In this chapter, the relevant literature will be illustrated and critiqued, with particular attention to the works on phonological systems and writing systems across languages and the interaction between the two systems. In addition, the influence of affective factors on the productive skills of Thai learners is reviewed. The first section of the chapter is a discussion of the relevant literature on the differences between phonological and writing systems across languages and the resulting language learner output, followed by a review of the framework to be used in this study.
The final section is a review of the effects of the affective factors on the learners’ language achievement and language performance as well as on language learning. 1. The related literature The findings reported in the literature that the majority of Thai learners of English demonstrate a low degree of proficiency, especially in the productive skills (speaking and writing) reflect the fact that “English language pedagogy in Thailand … is still in its infancy” (Wongsothorn, A. , Hiranburana, K. & Chinnawongs, S. 2002; Laopongharn & Sercombe, 2009, among others). As reported in the national survey (1999) conducted by the Office of Educational Testing of the Department of Curriculum and Instruction (in Khamkhien, 2010), “high school sstudents’ productive skills were generally below 50 per cent, i. e. below average, leading to the recommendation in the report for immediate improvements of writing ability in all educational institutions in the country” (Wongsothorn et al. , 2002: 112). There are several factors that could prevent Thai learners of English from aining a sthrong command of productive skills in English. In terms of the phonological system, Yangklang (2006) investigated the improvement in pronunciation of English final [l] in 40 Thai sstudents in Matthayom Suksa 4 at Assumption Convent Lamnarai School who had used the Computer-Assisted Instruction (CAI) programme she was testing. The CAI in the study, as described by Yangklang: “contained drill and practice exercises. Drill and practice activities aimed to provide learners with adequate practices and also to review items that were new to learners.
The goal of the drill and practice activities was to teach sstudents to pronounce words with final /-l/ accurately and automatically. ” These practices and exercises were, therefore, provided by the computer programme as part of the post-lesson activities. The participants were divided into two groups according to their competence of pronunciation of English, one with good pronunciation and the other with poor ability. Prior to the experimental task, each participant had taken a placement test to classify their pronunciation competence.
Subsequently, they all took three pronunciation post-tests every week for three weeks. It was found that both groups of sstudents improved their pronunciation significantly after they used the CAI programme. In general, both groups had positive reactions to the use of the CAI programme for improving their pronunciation. Obviously, the programme helped the participants improve their pronunciation, given that they had intensive training on pronunciation practice of specific English consonant sounds, and the post-test of pronouncing the sound was administered immediately after the programme training.
As Graham (1997) and Macaro (2006), among others, have pointed out, effective language learning is about mastering communicative skills, i. e. speaking, writing, listening, and reading. This indicates that the more learners practise, the better their communicative performance in a language. In line with this, the results did not show anything unexpected. The point at issue, which I shall leave for future study, is how can accurate pronunciation, e. g. of the /l/ sound, be maintained after short-term, intensive training?
In Yangklang’s findings, the English consonant sound /l/ which constitutes the coda, i. e. word-final ending in /l/, (see Figure 1 below) could be realised as [l], [n], and [w] by the participants. The participants with good pronunciation appeared to produce the [l], whilst the participants from the poorer group appeared to generate [n] and [w] instead of [l] or [? ]. The [n] was produced more frequently than [w] by these participants, however. Yangklang did not discuss why [n] and [w] constitute allophones of /l/.
In the study, it was the poorer group who produced such allophones, and it was reported from interviews that the participants had not yet been exposed to an environment where English is used as a medium of communication. In the word list, part of the research instrument which consisted of pronunciation tests contained words like ‘ball’, ‘mile’, etc. which are English loanwords. I assume that there is some force of segmental alternation with regard to English loanwords.
As /l/ is not available at coda position, the notion of final consonant phoneme alternation appears plausible. That is to say, [n] and [w] appear to be alternatives. Paradis (1996) in Kentowicz & Suchato (2006), claimed that the location of such segments in feature geometry and prosodic structure was relevant in dealing with segmental alternation in loanwords. Based on the data, /l/ and /n/ fall into the same natural class, in that both of them are sonorant consonants, which can be represented by the feature matrices