THE PHONOLOGICAL ANALYSIS OF A JAPANESE SINGER : THE CRITICISM OF USING SONGS IN ENGLISH

This study deals with the phonological analysis of a Japanese singer in some English sounds in live performance music videos. It aims to analyse and describe the pronunciation process of Japanese singer in producing tense vowels, diphthongs, and liquids. It also criticizes the use of songs in English teaching based on the findings. The data are from live performance videos of Japanese singer. The researcher watched and listened to the videos. After that, the researcher listed the words in the lyrics of the songs containing tense vowels, diphthongs, and liquids. The words were transcribed into standard phonetic transcription. Then, the researcher transcribed the words into the phonetic transcription according to the pronunciation of the singer. After that, the data were classified into three parts which were the words containing tense vowels, diphthongs, and liquids. After the researcher analyzed the videos, the findings of this study showed that the Japanese singer as an EFL speaker had different pronunciation process in producing tense vowels, diphthongs, and liquids sound. The findings also gave the criticism on the use of songs in English teaching in a form of caution and contribution


Introduction
As an international language, English has been learned by many people around the world.It is in accordance with Crystal's statement (2003) that English becomes the language which is most widely taught as a foreign language so that the number of second language speakers increases vastly.In this case, there will appear a lot of accents which are produced by the foreign speakers of English such as Japanese accents, Chinese accents, Korean accents, Spanish accents and so on.It is a clear sign that the sound patterns or structure of their native languages influence the speech or production of their second language.In short, it is very equitable to say that the nature of a foreign accent is determined to a large extent by a learner's native language (Avery & Ehrlich, 1992).
The sound patterns of language or usually called as phonology of a certain language raises some problems for the foreign speakers in pronouncing English words since the phonology of a certain language is different from the phonology of English.The difference can appear from the absence of sounds in a certain language but they are present in English.One example of the language is Japanese.As Kenworthy (1987) said, Japanese has only five vowels in its vowel inventory namely /i/, /e/, /a/, /u/, and /o/.A system is quite common among many natural languages in the world.It is different from English which has fifteen vowels namely /i/, /ɪ/, /e/, /ᴈ/, /ᴂ/, /ə/, /ʌ/, /u/, /ʊ/, /o/, /ɔ/, and /a/.The difference of Japanese and English vowel systems is also indicated from the existence of lax and tense in those two vowel systems.The difference between tense and lax vowels is made according to how much muscle tension or movement in the mouth is involved in producing vowels (Ladefoged, 1982).Some of English vowel systems are tense such as /a/, /i/, /u/, /e/, /o/ and all of Japanese vowel systems are lax.
In addition, there are differences in consonantal distribution between Japanese and English.According to Avery and Ehrlich (1992) and Kenworthy (1987), there are more consonants in English than Japanese.Japanese does not have fricatives and affricates which are much more widely distributed in English such as /f/, /v/, /θ /, /ð/, / ʃ /, / ʒ /, /ʧ/, and /ʤ/.In addition, Japanese has a liquid /r/ which is different from the /r/ or /l/ sound of English.The exact of articulation point of /r/ sound in Japanese is not specific.
This study discusses the phonological systems in Japanese which influence the Japanese singer in pronouncing the English words.The study describes the process of Japanese singer in making the sounds of tense vowels, diphthongs, and liquids.From the findings, the study gives the criticism on the using of songs in English teaching.As it is known that many believed that using songs in teaching English is good, the finding of the study notes that not all English songs can be used to teach English.According to the finding, a singer did some mispronunciations in English words containing liquids, vowels and diphthongs.Therefore, the study gives the implication and caution for teachers to be more careful in selecting the songs will be used to teach English.

Phonological Analysis
Phonological has meaning as relating to the nature of sounds of certain language, in this case is English (Fromkin, Rodman, & Hyams, 2007).Therefore, phonological analysis can be defined as the theoretical and practical steps in analyzing the sound patterns of English produced by foreign speakers and the strategies they use to solve the problems in producing the English sounds (Jenkins, 2000;& Nathan, 2008).
Steps of phonological analysis can be divided into three major steps.As Veen and Mve (2010) said, the first step of phonological analysis is carrying out preliminary inquiries in which the researcher has to determine the clear goals of the research and the sound representations.The second step is performing phonetic transcription.This is the necessary step yet it consumes a plenty of time to work on.The final step is analysis.Veen and Mve (2010) also stated that in working on phonological analysis, there are some important things to consider.Being rigorous, meticulous and accurate are must in working on phonological analysis, especially in terms of transcription, data management and storage, analysis, and description.They are very important since doing a phonological analysis is a task that combines specific analytical skills, techniques and reasoning (describing problems and giving solutions).The ability to manage the time and being patient are also important since when the researcher does the phonetic transcription it will take a plenty of time.It requires careful observation in watching and listening.In addition, it requires repetition in identifying the exact pronunciation of the subjects.

