Sezgi Acar, University of Alabama, Tuscaloosa, Alabama, USA
Susan Spezzini, University of Alabama at Birmingham, Alabama, USA

Rationale

When words are borrowed into a new language, their ensuing pronunciation is based on the receiving language’s phonological system. As such, loan words in Turkish are pronounced based on Turkish phonology. Instinctively, Turkish speakers continue to employ their Turkish pronunciation when pronouncing these same words in English. The resulting L1 interference greatly reduces these speakers’ L2 intelligibility. This ever-present challenge with loan words has also been experienced by the first author, an L1 Turkish speaker. Her experience with this loan word challenge served as the catalyst in designing the current study.

The challenge of pronouncing words in English that exist as loan words in Turkish can be illustrated by the word program. Though spelled identically in English and Turkish, program is pronounced /ˈproʊ-gɹæm/ in American English, with stress on the first syllable, and /pro-ˈgrɑm/ in Turkish, with stress on the final syllable which is where most Turkish words are stressed (Thompson, 2001). Due to distinct stress patterns between English and Turkish, when Turkish-speaking English learners try to pronounce program in English, they usually stress the second syllable as in Turkish rather than the first syllable as in English. They also usually substitute /oʊ/ with /o/ and /æ/ with /ɑ/. They tend to aspirate the /p/ softly as in Turkish, rather than strongly as in English. They also tend to pronounce a Turkish /r/ with the tongue touching the alveolar ridge rather than an American English /ɹ/ with the tongue curled back.

Study Design

In an effort to reach a more thorough understanding of this pronunciation challenge, we conducted a study of how Turkish speakers pronounce words in English that exist as loan words in Turkish. Our study’s preliminary analysis is provided in this article. We explain how four Turkish-speaking English learners initially pronounced target words in English, how their L1 phonology might have affected their English pronunciation, and how they improved their English pronunciation of these words after 6 weeks of pronunciation instruction. We also describe instructional strategies used for helping these English learners improve their pronunciation.

We conducted this study at a major research university in the southeastern United States. For our study, we selected words that frequently cause communication breakdowns between Turkish speakers and their English-speaking classmates and professors. To guide us in selecting these words, we solicited input from several Turkish-speaking English learners regarding the loan words in Turkish that they felt were most difficult to pronounce in English. Based on their input, we selected 22 target words: program, campus, transfer, internet, method, data, sponsor, theory, category, conference, analyze, problem, perspective, certificate, tunnel, control, status, hypothesis, address, mathematics, seminar, and graphic. For each of these words, we created a sentence using that word in context as illustrated by the word program in the following sentence: Many TV programs have a bad influence on children.

Our participant sample consisted of four Turkish speakers, two females and two males, whose ages ranged from 25 to 32. These participants had come to the United States for different reasons and, at the time of the study, had been living here from 1–10 years. Three participants had started learning English in fourth grade at Turkish public schools; they were taught by nonnative English speakers. The fourth participant had started learning English at age 4 in a Turkish private school; he was taught mainly by native English speakers. All participants received additional English instruction upon arriving in the United States. All participated in this study because they wanted to improve their English pronunciation.

In the first stage of our study, the first author met individually with each of the four participants to collect preassessment data. In our study’s second stage, the first author met weekly for 6 weeks with all four participants together and provided pronunciation instruction; each session lasted approximately 1 hour. In the third and final stage, the first author again met individually with each participant, but now to collect postassessment data. All sessions (preassessment, instruction, postassessment) took place in an unoccupied university classroom.

Preassessment

During the preaassessment session, each participant signed the IRB consent document and completed a 13-item questionnaire about their educational background and English learning experiences. Each participant then read aloud two sets of written prompts. The first set included 22 sentences, each containing one of the target words. The second set included these 22 target words in isolation. Both sets of responses were recorded. This constituted the preassessment data. The first author listened to these preassessment recordings, phonetically transcribed the 22 target words in context and isolation, and analyzed the results.

During the preassessment for reading the target words in context, participants tended to pronounce these words more like the words were Turkish than English. Though uninformed regarding which words were in this study, participants tended to pause or hesitate before reading aloud the target words. These pauses seemed to suggest that participants felt unsure of pronouncing these words. When asked later why they had paused before reading aloud these words, participants explained that they were aware of using Turkish pronunciation for these words and were trying to produce a more English-like pronunciation.

During the preassessment for reading target words in isolation, participants often produced a more English-like pronunciation. This might suggest that participants felt less intimidated when pronouncing a difficult word in isolation. This might be because the target word was neither surrounded by other words nor associated with contextual meaning. Further research is needed to better understand differences with the participants’ pronunciation of target words in context and in isolation.

