On the aquisition of grammar and meaning in instructed second language learning: a case study of the development of past verb forms by adult French learners of English as a foreign language.

摘要

A corpus of written English produced by three groups of adult Frenchudbeginning learners of English as a foreign language over a period ofudapproximately eight months was examined for evidence relating to theudacquisition of past tense forms and related meanings.udThe findings provide evidence to support several hypotheses which canudbe usefully grouped within a single framework which sees languageudacquisition as a process of hypothesis formation and testing whoseudconstraints are both first and second language in origin. Theseudhypotheses can be summarised as follows:ud(1) Language learning involves the acquisition of a new system ofudexpressing meaning. As a result, the learner engages in a process ofudmatching linguistic form to underlying meaning both within andudbetween languages. Only such a hypothesis, we believe, can satisfactorilyudexplain the apparently random variation that was observed inudour subjects' acquisition of past tense.ud(2) Language transfer is thus necessarily a widespread phenomenon,udconstraining learners' formation of hypotheses, but is itself constrainedudby the inter- and intra-language form-meaning transparencyudof the language item in question. In other words, whenever formmeaningudrelationships are not wholly transparent, transfer is to beudexpected.ud(3) Moreover, even when form-meaning relations are transparent,udtransfer may take place due to the learner's shortage of processingudcapacity. When this is lacking, learners tend to maintain communicationudby relying on existing procedural knowledge, which, at least inudthe early stages, means well-established first language procedures.udThis is because, even though humans can process at phenomenal speeds,udthis is only possible with procedures which are solidly in place. Forudthe vast majority of language learners this implies that firstudlanguage procedures will always take precedence over weaker secondudlanguage procedures because they were later traced and are lessudfrequently used.udUnder the circumstances, where demands are made on the learner, forudexample, to produce language in real time, there will be a tendencyudto bypass second language networks and rely on first languageudcircuitry. Consequently, learners make use of their ability toudoperate a number of strategies (such as planning and monitoring theirudlanguage output) to produce comprehensible language. This ability,udlabelled strategic competence, is identified as a key aspect inudlanguage use in general.