机器翻译与人工翻译相比的局限性分析——以第七届云南省“LSCAT·云译网杯”笔译大赛(英语)决赛译文为例

Analysis of the Limitations of Machine Translation Compared to Human Translation—A Case Study of the Reference Translation of the Final of the 7^th Yunnan Province “LSCAT∙Yunyi Online Translation Contest (English)”

摘要: 在技术推动下，机器翻译与人工翻译成为主要翻译模式，但机器翻译尚不成熟。本文以第七届云南省“LSCAT·云译网杯”笔译大赛(英语)决赛参考译文为例，剖析机器翻译局限性。在文化差异方面，机器难以传达如中文时间词汇蕴含的独特文化意涵；语境理解上，处理隐喻等表达易生歧义；同义词处理常因未结合语境而致不准确；语法和结构转换时，复杂句子易出错；专业术语翻译在准确度与专业性上有欠缺；语言表达习惯差异处理也不佳。对比决赛译文，机器翻译多“硬译”，人工翻译则在措辞、意译、情感传达及技巧运用上更优。因此，机器翻译处于发展阶段，在重要场合和专业领域仍难与人工翻译媲美，后者在处理复杂语境、文化内涵和语言美感方面具有不可替代的优势。未来，人机协同翻译将成为翻译的主要模式，以实现人类智能和人工智能的优势互补。本文有助于深化对机器翻译局限认识，推动人机协同翻译发展。

Abstract: As technology progresses, machine and human translation have emerged as primary translation modes. Nevertheless, machine translation remains in its infancy. This paper examines the limitations of machine translation by analyzing the reference translation of the final of the 7th Yunnan Province “LSCAT· Yunyi Online Translation Contest (English)”. Regarding cultural nuances, machine translation struggles to convey the unique cultural connotations embedded in Chinese temporal vocabulary. In terms of contextual comprehension, machine translation often misinterprets metaphors and other expressions, leading to ambiguity. When dealing with synonyms, machine translation frequently lacks accuracy due to a failure to consider context. In terms of grammatical and structural transformations, complex sentences are prone to errors. Regarding professional terminology translation, machine translation falls short in terms of accuracy and professionalism. Additionally, machine translation is inadequate in addressing differences in language expression habits. When compared to the final round translations, machine translations tend to be “literal” and “overly direct,” whereas human translations excel in diction, paraphrasing, emotional expression, and skillful application. Consequently, machine translation is still in its developing stage and cannot compete with human translation in important occasions and professional fields. Human translation possesses irreplaceable advantages in handling complex contexts, cultural connotations, and language aesthetics. In the future, human-machine collaborative translation will become the primary mode of translation, leveraging the complementary strengths of human intelligence and artificial intelligence. This paper aims to deepen understanding of the limitations of machine translation and promote the development of human-machine collaborative translation.