Is Technical Translation Really a Collaborative Activity?

By Steve Vitek, technical translator

This article was published in the May 2003 edition of the ATA Chronicle, 
the magazine of the American Translators Association. ( 

“Whereas other translation services send their materials to professional translators who work in many fields, XYZ sends its translation projects to bilingual category experts - people who trained and worked in the field. Translations are reviewed by a second bilingual category expert for technical accuracy, and finally the translations are brought in-house where an editor reviews the document for readability. If graphics or formatting are required, that is also accomplished by a bilingual professional. All our translators and editors have advanced degrees (such as, MD’s, Ph.d’s, MHA’s, RN’s) in their fields, and virtually all are natives of the target language country....” 
“There are five independent quality control points in our translation process, so we always certain that the documents being read replicate the intent and style of the original.” 

   The statements above were on the website of a translation agency under the heading: How We’re Different (including the “so we always certain part”).
   There are several schools of thoughts on the topic of this article. Some translators enjoy having their masterpieces dissected, criticized, and/or praised by editors and/or colleagues. Beginners especially fall into this category, and for a very good reason. But most translators, including this one, dislike few things more than when other people try to change anything in their translation, other than fixing a typo or an omission. This article will attempt to explain the reasons why technical translation, and in particular the translation of patents from foreign languages such as Japanese, German, or French for litigation and filing purposes, is probably not a suitable candidate for the collaborative approach described by the agency quoted above, as opposed to perhaps other translation fields, such as the translation of software or operating manuals.
   It always makes me laugh when I read statements like the ones above that often appear on the websites of translation agencies. If you run a search, you can easily find dozens of agency websites that are all “different” in the same way - they don’t trust a hapless translator to translate anything accurately until his or her translation has been checked by another translator, then rechecked by yet another translator, then proofread by a proofreader who has at least a PhD., and then checked one more time “for readability” by yet another editor who is possibly God Herself. The assumption here seems to be that since more heads know more, the company will use as many heads as it takes, all of them incredibly wise heads, to attain The Perfect Translation. Why, even the graphics are input by a bilingual expert (this is optimal for superior scanning resolution). The XYZ agency above (having been brought up in the humanist tradition in the old country, I am not using their real name) seems to display a healthy amount of contempt for translators, especially given the fact that they are in the translation business. They don’t use “translators who work in many fields” at all. They prefer “bilingual category experts,” whatever that means, who work in the field, presumably a field that these “bilingual category experts” know thoroughly. One can probably assume that since these experts are not translators, they did not study their language for many years at a university, but simply “picked up the language” somehow instead of majoring in it. It is not really important how they “picked up the language” - a beautiful expression common in this country. In most other countries, people do in fact study languages rather than simply picking them up like a piece of unwanted garbage lying on the street. As long as they are not really translators, agency XYZ will trust these “bilingual experts” to know their languages. (When a doctor asks you next time, “How did you pick up your German, Ms. Interpreter” [misinterpreter], answer by saying: “How did you pick up your medicine, Doc?” Will he get it?) 
   So now we know that as far as the general public and some translation agencies are concerned, you don’t really have to be a translator who has been studying languages all your life, a translator with a degree in translation who has been putting his or her brain through mind-bending exercises in more languages and more highly technical subjects than just one for the last few decades. As long as you are a “bilingual category expert,” you’ll do.

Where Have All Those Bilingual Category Experts Been Hiding All This Time?

   The thing is, most of the time, there are no bilingual category experts around when you need them. That is, when you happen to be a hapless monolingual category expert, such as a patent lawyer or doctor who has to figure out what the heck is in that Japanese document that was just delivered by FedEx. Had we had a few more of these experts where they should have been in September 2001, two very tall buildings would still be standing in Manhattan and a lot of innocent people who lost their lives would still be alive today. 
   A bilingual category expert in my field would be a patent lawyer fluent in at least Japanese, German, and English, with a degree in chemistry and physics, etc. If he really knows his stuff and at the same time knows Japanese or German, he can be hired on the spot by any number of major patent law firms who deal with hundreds of patents and other documents in foreign languages, especially Japanese and German. The problem is, a good patent lawyer or MD, monolingual or trilingual, will make at least twice what a good technical translator can make, which is probably a good explanation why not too many patent lawyers or doctors are clamoring to become translators. There may be some exceptions, but they are extremely rare. Thus, the only expert who really knows his stuff in several languages will usually be an experienced patent translator who has translated thousands of patents from a complicated language such as Japanese or German in a number of fields. If you find a good one, there is no need for five layers of additional checking and proofreading, because the highly technical material will be translated accurately the first time around. On the other hand, 5 layers of checking and proofreading, or 25 layers, are unlikely to result in an improved translation if the original translation was not accurate, even if a company could afford such a wasteful use of the most important and expensive asset that any business has - human intellect. 