Vowels
McMahon (2002) said that vowels refer to the sounds which are produced on a pulmonic aggressive airstream, with central airflow.Vowels can also refer to an unobstructed sound produced by unimpeded airflow that moves up from the larynx and out through the lips (Roach, 2009, p. 10).Vowels have some types, three of which are diphthongs, tense and lax A diphthong is a sequence of two sounds, vowel plus glide.For example: bite [bajt]; there are [a] vowel and glide of [j].Another example is bout [bawt]; there are vowel of [a] and [w] glide (Fromkin, Rodman, & Hyams, 2007).This kind of vowels do not exist in Japanese since as Kenworthy (1987) said, Japanese has only five vowels.
Tense vowels are the vowels which are produced with greater tension of the tongue muscle than its counterpart, and they are often a little longer in duration (Fromkin, Rodman, & Hyams, 2007) [u] and diphthongs are tense in English, while in Japanese, there is no tense or lax differentiation.In his journal, Ohata said that the tense/lax vowels pairs of English such as /i/ vs. /ɪ/, /e/ vs. /ε /, /u/ vs. /ʊ/, do not exist in the five-vowel system of Japanese.
According to Fromkin, Rodman, and Hyams (2007), lax vowels refer to the vowels which are produced with smaller tension of the tongue muscle than its counterpart, and they are shorter in duration.Some examples of lax vowels are /ʌ/, /ɪ/, /ʊ/, /ᴈ/, and /ɔ/.

Liquids
Liquids refer to the sounds produced by some obstruction of the airstream in the mouth, but not enough to cause any real constriction or friction (Fromkin, Rodman, & Hyams, 2007).There are two liquids in English: [l] and [r].In Japanese liquids are not exactly the same as in English.This issue stems from the lack of a separate /r/ and /l/ sound and the difference in place of articulation between Japanese and English /r/ sounds.There are two types of liquids namely trilled and flap.
A trilled "r" can be found in the word "perro" which means dog in Spanish.They produce the /r/ very clear.Therefore, according to Fromkin, Rodman, and Hyams (2007), a trilled "r" is the sound produced by rapid vibrations of an articulator.
A flap is produced by a flick of the tongue against the alveolar ridge.It sounds like "a very fast d (Fromkin, Rodman, & Hyams, 2007)."This kind of sound can be found when American pronounce the word "writer" in which it is almost the same as pronouncing the word "rider".A flapped "r" is symbolized as [ɾ], so the word "writer" will be pronounced as [raiɾər].It also happens when Japanese students say an English word which has liquids.It is different from English in which /l/ is an alveolar lateral or approximant and the /r/ is a post-alveolar approximant.Japanese students produce a single liquid voiced consonant that combines the two sounds of /l/ and /r/ as it is produced by a very quick tap (also called a "flap") of the tongue tip on the alveolar ridge (Suski, 1931, p. 70).

Method
This study dealt with the description of sound production analysis, which focused on the pronunciation process when Japanese produced liquids, diphthongs, and tense vowels.This study described how the sounds were produced.
In this study, the sources of data gathering can be specified into document analysis.As Ary, Jacobs, and Razavieh (2002) said that document analysis refers to a method of research utilized to written or visual materials in a purpose of identifying specified characteristics of the material or a project that focuses on analyzing and interpreting recorded materials within its own context.From the description of document analysis, live performance videos from Japanese singer can be indicated as the visual materials which are going to be identified and analyzed the errors.
As Ary, Jacobs, and Razavieh (2002) stated, there are six steps in document analysis which should be done.The first step is specifying the phenomenon to be investigated.In this step, the researcher specified the phenomenon of pronunciation process of Japanese singer in producing tense vowels and liquids.Secondly, the researcher selected the media to analyze which are the videos.The third step is formulating exhaustive and mutually exclusive coding categories so that the verbal or symbolic content can be counted.Then, the researcher specified the sample in this case video based on the year or period when they performed.It aims to obtain a representative sample of the documents.The next step is training the coders so that they can consistently apply the coding scheme that has been developed and thus contribute to the reliability of the content analysis.Finally, the data was analyzed and it involved the descriptive accounts since this study used qualitative research.
The materials could be analyzed from the textbooks, newspapers, web pages, speeches, television programs, advertisements, videos, musical compositions, or any of a host of other types of documents.The reason why the researcher wanted to use the live performance videos as the data because they showed the originality of the singer when he sang a song in English.The videos did not use the process of editing and it was as real as it was.In addition, from that originality the researcher was able to see the authentic process when the Japanese singer pronounced the English words.The pronunciation of the Japanese singer was analyzed from the lip movement and mouth so that the clearer sound transcription could be interpreted.
In analyzing the data, there were some steps which should be undergone by the researcher.The following is the steps in analyzing the data.1.
The first thing to do was watching the videos and listening to the songs 2.
Secondly, the researcher found the words containing liquids.

