A Comparative Study Of Machine Translation And Human Translation In Legal Text Translation From The Perspective Of Error Analysis Theory

Translation of non-literary works has always been the core of machine translation and human translation.With the continuous development of China's economy and society,and the increasingly frequent cultural exchanges,people's demand for English Chinese translation is increasing,and its importance is also growing.This paper will focus on the case study of Foreign Investment Law,which is one of its representative categories,and compare the quality of different machine translation from the perspective of EA thoery.According to Corder's error analysis theory,this paper presents a comparative study on the three major machine translation systems,Google,Baidu and Youdao translation and the official human translation,and put forward a quality evaluation model for machine translation.By using the method of literature research,quantitative analysis and qualitative analysis,this paper analyzes the three different levels of different machine translation systems in substance,text and discourse.The corresponding solutions and countermeasures are proposed with regards to different errors.The results show that there is still a gap between the quality of machine translation in legal texts and that of human translation.Machine still needs to expand the capacity of corpus,enhance the ability of text processing and analysis,and enhance the ability of context consideration,which cannot replace human translation in a short period of time.