Too Many Cooks Will Spoil The Broth

   To try to change terms in a highly technical translation is about as wise as to pull out bricks from the construction of a house in order to replace them with bricks that look better to you. The problem is, the original construction will collapse even if you pull out just a couple of bricks. Every translator makes hundreds of split-second decisions during his or her work. After the translation is finished, the translator must make dozens of additional instantaneous decision during proofreading. Whether these decisions are right or wrong will depend in part on how well the translator knows and understands the subject area and the terms, but that is only one segment of the total package. The most important factor, in my opinion, is whether the person making these decisions is a good translator - a person who has the required education, skills, and experience, and who is also a very good writer. Most bilingual experts are not translator material if they lack education emphasizing linguistic skills and translating experience. I once had to retranslate a biomedical patent that was “already translated” by a Japanese doctor who knew all the terms perfectly well and whose command of English was incredibly good for a Japanese person. This doctor would have made a superb translator, provided that money was not important to him. But for some reason, his translation was really a summary rather than a translation. Some parts were left out and other parts of the patent were rewritten to some extent, so that the resulting text was not really a translation but an edited version that was not accurate enough to be used for filing a patent in English in this country. I am not sure why he translated the patent the way he did, but I had to retranslate the whole thing to provide a real translation so that the patent could be modified and filed as a U.S. patent.
   On another occasion, I once received a short medical patent in a language I don’t really know well enough to translate from. This was only the second time I received a medical patent in this language, and the translator I had worked with previously was not available. A translation agency recommended a bilingual doctor to me, so I sent her the patent, even though she charged more than what most freelance translators are charging. She did a terrible job. It took me several hours to correct the mistakes and omissions in the translation, and I was only able to do that because I understood the original well enough based on the poor English translation of that doctor and because of my somewhat limited ability to follow the text in the original language. A bilingual expert is not necessarily a good translator, and a good translator is much more than a bilingual expert. From now on, if I have a choice between a doctor and a professional translator, I will always choose the latter.

When I See Other People’s Translations, My First Reaction Is Usually Negative

   Although I try to translate as much of the patents that I receive from law firms myself, sometime I send work to other people if I don’t have enough time to finish everything myself. When I see other people’s translations, my first reaction is usually negative. But I realize that the main reason why I don’t like translations done by other translators is mostly that their style may be different from mine, but not necessarily wrong. Fortunately, I almost always work with people who are experienced translators that I have known for many years, and once I get used to their style, I hardly ever change anything in their work. Sometime I fix an occasional typo or an omission, but I am consciously not trying to change much more beyond that because I respect the interpretation of the original of this particular translator. Who am I to say which nuances in the interpretation of the text are better and make more sense? Five layers of additional proofreading and editing would most likely destroy a good translation, and they would definitely not save a bad one, because nothing can. We know good translation when we see it. Why should we try to change a good thing? Except, perhaps, if we get paid for doing so by a dumb boss who thinks that you can keep changing a good thing several times until you get an even better thing.

Those Who Can Translate Usually Translate, And Those Who Can’t .... Edit

   Which is not to say that proofreading is not useful and in fact indispensable, even proofreading a translation that is excellent. Everybody makes mistakes. But a concept that translation should be approached like a public meeting at the City Hall, where we all put our heads together and arrive at the perfect solution in our collective wisdom, is fatally flawed. 
   This concept is, in my humble opinion, nonsense, even if it were economically feasible to have five qualified expert translators check a translation several times. Based on my experience over the last two decades here and in Japan, the reality is that coordinators and proofreaders who work for translation agencies are rarely expert translators. The reason is similar to the reason why good patent lawyers and doctors usually don’t become translators. While technical translators usually make less than lawyers and doctors, a good translator of Japanese or German patents can easily make more than $100 an hour when translating. Why would such a person want to make a fraction of that amount by checking other people’s work, which is not nearly as interesting as translating? So what really happens is that those who can translate, translate, and those who cannot, edit. It is not unlike the dichotomy that has been observed to exist in the writing profession. People who are good writers write books, people who like writing but maybe are not that good at it, teach creative writing at college. Even if the editor is bilingual, which does not happen very often in my language combinations, he or she will usually not be as good a translator as the one who actually translated the thing from the scratch. On the other hand, even a monolingual editor who knows a technical subject very well can catch potential mistakes in a translation, and when you deal with “exotic” languages such as Japanese or Czech, your editor is likely to be monolingual.
   I am not against proofreading and proofreaders. (Some of my best friends are proofreaders…. Well, not really, but close enough). But seriously, several layers of collaborative proofreading make sense in some cases. But there is a right way to do it, and then there is the way that is proudly featured in the website propaganda of a great number of agencies, which involves checking by layers upon layers of supposed experts who are not really translators. I sometime get work from a small agency that organizes proofreading in an intelligent way. Because they always proofread my translations very carefully and nobody there knows Japanese or Czech, they sometimes call me with editing questions. One of the persons working there has a degree in biology and has actually worked in the field, so she is usually the one who proofreads my translation of biomedical patents. But if I ask them who is proofreading a huge biomedical monster patent, 150 pages of DNA slicing and dicing, the answer is usually: “Oh, well, we all read it. We catch more mistakes that way.” This is absolutely true, provided that the proofreaders realize that their job is to look for typos, inconsistencies, and omissions, and not to try to “improve” my translation. And these people do realize this, which is why I still work for them.
   According to the European Patent Office, about one million patents are issued every year on this planet, and about one-third of them are patents in Japanese. Japanese scientists, inventors, and patent lawyers have one great advantage over their counterparts in this country: they can usually read patents in English, while their counterparts almost never read Japanese. Although every Japanese patent application is provided with an English summary, these summaries are sometime very short, usually about 50 words or so, even if the patent is more than ten thousand words long. U.S. patent lawyers, or in fact their clients, thus have to pay a lot of money for translations, but first they have to be able to identify which patents need to be translated. This is no easy task when you deal with a language as complicated as Japanese. 

