second score. If students can score at least 75 out of 100, they are eligible to move to higher levels. If their final grades are below 75, they have to take that semester again till they can pass it.
All participants were 18 to 27-year-olds. To make sure of the homogeneity of participants in terms of English proficiency, listening comprehension ability and pronunciation, an oxford placement test (OPT) was conducted on 80 upper Intermediate students. Having calculated the mean and the SD, 40 participants with the score of 1 SD above and below the mean (1SD ± mean) were selected to conduct the study. Then students were randomly selected into two groups of experimental and control. In both experimental and control groups there were 20, in total 40 students. Figure 4 shows the details about the participants.
Experimental Group Control Group Total
Female 20 20 40
Figure5. Participants of the study
All the students were following the same book; the core course book was the Iran Language Institute’s book.
3.3. Instruments and Materials
The instruments used in this study were:
1) An Oxford Placement Test (OPT) (Appendix A) which is a standardized Cambridge exam so reliability of the test is not needed to be tested.
2) Listening tests used to determine the students level of listening comprehension before and after the contrastive based pronunciation training (Appendix B). The tests contained 35 multiple choice questions requiring the participants to choose the best response from the options according to the statement they hear in the recording. The tests were developed using the TOEFL test which made 70 questions in total.
3) The material (see Appendix C) for the training employed was a pronunciation training pack collected by the help of the researcher’s experienced teachers from three different sources based on their past experience and observation as teachers within classes regarding the common problematic areas experienced by the learners as will discuss in this chapter.
These areas as problematic areas were also pointed out by several researchers (e.g., Brown, 2006; CoĢkun, 2011, 2008; Demirezen, 2011; Field, 2005).Acknowledging these problematic areas, the researcher provided brief explanation and examples followed by exercises. The sources used were Longman complete course for TOEFL test (by Deborah Philips), A course in phonetic ( by Peter Ladefoged), American Accent Training: a guide to speaking and pronouncing American English for everyone ( by Ann Cook) and Clear Speech: pronunciation and listening comprehension ( by Judy B. Gilbert)
3.4. Data Collection Procedures
Firstly, the researcher asked for the consent of Iran Language Institute. After agreements were granted, the researcher decided on which features of pronunciation to include in the training before preparing the materials. The researcher selected the features in terms of her teachers’ experiences within classes regarding the common problematic areas experienced by the learners. Some of the most frequent errors for Iranian EFL learners were replacing those sound which are absent in their own language with those present in Farsi, replacing short vowels with long ones, difficulty in producing diphthongs, producing initial and consonant cluster as barriers for speaking and making them easy by transferring to some similar sound in their own language and making new words and the result of such transfers on their listening comprehension ability. Such observation of L2 pronunciation errors mentioned would naturally suggest the critical need for ESL/EFL teachers to become more aware of the impact that the learners’ L1 sound system and syllable structure will bring to the learning of English pronunciation. To achieve this awareness, Contrastive Analysis can convey insights into the differences and similarities between the L1 and L2 phonological characteristics. This assumption was clearly stated by Lado (1957) in Linguistics across Cultures as follows:
Individuals tend to transfer the forms and meanings, and the distribution of forms and meanings of their native language and culture to the foreign language and culture –both productively when attempting to speak the language…and receptively when attempting to grasp and understand the language… as practiced by natives. (p. 2)
Like English and Persian, many languages of the world, are alphabetic in the sense that they represent their vowels and consonants in the form of letters in their orthography. In these languages, words are composed of one or more syllables. According to Windfuhr (1979), Persian is characterized as a syllable-timed language. In other words, the syllables are said to occur at approximately regular intervals of time, and the amount of time it takes to say a sentence depends on the number of syllables in the sentence, not on the number of stressed syllables as in stress-timed languages like English and German ( Nosratnia, M., & Zaker, A., 2014, p. 100). Persian and English belong to the Indo-European language family. They are different in alphabet, sound system, and syllable structure.