3.
Next, the words found were transcribed into phonetic symbols.

4.
The writer inserted those in the table of variation of phonetic transcription, as can be seen in table 1, so that the difference from the standard phonetic transcription could be seen.5.
The next step was categorizing the findings into three major groups, which were word or word group with tense vowels, word or word group with diphthongs, word and word group with liquids and combination.6.
After inserting the phonetic transcription, the researcher tried to find the difference between the standard phonetic transcription and variation.7.
From the differences, the researcher described the process of pronunciation of the Japanese singer when producing tense vowels and liquids by relating the theories how the process of pronouncing the tense vowels and liquids were with the pronunciation produced by the Japanese singer.

8.
Describing what made the Japanese singer produce the sounds of tense vowels and liquids differently in which the strategies of how they made the sound could be indicated in table 1.

Table 1. Sounds Production
Finally, the study will give the criticism on the using of songs in English teaching based on the finding.By providing the table 2, the study will discuss mispronunciation cases found in English words containing liquids, diphthongs, and tense vowels which will affect the use of songs in English teaching in table 2.

Findings and Discussion
The findings of the study answered the first research problem by describing the process of pronunciation of the Japanese singer in producing tense vowels, diphthongs, and liquids.Then, the second research problem was answered by criticizing the use of songs in English teaching based on the mispronunciation done by the Japanese singer when he produces the tense vowels, diphthongs, and liquids.

The Pronunciation Process of Tense Vowels, Diphthongs and Liquids Produced by the Japanese Singer
This part discusses the pronunciation process of Japanese speakers which are represented by the Japanese singer.This part has three sections of discussion containing the pronunciation process when he produced tense vowels, diphthongs and liquids.

Tense Vowels
In two videos, the researcher found thirty nine single words containing tense vowels.From the videos, the researcher analyzed that the singer of the band L'Arc-en-Ciel produced words containing tense vowels in a different way.Even though Japanese vowel system has sounds /a/, /i/, /u/, /e/, and /o/, it does not have exact pronunciation between tense and lax.It means that in English, there are specific differences between tense and lax vowels, such as /i/ and /ɪ/, /u/ and /ʊ/, /e/ and /ᴈ/ and so on.However, although long vowels of Japanese are sometimes analyzed as having the same quality as English tense vowels, this claim is difficult to support because those vowels of Japanese are not always contrastive in nature as the English tense/lax vowel pairs (Vance, 1987).This case occurred when the singer produced /i/ sounds.Several examples are provided in Table 3.The table showed that the singer produced /i/ sound in lax.In English, it is usually called short /i/ which is symbolized as /ɪ/.When the singer said the word me, he pronounced it as /mɪ/ instead of /mi/.It also happened when the singer pronounced street, knee, and see.He pronounced all of them as /strɪt/, /nɪ/, and /sɪ/ in which the /i/ sound which is supposed to be the tense vowel became /ɪ/.There were also other examples such as the words eternity and venality.In English, eternity and venality are pronounced /ɪtᴈ:nəti/ and /vi:nᴂləti/, yet the singer pronounced them as /etᴈ:nətɪ/ and /vənalətɪ/.Those words also showed the same results that the singer produced /ɪ/ sound instead of /i/ especially in the final syllable from the other words such as deeply and heavenly.In the videos, the researcher also watched the vocalist did not totally do unrounded lips.In fact, Fromkin, Rodman, and Hyams (2007) said that when someone produces /i/ sound he or she will have the lips in the shape of smile which is totally unrounded.Meanwhile, the vocalist's teeth were still not closed which means the lips were not in the shape of smile.
In addition, when the singer pronounced words me, knee, and see, he actually had produced impossible words in English.It means that lax vowels mostly did not appear at the ends of English words.For example the words me, knee, and see should be pronounced as /mi/, /ni:/, and /si:/ in English yet the singer which is a Japanese pronounced them as /mɪ/, /nɪ/, and /sɪ/ all of which have no meaning in English words.It comes from Fromkin, Rodman, and Hyams (2007) who stated that lax vowels do not occur at the ends of words, so [sɪ], [sᴈ], [sᴂ], [sʊ], and [sʌ] are impossible words in English.
From the data result, it can be concluded that the singer who is a Japanese speaker did not pronounce the tense vowels clearly.The singer produced the tense vowels in lax.It means that the singer did not spread the lips when he produced the words me, knee, and see.In fact, Ladefoged (1982) stated that the difference between tense and lax vowels is made according to how much muscle tension or movement in the mouth is involved in producing vowels.In this case, the muscle tension or movement in the mouth occurred when the singer spread his lips.