Once We Accept Your Translation, We Are “Stuck” With It

   It is thus quite important for a patent law firm to have access to a reliable translator who can be relied upon to provide competent expert testimony in court about his translation or other translations done by other translators. The first question to a translator testifying in court about his translation would be: “Is this your translation?” When you are under oath, you better tell the truth. Let us assume that you have been hired by XYZ agency to translate a patent that happens to be crucial to a lawsuit, and the agency had your translation proofread by several proofreaders who have made significant changes instead of just correcting typos or omissions as a sane person would do. Since you are under oath, you would have to say: “Yes, I translated it, but I did not use these exact words. Somebody changed this word and that word at the agency. They check and edit all translation in five stages for readability. It is a company policy.” In other words, your translation has just been rendered worthless as a crucial piece of evidence in a very expensive lawsuit. 
   On the other hand, a professional translator who translates in several fields, and preferably from several languages, and avoids agencies of the XYZ type like a plague, will be able to tell the lawyers why a so-called causative mode construction followed by a mode called “renyokei” in Japanese was translated in a certain way into English. Such a translator will be able to explain these unique elements of Japanese grammar and then compare his English translation to translations of the same Japanese patent into French and German that appear on the EPO website and websites of the patent offices of respective countries. So instead of trying to invalidate a translation by arguing that the Japanese word “ki” means device and not apparatus or vice versa (it depends), and what the Japanese word “mono” means in this particular patent (it can mean almost anything you want it to, trust me), the lawyers will concentrate on the actual elements of the design rather than arguing about semantics.
   Once a law firm accepts a certain translation, they are “stuck” with it, as one patent lawyer put it to me. That is why sometimes they have to go over it together with the translator with a fine-tooth comb before submitting it as evidence. I was asked once to compare five different translations of a patent, one of them mine, in order to try to establish the meaning of a single sentence. On another occasion, I was asked by a U.S. inventor to translate a Japanese patent related to his invention. He told me he was going to order at least three translations of the same Japanese patent because he believed he was provided with an incorrect translation. I remember a conference call between myself and two patent lawyers who were going over my translation, looking for potential problems before submitting the translation into evidence. The patent was so complicated that two patent lawyers with two different professional backgrounds were needed to grill me for about an hour. It was not very pleasant for me, but one thing I learned about my work from that grilling session is that a patent translator has to be responsible for his translation. He has to be able to defend the meaning of every term, every sentence, and every word based on his education and professional experience. 
   Only an experienced, professional translator can do that. A “bilingual subject expert,” even if there were such a thing, would probably fail the test. And five such experts would most likely make a total mess of what originally might have been a good translation. Fortunately, we know that, in reality, these layers of bilingual experts who proofread and check and recheck everything five times do not exist. They are about as real as the housewife (I mean, homemaker…wrong continent and/or decade) on a TV commercial for a new detergent or margarine brand. The actress has an apron and looks just like a real housewife (I mean, homemaker), but she is really an actress who projects the right kind of domestic astuteness and bliss. She does not use this detergent or margarine brand at all. In fact, her husband probably does all the household chores anyway because he works at home, composing music for movies and commercials, since at this point, she makes much more money than he does, being so terminally cute and talented. 
   What you just saw, folks, was just a TV commercial that has really nothing to do with reality. (I know this for a fact because I used to know a couple just like the one described above when I lived in San Francisco in early 1990s.)

Not all technical translation is suitable for the collaborative, multi-layered approach to proofreading by “bilingual category experts” that is supposedly practiced by some translation agencies

Steve Vlasta Vitek received his master’s degree in Japanese and English studies from Charles University in Prague, Czechoslovakia, in 1980. He worked as an in-house translator for the
Czech News Agency in Prague (1980-81) and for Japan Import Center in Tokyo, Japan (1985-86). He has been a freelance translator specializing mostly in the translation of Japanese, German and French patents and articles from technical journals for patent law firms in the U.S. and Japan since 1987. Before moving to Chesapeake, Virginia, he spent almost two decades in Northern California. Contact:

Back to Top (formerly JATCO)
1304 False Creek Way
Chesapeake, VA 23322

Phone: (757)-312-8515
Toll Free: 1-800-40-JATCO (1-800-405-2826)