When participants produced English words with Turkish sounds, they routinely used Turkish stress patterns. For example, instead of pronouncing method as /ˈmɛ-θəd/ as in American English, two participants pronounced it as /me-ˈtod/, which not only approximates Turkish sounds and stress patterns but also represents Turkish spelling. The participant who had lived in the United States for 10 years produced category as /ˈkɑ-tʌ-go-ˈri/, resembling its Turkish pronunciation. Two participants were a married couple and had been living in the United States for almost a year. Both paused before pronouncing target words, especially in context. Yet, though concentrating before reading these words aloud, both produced these target words using Turkish pronunciation. The fourth participant most closely approximated English pronunciation for all target words. As a PhD student, she was cognizant of her challenge with pronouncing words in English that are loan words in Turkish. Shortly after arriving in the United States, she began working at attaining more native-like English pronunciation and thus demonstrated the least difficulty when reading these words in both context and isolation.

Instruction

The first instructional session began with participants sharing their experiences from the preassessment session. All described a major challenge with accurately pronouncing words in English that exist as loan words in Turkish. They felt this difficulty stemmed from how they had been pronouncing these words their entire life with Turkish pronunciation. They concurred that their prior use of these words as loan words in Turkish influenced how they pronounced these words when using them in English. They identified this as the main reason for pausing and thinking before actually attempting to say these words in English. Participants also shared their experiences at being misunderstood when trying to use these words in English.

During each of the six instructional sessions, the first author followed a communicative framework for teaching pronunciation (Celce-Murcia et al., 2011). In her role as instructor, the first author described the target sound or stress pattern; provided listening discrimination activities; led participants in controlled, guided, and communicative practice; and offered cyclical, explicit feedback. Instruction was focused on the sounds and stress patterns represented in the 22 target words rather than on the words themselves. Instructional strategies included listening and imitating, audio and video recording/replaying, phonetic training, minimal pairs in isolation and context, role-plays, tongue twisters, visual aids, vocal tract drawings, websites, the color vowel chart (Taylor & Thompson, 2009), and the International Phonetic Alphabet (IPA).

Postassessment

After 6 weeks of instruction, each of the four participants again met individually with the first author. They responded to the same prompts as in the preassessment. Their recorded responses constituted the postassessment data. Participants also offered opinions on how their pronunciation had improved and on which strategies had been most beneficial. They felt that explicit instruction and feedback were useful in helping them improve their pronunciation and that the instructional sessions had also served to boost their motivation and self-assuredness. Explicit demonstrations with the IPA chart, color vowel chart, and other visuals had enabled them to realize how seemingly similar sounds were actually quite different. By using sensory perceptions other than hearing, they could visually and even kinesthetically ascertain how problematic sounds were articulated and hence different from each other. Moreover, after seeing the visual differentiation of distinct sounds, such as /ɑ/ and /æ/, and of distinct stress patterns, participants felt that they were finally able to begin hearing differences, which heretofore had remained elusive.

After all data were collected, the first author listened to postassessment recordings, phonetically transcribed target words in context and isolation, and analyzed results. Postassessment results were compared with preassessment results to determine whether participants had improved their English pronunciation of the 22 target words. Comparisons indicated that all four participants attained a more English-like pronunciation. In addition to their increased accuracy with producing English sounds and stress patterns in the target words, the participants also demonstrated greater confidence during the postassessment session.

Conclusion

Our preliminary findings suggest that explicit instruction and feedback helped four Turkish speakers attain a more English-like pronunciation of words that exist as loan words in Turkish. These English learners increased their awareness of pronunciation challenges and improved their English pronunciation. Our study highlights the challenge faced by English learners with pronouncing words in English that exist as loan words in their L1. It also demonstrates the importance of explicit pronunciation instruction with adult language learners. Further analysis is underway to expand findings and identify implications.

References

Celce-Murcia, M., Brinton, D. M., & Goodwin, J. M. (2011). Teaching pronunciation: A course book and reference guide (2nd ed.). Cambridge University Press.

Swan, M., & Smith, B. (Eds.). (2001). Learner English: A teacher’s guide to interference and other problems (2nd ed.). Cambridge University Press.

Taylor, K., & Thompson, S. (2009). The color vowel chart. American English.https://americanenglish.state.gov/resources/color-vowel-chart

Thompson, I. (2001). Turkish speakers. In M. Swan & B. Smith (Eds.), Learner English: A teacher’s guide to interference and other problems (2nd ed.), pp.214–226. Cambridge University Press.

Yoshida, T. M. (2016). Beyond repeat after me: Teaching pronunciation to English learners. TESOL Press.

Susan Spezzini, PhD, is a professor of English learner education at The University of Alabama at Birmingham, program director of secondary education and ESL, and principal investigator for professional development grants from the U.S. Department of Education's Office of English Language Acquisition (OELA).