Diphthongs
The second section is diphthongs.In order to make the discussion clearer, the meaning of diphthongs was explained concisely here.As Fromkin, Rodman, and Hyams (2007) stated, diphthongs are the sounds produced by the English speakers by adding the short /j/ and /w/ glides when they produce front and back vowels.For example, English speakers pronounce survey as /sᴈveɪ/ in which the front vowel /e/ is followed by a short /j/ glide and they are symbolized as /eɪ/.
Discussing the analysis result, there were fifty eight single words and word groups containing diphthongs found from two videos.The researcher analyzed that there were several different pronunciation done by the vocalist when he produced diphthongs.The different pronunciation occurred because the vocalist often pronounced diphthongs only the tense vowels without the /j/ or /w/ glides.There were several examples of the different pronunciation provided in Table 4. From the table, it could be seen that most of different pronunciation occurred because the vocalist did not pronounce the glides /j/ or /w/ clearly.It occurred because there are only five vowels in Japanese vowels inventory (Kenworthy, 1987).He pronounced the words rolling, cold, and twilight as /lɔ:lɪŋ/, /kɔ:ld/, and /twɑ:let/ in which he only pronounced the vowels.One example came from the word down.In the word down, /a/ becomes /aʊ/ since /a/ is added with /w/ glide so the word is pronounced as /daʊn/.However, the singer pronounced it as /dɔ:n/.The singer did not pronounce the glide clearly, so that the sound was like /o/ in lax which becomes /ɔ/.Another example is snake./e/ became /eɪ/ since /e/ is added with /j/ glide, so the word is pronounced as /sneɪk/.However, the singer only pronounced the tense vowel [e] without the [j] glide and it became /snek/.
The different pronunciation also occured in the other words such as so and wake.The words should be pronounced as /səʊ/ and /weɪk/.Yet, the vocalist pronounce them differently and the words became /so/ and /wek/.It happened since there are no vowels which are added with the glides in Japanese vowel system.It is supported by Okada's statement (1991, p.94) which is showed in the vowel chart that the vowels of standard Japanese are only and [o̞ ].They only have simple vowels usually called as monophthongs (Fromkin, Rodman, and Hyams, 2007).
The data result indicated that the singer did not pronounce the diphthongs clearly.In this case, the singer had produced the diphthongs only in one sound which means that the glides were not produced directly after the vowels.For example, when the singer said the word down, he pronounced that word as /dɔ:n/ instead of /daʊn/.Therefore, it is different from what Fromkin, Rodman, and Hyams (2007) had stated in their book that diphthong is two sounds containing vowel and glide which are produced in a sequence.

