An Exploratory Study of the Non-Conventional Use of English Structure in the Writing of Thai Academics

Abstract

This study investigated the factors which have influenced the level of English of a sample of 26 Thai academics from various fields, who publish work in English. In particular the study investigated the interplay of their linguistic and cultural background, their characteristic traits and adaptive behaviors and their experiences of learning and using English. As proxies for level of English, the study analyzed non-conventional English use (NCU) in extended samples of both their writing and speech in English and also assessed their ability to recognize problematic areas of language use in their writing. The writing samples analyzed were 26 pre-publication manuscripts written by the authors, who were also interviewed regarding their experiences of learning and using English, following which a similar analysis of NCU in a sample of their speech was conducted. They were also asked to identify the main language areas which they had difficulty in using and their ordering of those areas was compared with the actual order derived from the analysis of NCUs in their written work, as a measure of their meta-linguistic awareness. The information relating to their English learning and use experiences derived from the interviews was analyzed quantitatively and those variables which showed the highest correlations with the written and spoken NCU data and the meta-linguistic awareness coefficients were used as independent variables in multiple regression analyses, with the written and spoken NCU data and the meta-linguistic awareness coefficients as dependent variables. Significant models were derived in all three cases accounting for up to 88% of the variance in the dependent variables. The analysis of the manuscripts found high correlations between authors in the areas which produced the largest numbers of NCUs, indicating that the main areas giving rise to NCU were common to all the authors. The main areas were articles, prepositions, verbs and nouns which together accounted for around 80% of the total structure NCUs. It was noted that these areas correspond to areas identified as problematic in previous studies of learners at earlier stages of their education and are all areas where Thai and English differ markedly in their structure. These findings suggest that the factors giving rise to the problems are widely shared by Thai learners of English at all levels and are therefore likely to be due to factors which are common to all or most Thais. Moreover, the consistency of the pattern of NCU and its agreement with earlier studies of learners at earlier stages of development, suggests that these problems are persistent over time and that while their effect may be ameliorated by learning and use experience, they are relatively immune to being entirely obviated. The study concludes that the main factor causing the participants' NCU of English is the effect of structural differences between the Thai and English languages in those areas giving rise to the largest numbers of NCUS. Further, the regression analyses identified 30 learning and use factors as having influenced aspects of the NCU or meta- linguistic awareness variables, and that combinations of those factors accounted for between 71 and 88% of the variance in those variables. This finding suggests that most of the variance in the rate of occurrence of NCUS between the participants was the result of aspects of their learning and use experience, with the balance being attributable to their characteristic traits and adaptive behaviors. This finding is broadly supportive of the appropriateness of complexity theoretic approaches to the study of the learning of second languages, which view language learning as a complex system which develops due to the non-linear interaction of variables over time and is dependent both on initial conditions and on learning resources. The main implication to be drawn from the study's findings is, therefore, that SLA theories proposing that individual factors substantially influence second language learning may misjudge the complexity of the process and the importance of the situational and linguistic contexts, and that individual learning is more likely to result from a combination of many factors with none being dominant. Other implications noted relate to how the teaching of English in Thailand and in particular to the teaching of academic writing should be conducted. The study thus contributes to the fields of SLA, error analysis and pedagogy in Thailand by focusing on a relatively homogenous sample of Thai academics, whose socio- cultural and educational backgrounds were similar and who shared a similar motivation to learn English, based on their need to use it as a means of participation in their academic communities. It is apparently the first to analyze pre-publication academic journal articles for evidence of language learning difficulties and to compare the numbers of errors with numbers of particular word types in texts as a means of assessing the degree of problematicity of errors. Moreover as no previous study identified has yet done, it offers empirical evidence of the appropriateness of a complexity theoretic approach to second language learning by identifying the wide variety of personal, situational, linguistic and experiential factors influencing the English structural accuracy of this group of participants