TRACING LANGUAGE TRANSFER : PATTERNING ENGLISH PRONUNCIATION THROUGH ARABIC SOUNDS IN INDONESIA

This research aims to reveal how certain consonants existing in Arabic and English accommodate English for Foreign Language (EFL) students in Indonesia to pronounce English words. The data employed in this research were the sound transcriptions of two groups of English Language Education Department (ELED) freshmen of the University of Muhammadiyah Malang (UMM) when they pronounced several targeted consonants that were embedded into several chosen words. Based on the speech sound transcriptions of the first group of students who graduated from Islamic-based schools, the students were able to pronounce the targeted English sounds correctly. However, the second group showed that many of the students graduating from non-Islamic based schools made numerous errors; they replaced the targeted sounds with the nearest speech sounds.


INTRODUCTION
This study is instigated by a brief review on how each language in this world is different from one another.Several of these distinctions are depicted structurally in each language's linguistic aspect such as phonology, morphology, syntax, semantics and pragmatics.With regards to this issue, experts classify languages into some categories according to the closeness of their features (Slobin, 2006;Shopen, 2007).
Bahasa Indonesia, which belongs to Austronesian language family, obviously has many different features which do not exist in English, a Germanic language, and vice versa.

This somehow results in challenges faced by
Indonesian students in learning English.One of these major challenges is seen in pronunciation (Nguyen, 2011;Pallawa and Alam, 2013).Many of Indonesian students find it problematic to pronounce some words which contain certain sounds such as these two dental sounds /θ/ and /ð/.
Despite the fact that languages in the world are categorized into several language families due to their differences, they share several resemblances; one of which can be noticed in their phonetic features as different language may have similar speech sounds (Lestiono and Gusdian, 2017).A comparative study conducted by Javed (2013)

METHODOLOGY
This study was preceded by a preliminary study of which result was that Arabic positively accommodates Indonesian EFL students to pronounce English words.In

Findings from Islamic-based School Students
Departing from the transcriptions obtained from the students graduating from Islamic-based schools, the ways they pronounce the targeted English words are presented in Table 1.
From the table above, it can be seen that all students pronounced palato-alveolar sound /tʃ/ correctly.Meanwhile, for the transcription of the other four sounds, i.e. /θ/, /ð/, /∫/, and /ʤ/, it was found that some students still made mistakes although the number was very few.For instance, some

Findings from Public School Students
Based on the transcriptions obtained from regular school graduates, the produced consonant sounds of the targeted English words are depicted in table 2 as follows.

Table 2. Consonant Sounds Produced by Students from Public Schools
From the table above, the students were found to commit numerous errors (the writers used 'error', not 'mistake' as the transcription findings of this group showed that the students did not seem to be capable of producing the targeted sounds correctly; they kept producing wrong consonants for different words).
Furthermore, it is also clear that /θ/ and /∫/ deemed to be the most difficult speech sounds to pronounce.For example, all of the students in this group failed to pronounce /θ/ in 'thirty' and 'thought'; then, to cope with the difficulties, they used an alveolar sound /t/ in the replacement of /θ/.In addition, many of them also mispronounced /∫/ sounds in 'she' and 'shy'; instead, they pronounced both words using /s/ sounds as the replacement.
Despite the fact that the students in this groups made a number of erroneous pronunciations, they showed a similar strategy to that employed by the students from the previous group; they produced the sounds that come from the nearest articulator.For instance, they used alveolar sound-/s/ sound to pronounce the words 'she' and 'shy' that should have been using palato-alveolar sound; the articulator of this palato-alveolar speech sound is replaced with its neighbouring articulator, alveolar ridge, on which /s/ sound vibrates.
Their failure in pronouncing the aforementioned consonants may be due to the students' lack of knowledge in these targeted sounds.Furthermore, the sounds' absence in Bahasa and their limited exposure to Arabic consonants may become the plausible reasons of this issue.Accordingly, the adjacent speech sounds such as /s/ are used as they are the closest ones to the targeted sounds.
The realization of using the closest speech sounds also represented a cross-linguistic transfer that the students did (Jarvis and Pavlenko, 2008).In this case, the students apparently employed their phonological knowledge in Bahasa in order to cope with the difficulty they encountered when coming across with certain English words.

CONCLUSION
In the light of the data analysis, the writers summed up that the transcriptions of the students who graduated from Islamicbased school confirmed that they were able to pronounce the targeted English sounds correctly.This showed a link to their intensive exposure to Arabic sounds in their past.On the other hand, the transcription results from the second group exposed that these students failed to produce the expected sounds.Again, this might be connected to their limited exposure to these speech sounds both in Arabic and English before.
Meanwhile, both groups also demonstrated a resemblance in the way they dealt with the pronunciation mistakes/errors; they tended to do a language transfer, in which they employ their phonological knowledge of producing speech sounds in Arabic (for the former group) and Bahasa (for the latter group) in order to pronounce the targeted English words.
detailed, twelve transcripts taken from two different groups of ELED freshmen of the University of Muhammadiyah Malang (UMM) were used as the data.The first group were six students graduating from Islamicbased schools; meanwhile, the second group consisted of the other six students graduating from public schools; whose exposure to Arabic was very limited.These students were to record their voice while reading a short story containing several targeted sounds, i.e. /∫/, /θ/,/ð/,/tʃ/, and /ʤ/.After being recorded and transcribed, these transcripts were eventually highlighted and analysed in order to find out how these students produced the targeted sounds in pronouncing several specific words.

Table 1 .
Consonant Sounds Produced by Students from Islamic-based Schools students mispronounced the word 'birthday' and 'brother'; instead of using /θ/ sound in the former and /ð/ in the latter, they employed some alveolar sounds such as /t/, /d/, or /s/ sounds.The use of these alveolar sounds are quite understandable as the position of alveolar ridge as the passive articulator of alveolar sounds is just right behind the upper teeth as the passive articulator of /θ/ and /ð/ sounds.The other example was also grasped from /∫/ sound.Some students replaced this palato-alveolar sound with alveolar sound /s/.This realization also indicated that the students employed the sounds which are close to the actual sounds-as it is seen that palate as the articulator of /∫/ is just behind the alveolar ridge, where /s/ sound takes place.e-ISSN: 2549-8673, p-ISSN : 2302 -884X 5 Regardless of some of the mispronounced words, most students did pronounce the targeted words correctly as the mistakes found were very rare (the writers picked up 'mistake' instead of 'error' because the transcription of this group revealed that the students were able to produce these targeted sounds in different words properly; this has shown that they know how to produce these phonemes).It depicts the ability of the students to employ their knowledge in Arabic consonant sounds in pronouncing English words.In other words, they may not find any difficulty in pronouncing the targeted words in English since they are already familiar with the sounds when learning Arabic back at their high schools.This result confirms the previous study by Lestiono and Gusdian (2017) that Arabic consonant sounds assist Indonesian EFL students to pronounce English words.The link between the produced sounds and their exposure back at school is simply a result of a cross-linguistic transfer-applying knowledge of one language to the use of another language (Jarvis and Pavlenko, 2008).