学习者翻译语料库与汉英文本翻译测试

Learners’ Translation Corpus and Testing of Chinese-to-English Textual Translation

本文以《中国高校外语专业多语种语料库建设与研究——英语语料库子项目》(CEM)第1期中的八级考试汉英翻译语料为基础,通过综合分析和判别分析,初步展示学习者翻译语料库在汉英文本翻译测试上的应用。本文认为,学习者翻译语料库可以改进汉英文本翻译测试的命题和优化评分量表的制定,使汉英翻译试题的内容更具代表性和区分度,使评分量表更加客观、可靠,更具概括性。

Learners’ corpus of translation into a second language is yet a research niche in the undertaking of corpus linguistics. This study taps into a particular learners’ translation corpora—CEM-T—and explores its possible applications in the Chinese-to-English textual translation testing for undergraduates majoring in English in China. With an integration of global and discriminate approaches to a body of English major students’ Chinese-to-English translation scripts from TEM8, the test-takers’ general patterns of translation errors and the discriminating features of different score levels have been highlighted. The results add support to the claim that corpus of such nature could pave the way for a translation test practice of enhanced quality. Specifically, they can lend to the streamlining of test design through a better-selected L1 text and optimization of the related rating scale and criteria, making sure the test-taker’s scripts are more objectively-judged. The direction for further research in this area has also been mentioned.