Liquids
From the two videos, the researcher found sixty four single words and word groups containing liquids.Each word had the liquids which were different in place.The liquids were placed at the beginning, within, and the end of the words.Even there were some words containing cluster with /r/ and /l/.Even though Japanese has liquids, there are no specific differences in producing the liquids because of the lack of a separate /r/ and /l/ sound and the difference in place of articulation between Japanese and English /r/ sound.It means that in English, /l/ is an alveolar lateral or approximant and the /r/ is a post-alveolar approximant, yet Suski described "the Japanese produce a single liquid voiced consonant that combines the two sounds of /l/ and /r/ as it is produced by a very quick tap (also called a "flap") of the tongue tip on the alveolar ridge" (as cited in Smith, 2012, p.202).Akamatsu (1997) also stated that /r/ is an apical postalveolar flap undefined for laterality.Therefore, it is specified as neither a central nor a lateral flap.Thus, it raises some evidences from the videos.The examples of evidences are provided in the Table 5.The table showed that the singer had difficulty in pronouncing /r/ and /l/.When the liquids at the beginning of the word like the word rolling, the vocalist pronounced it as /lɔ:lɪŋ/ in which the liquid /r/ became /l/.The singer even sounded like omitting the liquids when they were within and at the end of the words.For example, survive and holding, the words should be pronounced as /sərvaɪv/ and /hoʊldɪŋ/ yet the singer pronounced the words without the liquids.Therefore, the pronunciation became /səpɑ:/ and /hodɪŋ/.It also occurred in the words your and dare in which the singer pronounced them as /jᴈ:/ and /dᴈ:/.Other examples are the word grown and blood.In the word grown /groʊn/ the singer sounded like pronouncing /r/ as /l/ which became /glon/.In addition, the vocalist could not produce the liquid clearly in the word blood /blʌd/ then it became /bʌd/.The inconsistency happened because of the absence of exact sound for liquids in Japanese consonants system.
Thus, it can be concluded that the singer which is a representation of Japanese speaker had problem in pronouncing /r/ and /l/.The singer often exchanged /r/ to /l/ or even he did not pronounce /r/ or /l/ sound.This problem raised because of a great deal of difficulty for Japanese speakers as represented by the singer both in producing and distinguishing /r/ and /l/ which is caused by the lack of a separate /r/ and /l/ sound in Japanese (Smith, 2012, p. 202).From the results, it can be concluded that the Japanese singer produced tense vowels, diphthongs, and liquids in different ways.In English his pronunciation can be clear, yet he produced non-standard pronunciation.

The Criticism on The Use of Songs in English Teaching
Using songs in teaching English is believed to be effective and fun, in this case teaching pronunciation.As Kramer (2007) said, songs are funny and interesting for students.The students can learn English pronunciation very easily by singing the songs.Paul (2003) also stated that by listening to the song, the students can remember the pronunciation easily and draw more deeply into the lesson.In addition Philips (1993) said that by imitating the singer in singing the song, it can help the students know how to pronounce English words correctly.
Despite the effectiveness of using songs in teaching English pronunciation, the findings of the study gave the caution for the teachers in selecting the English songs for the students.The findings provided the cases showed in the table 6.The cases showed in the table 6 can be indicated that the singer did mispronunciation in the words containing tense vowels, diphthongs, and liquids.Considering these findings, the teachers should be more careful in selecting the songs.The teachers must listen and pay more attention to the singer's pronunciation in singing the English songs whether the singer pronounce the words correctly or not.This is important since the students will listen to the songs and imitate how the singer pronounce the English words.Therefore, choosing a singer with good pronunciation will lead the students to produce correct pronunciation in English words.

Conclusion
The study concluded that the Japanese speakers, as represented by the singer of a group band, have different pronunciation process from English speakers in producing English words containing tense vowels, diphthongs, and liquids.These differences are caused by the non-existence of some sounds in their phonological system.When the singer produced English words containing tense vowels the data showed that the singer pronounced them in lax because the vowels in Japanese phonology system do not have contrast sounds in nature as the English lax or tense vowel pairs (Vance, 1987).The singer also produced the diphthongs without pronouncing the glides /j/ or /w/sue to the limited vowels of Japanese which only have five vowels ten sounds compared to English with five vowels twenty four sounds (Tsujimura, 2007).In addition, when the singer pronounced the liquids /r/ or /l/, he did not pronounce them clearly.This problem occurred because of the dilemma experienced by the singer since there is no clear separate /r/ and /l/ sounds in Japanese (Suski, 1931) Based on the findings, the researcher wants to give the insights of the diversity in English accent to the EFL (English as a Foreign Language) teachers.Nation and Newton (2009) stated that the EFL speakers often have pronunciation problems in speaking English influenced by the first language of the speakers.The researcher hopes, the EFL teachers can consider the barriers faced by Japanese learners when they speak in English because of their first language.Therefore, this will help EFL teachers not to misunderstand the words uttered by the Japanese learners.
Finally, the study criticized the use of songs in English teaching.In spite of some beliefs in the effectiveness of using songs in teaching English pronunciation (Kramer, 2007;Paul, 2003;& Philip, 1993), this study gave the caution for the teachers in using songs to teach pronunciation in English.The teachers have to consider the singers first when they want to use the songs as the teaching media.The teachers have to listen and pay more attention to the pronunciation of the singer whether it is correct or not.As a result, this consideration can lead to the better understanding of the students in pronouncing the English words correctly..