The Persian alphabet is based on Arabic, which is a consonantal system and contains thirty two letters: twenty three consonants and six vowels as well as two diphthongs and a total of 29 phonemes (Samareh 2000; Windfuhr, 1979). Of the six vowels, there are three lax vowels (/a/, /e/, /o/) and three tense vowels (/æ/, /i/, /u/) as well as two diphthongs /ei/, /ou/ and a total of twenty nine phonemes (Samareh, 2000; Windfuhr, 1979). Persian vowels do not have any variation in length in formal speech; however, in informal speech when vowel length changes due to compensatory lengthening, the meaning of the word will not be affected. For instance, in these Persian words ‘begu’/begu/ ‘tell, we can change the length of vowels in each case and L1 Persian speaker will understand it.
At the beginning of the research process, the participants were informed of the purpose of the study. Before starting the training, a pretest was applied to all students, and the treatment started the week after. The researcher conducted the lessons twice a week. The teacher/ researcher gave about 30 minutes of pronunciation instruction to experimental group.
For the purpose of this study the researcher conducted lesson base on introduction of vowels, consonants, numbers of them in each language and compering them as showed in the figures below. English alphabet contains 26 letters: 44 consonants, 12 vowels, 8 diphthongs and a total of 44 phonemes (Sousa, 2005). A comparison between the Persian consonant system and that of English consonant system revealed noticeable differences in the distribution of consonants. As stated in Yavas (2006), theoretically, the overlay of the Persian consonants on the English inventory results in the following:
Figure6. Overly of Persian consonants on the English Inventory (Yavas, 2006, p. 197)
Participants were taught that the four consonants squares are absent in the Persian consonantal system and the r-sound in the circle presents a phonetic differential with its allophones. As with the differences in the consonant systems, there are also noticeable expected differences in vowel systems between Persian and English.
Figure7. English and Persian vowels (Yavas, 2006, p. 197)
After explanation, in each session the researcher used the table below to illustrate one of these difficulties and errors make by learners. In each sessions, the researcher distributed necessary papers students needed to have to practice (Appandix C). After teaching these elements, in each session, the researcher asked students to repeat the correct forms of the absent words. Then they worked on training packed as mentioned.
Definition of the problem
Substitution of / ə/ (Mid and central vowel) with /e/ (Front, short vowel) or / /ɜ:/ (Mid-central long vowel), e.g. /e’baʊt/ for /ə’ baʊt/
The absence of / ə/ in Persian and substitutes resulting in ease of articulation.
Some diphthongs are replaced with a long vowel or a short vowel e.g. /aʊ/ pronounced as /ɒ/ e.g. or for our
In terms of length, they are similar (Roach, 2009, p. 17), but the glide or movement is eradicated. This results in ease of articulation by using long vowels in Persian instead of unusual diphthongs.
Substitution of /s/ or /t/ for /θ/ (Dental fricative), e.g. thank
The error originates from the absence of /θ/ sound in Persian. Learners will produce the nearest available sounds as they perceives.
Substitution of /z/ or /d/ for /ð/ ( Dental fricative), e.g. then
The error originates from the absence of / ð / sound in Persian. Learners will produce the nearest available sounds as they perceives.
Short vowels are pronounced as long vowels e.g. /I/ pronounced as /i:/, e.g. sheep for ship
Long vowels are different with short vowels not in length but also in quality
Substitution of /ŋ/ (Nasal velar) with /n/ (Nasal alveolar) and /g/ (Plosive velar), e.g. sing
The absence of /ŋ/ sound in Persian makes the learners produce two separate sounds
Substitution of /w/ ( Labial-velar approximate) with /v/ ( Labiodental fricative), e.g. work
The absence of /w/ sound in Persian.
Insertion of intrusive vowels e.g. /e/ before consonant clusters and in in the middle of clusters. Specially before clusters beginning with /s/, e.g. eschool instead of school
Consonant clusters do not occur within single syllable in Persian and learners tend to add a short vowel either before or in the middle of various English clusters.
Palatalization of velar stops, i.e. /k/ & /g/ ( both plosive velars), e.g. book and gather
The error originates from producing Persian equivalent sounds which differ from their English equivalent in the