作品原文
茅盾 《直译·顺译·歪译》
“直译”这名词，在“五四”以后方成为权威。这是反抗林琴南氏的“歪译”而起的。我们说林译是“歪译”，可丝毫没有糟蹋他的意思；我们是觉得“意译”名词用在林译身上并不妥当，所以称它为“歪译”。
林氏是不懂“蟹行文字”的，所有他的译本都是别人口译而林氏笔述。我们不很明白当时他们合作的情形是别人口译了一句，林氏随即也笔述了一句呢，还是别人先口译了一段或一节，然后林氏笔述下来？但无论如何，这种译法是免不了两重的歪曲的：口译者把原文译为口语，光景不免有多少歪曲，再由林氏将口语译为文言，那就是第二次歪曲了。
这种歪曲，可以说是从“翻译的方法”上来的。
何况林氏“卫道”之心甚热，“孔孟心传”烂熟，他往往要“用夏变夷”，称司各特的笔法有类于太史公，……于是不免又多了一层歪曲。这一层歪曲，当然口译者不能负责，直接是从林氏的思想上来的。
所以我们觉得称林译为“歪译”，比较切贴。自然也不是说林译部部皆歪，林译也有不但不很歪，而且很有风趣——甚至与原文的风趣有几分近似的，例如《附掌录》中间几篇。这一点，我们既佩服而又惊奇。
现在话再回到“直译”。
照上文说来，“五四”以后的“直译”主张就是反对歪曲了原文。原文是一个什么面目，就要还它一个什么面目。连面目都要依它本来，那么，“看得懂”，当然是个不言而喻的必要条件了。译得“看不懂”，不用说，一定失却了原文的面目，那就不是“直译”。这种“看不懂”的责任应该完全由译者负担，我们不能因此怪到“直译”这个原则。
这原是很浅显的一个道理，然而不久以前还有人因为“看不懂”而非难到“直译”这个原则，而主张“顺译”，这也就怪了。
主张“顺译”者意若曰：直译往往使人难懂，甚至看不懂，为了要对原文忠实而至使人看不懂，岂不是虽译等于不译；故此主张“与其忠实而使人看不懂，毋宁不很忠实而看得懂”。于是乃作为“顺译”之说。“顺”者，务求其看得懂也。
在这里，我们觉得不必噜噜苏苏来驳斥“顺译”说之理论上的矛盾（因为它的矛盾是显然的），我们只想为“直译”说再进一解：
我们以为所谓“直译”也者，倒并非一定是“字对字”，一个不多，一个也不少。因为中西文字组织的不同，这种样“字对字”一个不多一个也不少的翻译，在实际上是不可能的。从前张崧年先生译过一篇罗素的论文。张先生的译法真是“道地到廿四分”的直译，每个前置词，他都译了过来，然而他这篇译文是没人看得懂的。当时张先生很坚持他的译法。他自己也知道他的译文别人看不懂，可是他对《新青年》的编者说：“这是一种试验。大家看惯了后，也就懂得了！”当时《新青年》的编者陈促甫先生也不赞成张先生此种“试验”，老实不客气给他改，改了，张先生还是非常不高兴。现在张先生大概已经抛弃了他的试验了罢，我可不十分明白，但是从这个故事就证明了“直译”的原则并不在“字对字”一个也不多，一个也不少。“直译”的意义就是“不要歪曲了原作的面目”。倘使能够办到“字对字”，不多也不少，自然是理想的直译，否则，直译的要点不在此而在彼。
作品译文
Appropriation, Distortion and Literal Translation
Mao Dun
The term “literal translation” has become something of authority since the May Fourth Movement in clear opposition to Lin Qinnan’s “distortion in translation”. When referring to Lin’s translation as “free”, we do not intend to disparage him in the least. Nonetheless, the term “free translation” is not exactly applicable to Lin, and as a result, “distortion” is our denomination of his translation.
Lin had no knowledge of any European languages whose letters were no more than “squiggles” to him. All of his translations were done after they had been orally translated by others. We are not quite sure how they collaborated at that time. Was it that someone orally translated one sentence, which was then followed by Lin writing it down? Or was it only after one paragraph or passage was orally translated that Lin transcribed the oral version? In any event, this practice could barely avert a double distortion: in the process of the original text being orally translated into a colloquial form, some degree of distortion was unavoidable. Later when translating the spoken words into classical Chinese, Lin distorted the original text yet again.
Such distortion may be traced onto his “translation methodology”. Lin was a passionate champion of orthodox Confucian teachings, and “the doctrine of Confucius and Mencius” had been spiritually passed down to him. He would identify Scott’s style with that of the great Chinese historian Sima Qian… The corollary could only be yet another distortion. His collaborators, however, could not be blamed for the last type of distortion, which derived directly from Lin’s ideology.
Therefore, we opine that it is fairly befitting to call Lin’s translation “distortion”, by which we do not mean that every translation by Lin is a product of distortion. Some of his translations are not distorted too badly at all, and they certainly lack no sense of humour, which shows some resemblance to what is inherent in the original text, as is evidenced in some of the pieces in The Sketch Book, 1820. This has filled us with wonder and admiration.
Let us come back to the term “literal translation”. As stated earlier, insistence on “literal translation” after the May Fourth Movement was in opposition to distortions of the original text. The features of the original text should be kept in a translation. If even all the features are supposed to be retained, “intelligibility” is, without a doubt, a prerequisite for translation. But if a translation turns out to be “unintelligible”, needless to say, it must have lost the features of the original text, which then cannot be called “literal translation”. It is the translator who should be held responsible for the “unintelligibility” of his translation. We must not, therefore, blame the principle of “literal translation”.
This argument is pretty easy to follow. But it is odd that not long ago, some people maintained appropriation in translation on the grounds that, due to “unintelligibility”, the principle of “literal translation” would have to be abandoned. Those espousing appropriation in translation believe that obscurity and even “unintelligibility” are attributable to literal translation. If faithfulness to the original text leads to “unintelligibility”, translation is tantamount to non-translation. Thus, they argue: “‘Unintelligibility’ resulting from faithfulness is worse than faithlessness that makes translation ‘intelligible’.” For the sake of “intelligibility”, the theory of appropriation is formulated to “appropriate” a translation in order to make it “intelligible”.
At this point, we consider it utterly unnecessary to direct a prolix and verbose refutation at the contradictions in the theory of appropriation (for they are apparent contradictions). We merely wish to present a further exposition of “literary translation”. It is our view that the so-called “literal translation” is not necessarily “word-to-word” translation, or equivalence with neither a word more, nor a word less. Due to the structural differences between Chinese and Western languages, such narrowly strict equivalence is practically impossible.
Mr Zhang Songnian once translated an essay by Bertrand Russell. Zhang’s translation is an “one hundred percent” literal translation, omitting not even a single preposition. However, no one could understand this translation. But Mr Zang adamantly refused to give up his way of translation. Fully aware that his translation would be unintelligible, he made this remark to the editor of New Youth: “It was an experiment. When people get used to it, they will understand it!” The editor of New Youth at that time was Mr Chen Zhongfu, who disapproved Zhang’s “experiment”, and made untrammeled revisions to the translation in question. Zhang was understandably very upset with these revisions.
Maybe Mr Zhang has renounced his experimentation now, though I am not so sure. But what this anecdote proves is that the principle of “literal translation” does not entail “word-for-word” translation, not a word more, and not a word less. The underlying meaning of “literal translation” is simply “do not distort the features of the original text”. If it were possible to attain “word-for-word” translation, with nothing being added or lost, it would naturally be an ideal translation; however, this is not the main point for literal translation.