garden. I think it¿½¿½s a garden inside the house of the

holdster, and many guests [g.s] sitting around the round and chatting with each other,

and beside the guest [g.s], there is a band [b.n] and they really enjoy the music. At

the corner there are many gifts put aside. I think they are sented by the guests [g.s].

Description of C2: Picture 4 (See Appendix A)

Eh, this picture, describe that people are having [h.v.n], are having some kind of

activity in the open air [.a]. And then, and then it¿½¿½s a sunny day and so far away, there

seems to be a beach. And then there are many trees. Tree is very tall and green, and

surrounding the people, surrounding the ground and there is a grass land. There are

many people seems like they are having picnic [p.n.]. They are not having picnic

[p.n.k] for the picnic, seems like they are waiting for some act, activity which is gone

to happen because there is a very big scream in the middle of the crowd. And but the

activity haven¿½¿½t start yet, so the people is enjoying their dinner. Close to me there are

three people. And two fat guys is having their dinner and there are many bikes [ba.s]

lying [.a.n] on the ground. Maybe these people bike [ba.] to this place.

F2 had no problem in retelling the speech except she could not catch the word bikes in the last sentence.

Discussion

French participants were expected to have some difficulties in understanding the Cantonese, but it turned out that they understood their English speech well. The possible explanations might be: (1) sufficient previous input of Chinese-accented speech was helpful when they listened to the speech of the Cantonese. According to the study of Clarke (2000), the Chinese and Spanish participants who were provided with sufficient access to different voices of the same foreign accent of English and required to find the characteristics of the same accent of different voices would adjust to the foreign accent. (2) the speech produced by the Cantonese was more intelligible and comprehensible not only due to the higher accuracy of the pronunciation, but also because of the error type of their pronunciation. Bent, Bradlow, & Smith (2001; cf. Ocke-Schwen & Munro, 2007, p.349) reported that the mispronunciation of the vowels rendered more difficulties in perception than inaccuracy of consonants (e.g. omitting of certain onset or coda clusters) did. Though the Cantonese habitually omitted the consonant in the coda clusters, the French had no difficulties in identifying the words. Here L1 transfer was not a hinder for perception. Contrasted with the unintelligibility in consonants, the French mispronounced the vowels was more confusing.

Experiment 2: Reading Short Passage

Methodology

The compare and contrast between the transcription of two dyad and what they reported afterward, Experiment 2 was conducted to find out the exact phonological variances that hindered the perception. F2 was chosen to read the short passage because (1) C2 reported that the speech of F2 was more accented and she relied on guessing the word meanings from the context while C1 could understand the speech of F1 well. (2) Both C1 and C2 were fully comprehended by F1 and F2. They were familiar to Cantonese-accented speech. Thus, F2 was required to read a short passage created based on the inventory of minimal word pairs (See Appendix C). The recorded speech of F2 was later played to C2.

An inventory of vocabulary, based on English Pronunciation Practice written by Arnold and Gimson (1968), was devised to elicit potential segmental variance. As Yavas (2008)suggested, the missing interdental /¿½¿½/ and /e/ and phonemes in French give rise to the mispronunciation such as /s/ and /z/. The affricates /./ and /./ are rendered as /./ and /./. According to Markedness Differential Hypothesis (a), the missing phonemes in French indicated that the French might have difficulty in producing the English phonemes that are more marked (Eckman,1997; cf. Larsen-Freeman, 2000, p.102). Also, Yavas (2008) noted that voiceless stops such as /p, t, k/ are unaspirated in French, especially in the word final position, which is the same as in Cantonese English. In that case, it was supposed that the Cantonese would be able to perceive the phonological variances. Whereas with minor phonological

difference, /t, d/ are dental rather than alveolar, /t/ is not voiceless in French (Yavas,

2008). It is possible that French will produce /d/ instead of /t/ in English. In the information exchange task, F2 attempted the word chimney three times which indicated that /./ would be difficult for French speakers. However, F2 articulated [zis] for this but [d.] for the the. Therefore, /e/ was left out in the inventory. On the other hand, Bent, Bradlow, & Smith (2001; cf. Ocke-Schwen & Munro, 2007, p.349) found out that ¿½¿½errors in word-initial position were more detrimental to intelligibility than errors in other positions¿½¿½, which was taken into consideration when devising the inventory.

In terms of vowel, the phonemes that differ between Received Pronunciation and American English were not considered. Minimal pair words with the phonemes /., ./, /., i/, and /., u/ were devised due to the missing phonemes of /./, /./, and / ./ in French (Lodge, 2009) .

Inventory for Task 2: the minimal word pairs with potential variances for French

1.

/s/ & /¿½¿½/

sink think sin thin seem theme sank thank some thumb

boots booth face faith

2.

