Significantly， it is this last aspect of translation to which mechanical and computerized techniques are being applied with some prospects of limited success． Machine translation，whereby ultima- tely，a text in one language could be fed into a machine to produce an accurate translation in another language without further human intervention，has been largely concentrated on the language of science and technology，with its restricted vocabulary and overall likeness of style，for both linguistic and economic reasons．Attempts at machine translation of literature have been made，but success in this field，more especially in the translation of poetry，seems very remote at present.
Translation on the whole is an art，not a science.Guidance can be given and general principles can be taught，but after that it must be left to the individual"s own feeling for the two languages concerned.Almost inevitably，in a translation of a work of literature something of the author"s original intent must be lost；in those cases in which the translation is said to be a better work than the original，an opinionsometimes expressed about the English writer Edward Fitzgerald"s “translation”of The Rubaiydt of Omar Khayydm，one is dealing with a new，though derived work，not just a translation.The Italian epigram remains justified： Traduttore traditore “The translator is a traitor.”
Excerpts from the Entry “Translation” in New Encyclopaedia Britannica，Macropaedia.
3
Most successful and creative translators have little or no use for theories of translation.In fact，some insist that only those who cannot translate become theorists of translating.In reality，outstanding translators are born，not made，since without an innate potential for the creative use of language，the study of procedures and principles of translating is unlikely to produce outstanding results．According to G.V.Chernov of the Maurice Thorez Institute in Moscow，this institute has not produced a top-flight translator during the last 25 years．Those who have the ability for outstanding creative work（and there are many such Russian translators） simply have not felt the need for such instruction.The institute has，however，been quite successful in training translators for less significant levels of competence.
We should not attempt to make a science out of translating，since it is essentially not an isolatable discipline，but a creative technology，a way of doing something which employs insights from a number of different disciplines.Translating can never be any more holistic or comprehensive than the disciplines on which it depends．Furthermore，translating，like language，must be open-ended，since it must deal with different kinds of texts，designed for quite different audiences with very different presuppositions about what a translation should be.There is no way in which these multiplicities of use，so subject to change without notice，can be neatly categorized and quantified.No methodology can stipulate all the steps in procedure which should be followed in order to arrive at proper solutions.But decisions have to be made，and great translators instinctively grasp the solutions in unpredictable ways and in a manner often unrecognized by the translators.
In a sense translating is both discouraging and challenging：discouraging because there are no simple rules to follow and no way to know in advance if a solution is completely correct and acceptable，but also challenging because it is excitingly creative.The very tensions within translating bring out the best in those who instinctively learn to play the game．
Excerpts from Translation：Possible and Impossible by Eugene A.Nida，1991．In Translation Beyond the Boundaries of Translation Spectrum，Translation Perspecties IX 1996，Edited by M.G.Rose，Center for Research in Translation，State University of New York.
翻译的性质（综合编译稿）
[1]对于“翻译”，现有多种多样的定义，其中最重要的分歧在于翻译是科学，还是艺术，由此分成两派：科学派和艺术派。
[2]科学派着重研究对翻译过程以及语言形式和结构的描写，以揭示翻译过程固有的客观规律。而艺术派则强调翻译的效果，林语堂曾是代表人物之一。他在《论翻译》一文中说：“翻译是一种艺术。凡艺术的成功，必赖个人相当之艺术，及其对于该艺术相当之训练。此外别无成功捷径可言，因为艺术素来是没有成功捷径的。”
[3]《不列颠百科全书》（详编）解释“翻译”时认为：“翻译总的说来是一种艺术，不是一门科学。翻译可以加以指导，讲授一般的原理。但除此之外，译品的优劣要看译者本人对两种语言的感知了。”同时，该书编者也注意到机器翻译的事实：也许最终能够把一种语言的文本输入机器，然后产生另一种语言的确切译文，而不需要人的进一步润湿修改，不过这种方法现在主要集中应用于科技语言的翻译。该书也写到文学作品的机器翻译，认为“在这个领域里，特别在诗歌翻译方面，似乎目前离成功还很远。”
[4]综合两派的长处和短处，刘重德先生就文学翻译提出这样的观点：现在的趋势是把它们合而为一，因为文学翻译有双重性-- 有其自身规律与方法的科学的一面，又有艺术的一面。
[5]现在，译界普遍注意到美国着名翻译家和翻译理论家尤金·奈达对这个问题的观点有一百八十度的转变。奈达曾在1964年出版过专着《翻译科学探索》（Towards a Science of Translation），竭力推行翻译的科学论。但是时隔二十余年，奈达开始否定他原先的看法。在1991年发表的《翻译的可能与不可能》一文中，他写道：“我们不应该试图把翻译变成一门学科，因为它本质上不是一门独立的学问，而是一种有创造性的技艺，一种靠综合多学科的见识来处事的本领。”他还认为，跟语言一样，翻译是无止无境的，译者面对为不同读者而作的用途各异、变化多端的原文，不可能对此作明确分类和定量分析。在操作过程中也不能一步步地规定方法来解决翻译中的诸多问题。大多数成就卓着、富于创造精神的翻译家很少用得着，或者根本用不着翻译理论。
[6]总之，对于翻译是科学，还是艺术，抑或是两者的综合，至今仍是公说公有理，婆说婆有理。
注：本篇综译稿由三篇原文综合编译而成，各篇之末均已注明出处，来源文献不另列。