Fossilization In Oral English Learning Among Chinese English Majors

For decades, the charm of fossilization has never faded for SLA researchers. Starting from Selinker's interlanguage (1972), numerous scholarly theoretical opinions and empirical studies have conglomerated in such a field. Interestingly, from the author's former teaching practice, subjects ranging from junior school students, to university students, and to the middle aged, multitudinous facts about repeated errors are indeed discovered from them. This tallies with the phenomenon of fossilization. This is thus associated with the spoken language of English majors. Then some hypotheses about English majors'interlanguage surface in my brains for finding answers.Florence Myles (2005:374) suggests SLA research crucially depends on good datasets to work from. Researchers aim to build models of the underlying mental representations and developmental processes which shape and constrain L2 productions. He further points out two variables to guarantee the degree of generalization of data analysis, a lengthy period of time and very large cross-sectional datasets. Corpus, in no doubt, becomes a rich data resource which covers a considerable amount of time and a large enough quantity of data. As commented by Susan Hunston (2002: 23), one function of corpus is to offer evidence, from which learners'state of learning on levels of phonology, morphology, lexis, syntax and pragmatics can be identified, the mental processing of L2 production can be well perceived and the degree of differential success/ failure can be further diagnosed. This perhaps explains the reason why the author sets her mind on learner corpora SECCL to collect evidence from the spoken data, which can make the research analysis confident and convincing.This thesis sketches an outline of real problems that exist in the teaching cases and states the hypotheses to test as well as the purpose the study is designed to realize. Different understandings of fossilization in terms of definition and classification are given, followed by a brief review of representative fossilization viewpoints from abroad and at home. Multi-dimensional possible causes that lead learners to stay stagnant are analyzed, from learner's, teacher's and environment's perspectives respectively. Learner's factors, the internal causes, might be regarded as more influential than the other two types of external causes. In the part of empirical study, SECCL, the untouched data source is initially presented; then the theoretical support of using the corpora and the plan of sampling, verb be as point of departure as well as error tagging procedure are stated. The processed data has been fully compared and discussed, showing the gaps and similarities between the selected TEM4 and TEM8 corpora. Different graphs and tables demonstrate the degree of fossilization of each compared linguistic item. Particular examples are drawn from the selected corpora both (except that certain error cases appear in one selected corpus only, while unfound from the other). Implication for oral English acquisition from the perspective of fossilization, limitations of the study are also included for further improvement and exploration.