/./ & /./

chip ship cheer sheer cheat sheet cheap sheep

watch wash march marsh arch/ash

3. /./ & /./

ten tan mess mass pet pat bed bad bet bat

4. /e/ & /./

main men pain pen sailor seller

5. /./ & /i/

bit beat bit seat bin bean sick seek live leave ship sheep

6.

/./& /u/

books boots pull pool full fool

7.

/t/ &/d/

tear deer time dime

8.

/k/ & /g/

pick pig coat goat

9.

/p/ & /b/

pill bill pen Ben

Results

C2 noticed that she could not rely on the context as she did in the last experiment, which increased the difficulties in terms of identifying the phonological variances. Therefore, she could not catch the gist of the whole passage. The outcomes indicated that the intelligibility was mainly determined by the accuracy of the phonological core e.g. the accuracy of producing vowel and word-initial consonant. On the other hand, the intelligibility of speech might be influenced by the type of task. The speed of reading was faster than the speed of describing picture.

He knows where the pain is and he has to write down his feelings. He is writing a book about a sailor who puts all his time on watching and raising a sheep. One day, he is told [t..d] its¿½¿½ wool [hwu] is not better than an ordinary coat. He is so angry that he throws the beans [b.ns] and pills [p..s] away. His faith [feit] and money are lost but he doesn¿½¿½t want to give his only chip away. The only way to make him feel better is to go for a pool [p.l] and hear the fool cheer from the mass of people in the bar.

From the transciption, L1 transfer frequently occured when F2 attempted those

minial pair words.

Discussion

The nonexistent phonemes in French caused obstacles in producing certain English phonemes. Thus, it led to unintelligibility for NBES. F1 reported that there was reasons that the French usually do not produce the sounds such as /¿½¿½/. They consider putting the tongue outside the mouth as rather impolite. Other than L1 transfer, the habit formation of speech might explain why they did not produce those sounds. In terms of the type of task, Rogers (1997; cf. Ocke-Schwen & Munro, 2007, p.353) found ¿½¿½a significant positive correlation between percent of minimal pairs corrected identified and intelligibility of sentences read by Chinese¿½¿½ based on the dictation of native speakers. However, no correlation was found between minimal pairs identification and intelligibility of a read passage. The results of the study implied that higher accuracy of minimal pair identification of a reading text might not contribute to the intelligibility of a read passage. Therefore, factors such as suprasegmental including intonation and stress might affect intelligibility of a read passage.

Conclusion

In order to probe the phonological factors at the segmental level that might affect intelligibility and comprehensibility in the communication between Cantonese and French as Non-Bilingual English Speakers, two experiments were conducted. Experiment 1 focus on the speech production while Experiment 2 emphasized the speech perception. On the ground of Markedness Differential Hypothesis and the feature of French phonemes contrasted with English phonemes, the inventory of minimal pairs was devised to find out the exact phonological core. Based on the analyzed transcriptions and what the Cantonese participants reported, the accuracy of vowel affected intelligibility the most, then followed with the accuracy of word-initial consonant. The least detrimental to intelligibility were the mispronunciation of word-final consonant and onset or coda clusters.

First of all, to some extent, the research findings in terms of phonological core will contribute to both speech production and perception in the Non-bilingual English communication. Also, it will be revealing for the students who major in English and do not perceive the importance of improving their listening skills especially about how to adjust to listening to foreign-accented English other than Received Pronunciation or American English. Finally, the research findings in terms of phonological core will be helpful for English teachers who are in favor of error correction in the context of second language classes. They can highlight the phonological core rather than just repeating modeling and drilling words.

The recommended research in the future might be to investigate how the input would affect the perception of different foreign-accented English. The research might provide more participants with treatment and might seek the correlation between intelligibility and the natualness of suprasegmentals.

Appendix A

Pictures for Task 1

Appendix B Devised test For Task 2 Reading short passage (Words in the blankets are potential variances generated from the inventory of

minimal word pair) He knows where the pain (pen) is and he has to write down his feelings. He is writing a book about a sailor (seller) who puts all his time (dime) on watching (washing) and raising a sheep (ship). One day, he is told its¿½¿½ wool is not better than an ordinary coat (goat). He is so angry that he throws the beans (bins) and pills (bills) away. His faith (face) and money are lost but he doesn¿½¿½t want to give his only chip (sheep) away. The only way to make him feel better is to go for a pool (pull) and hear the fool (full) cheer from the mass (mess) of people in the bar.

Cite This Work

To export a reference to this article please select a referencing stye below:

Reference Copied to Clipboard.
Reference Copied to Clipboard.
Reference Copied to Clipboard.
Reference Copied to Clipboard.
Reference Copied to Clipboard.
Reference Copied to Clipboard.
Reference Copied to Clipboard.

Related Services

View all

DMCA / Removal Request

If you are the original writer of this essay and no longer wish to have your work published on the UKDiss.com website then please: