A Corpus-based Study Of Simplification In English-chinese Translated Economic News

Corpus-based research method is descriptive and empirical. It is based oncorpora which are built through collecting real-life data. In the facilitation ofcomputer soft wares and under specific goals, it analyzes the data, describes themobjectively with statistics and trying to generate rules in these corpora, if any. In thisthesis, a bilingual parallel corpus has been built consisting of two sub-corpora: Corpusof Original English Economic News (COEEN) and Corpus of Translated ChineseEconomic News (CTCEN). Each includes about50thousand words. Besides, areference corpus with about50thousand words has been compiled: Corpus ofOriginal Chinese Economic News (COCEN).Based on the parallel corpus and by making use of tools such as AntConc,WordSmith Tools4.0and BFSU PowerConc1.0, the present research makes acomparative and descriptive investigation on simplification as one of the “TranslationUniversals” through observing translated Chinese economic news texts. By testing thefour indicators of simplification, which are STTR for lexical variety, lexical densityfor information load, list heads and mean sentence length, the results turn out to bepartly contradictory to what the simplification hypothesis assumes. It finds: first,translated Chinese economic texts display a broader vocabulary range than both theirsource texts and original Chinese economic news texts. Second, the lexical density(information load) of translated Chinese economic texts is higher than their sourcetexts but lower than original Chinese economic news texts. This result provides someevidence to support my second research hypothesis while some goes against it. Third,sentences in CTCEN are generally much shorter than texts in COEEN and slightlyshorter than those in COCEN. It partly supports but partly opposes my thirdhypothesis.In short, the findings of the present thesis partly but not fully advocate theassumptions of simplification.The conclusion of this thesis is consistent with that of Wang Kefei&Qin Hongwu (2009). There are also differences between my investigation and Wang&Qin’s. First, the language data in this thesis are confined to economic news texts,while Wang&Qin’s investigation involves more comprehensive language datacovering several fields such as literature, fiction and science. In addition, the corpusemployed in this thesis is much smaller in size than Wang&Qin’s. This may be oneof the limitations of my investigation.