Posted by: patenttranslator | April 14, 2015

Editing of Patent Translations – The Good, The Bad, and The Ugly

The topic of today’s sermon by Mad Patent Translator is editing of translations. I realize that I am not exactly an expert on editing of translations in general because evidently, a different approach than what I am preaching in my posts may be required depending on the type and purpose of the translation. But since this is the 28th year that I have been translating patents and technical and medical articles from various languages for a living, while editing almost daily my own translations and those of other translators who work for me and getting paid for it, at this point I am probably something of an expert on editing of patent translations.

Editing is an important part of a somewhat mysterious translation process. Mysterious because regardless of what kind of equipment and tools a translator may be using, the actual process takes place in the central processing unit of human body called brain, a mostly unexplored territory that is not easily accessible to medical science.

There are things that I know and understand about this process, and then there are also things that I think I know about it, although I don’t really understand them.

For example, if at all possible, I edit my own translations the next day, although this is sometime not possible, especially with short translations which are often due ASAP. I don’t know how exactly it works, but I will catch more mistakes, omissions and typos if I can sleep on it. The same principle applies also to blogging – I simply don’t see my own typos just after I have finished a post. So I just quickly proofread the post before I lose my inspiration and click on the Publish button to start sharing it with the world, although I know that the real proofreading will have to wait until the next day.

I also know which mistakes I am most likely to make when I translate and I try to be particularly careful about my own potential mistakes when I proofread my own translations. For example, I tend to mistake number 3 for number 8 and vice versa in small font, especially if the document is not perfectly legible, which is perfectly understandable. But when I am tired, I sometime also mistake “dessus” (above) for “dessous” (below) in French. Oddly enough, I sometime tend to make the same mistake in Japanese and for some reason instead of 上 (above), I see 下 (below) in compound nouns containing this character. Especially in patents dealing with semiconductor manufacturing techniques, various layers and components may be located either below or above other layers or components, which means that the context is of no help in this case.

Many translation agencies emphasize on websites aimed at gullible customers that several highly qualified translators and proofreaders are working on every document that they translate to make sure that the result will be absolutely perfect.

It is of course not true. As I wrote in a long article published in the ATA Chronicle by the American Translators Association 12 years ago, for one thing, it would not be economically feasible to pay several highly qualified translators who would be obviously likely to charge high rates to work on the same translation, and then to add the agency’s margin, often more than 50%, on top of that. But even if it were economically feasible to work like that, the result would inevitably be a disaster.

More heads do not necessarily know more when it comes to proofreading of complicated “patentese”, which often does not make a whole lot of sense in the original document because one of the aims of the patent is to make a claim that will hopefully cover also new, emerging techniques that were not yet in existence when a given patent application was filed.

A good formula for achieving a really good translation would be as follows:

1 highly competent translator + 1 highly competent proofreader = 1 really good translation

There is a reason why the proverb “Too many cooks will spoil the broth” exists in so many languages.Unfortunately, the proofreaders that most translation agencies use are not qualified to proofread patent translations.

In 27 plus years that I have been translating patents, I have had some feedback from proofreaders of translation agencies who are also among my clients, along with direct clients, mostly patent law firms. But although I have had patent lawyers asking me questions about my technical terms used in my translations, in all those years, I have never had a question from a translation agency proofreader about a clumsy or wrong technical terms in my translation, although the chances are that there must have been a few occasions like that over a period of almost three decades, especially in the beginning.

Most patent applications have numbered paragraphs and the lines are also numbered to make it possible to quickly identify specific portions of the text, which is also very useful for proofreading. In more than 27 years, I have never had a question from an agency proofreader about a technical term that did not seem to make sense. The only time when an agency proofreader catches my mistake is when I miss something in the text that is there in English, for instance when instead of “The once-extracted aniline 5 is supplied from the first extraction stage C into the second extraction stage D”, my translation read “The once-extracted aniline is supplied from the first extraction stage C into the second extraction stage D”.

It would be obvious to anybody who can read Japanese, or German, or French, or one of the other languages from which I translate patents into English, that I somehow managed to miss the number 5 (which is bolded), and where it should be placed in the translation.

But whenever I do commit this egregious error, and it does happen sometime, in-house agency proofreaders always ask me about the missing number – and they need to know where exactly it should be placed in the English translation because they can’t read the text in the original language and thus are unable to make the correction themselves.

Some translation agencies use external proofreaders who do understand the language of the source document. But since the rates offered for proofreading are mostly very low, these proofreaders are usually beginners, which means that the results may vary also with this method.

A beginner proofreading a good translation may do a lot of damage to a really good translation if he or she does not realize that the real job of a proofreader is to simply look for typos, mistakes and omissions of the type mentioned above while making a conscious effort to change as little as possible.

Translation agencies also like to say that their method of using external proofreaders is superior to the way individual translators work who generally do not use external proofreaders and simply proofread their own translations.

This line of reasoning is very convenient for the agencies, but it is again based mostly on fiction. Agencies are simply trying to make a virtue out of necessity. Any individual translator can use an external proofreader, all he or she has to do is to pay for one, and most individual translators are much more qualified to evaluate the abilities of potential proofreaders than a translation coordinator who just works for a translation agency.

In fact, agencies use this method because they are unable to proofread the translation themselves, and because they have no idea whether the translation is good, passable, or potentially even horrible if they don’t understand the language of the source document and/or the subject matter.

It is similar to the reason why they send translation tests to new translators who want to join their stable of translators. If they lack the capacity to evaluate a relevant sample, they have to send them their own test, which may have mistakes in it that they are not aware of as I wrote in another post.

So, here is what I think about editing of translations: The most important ingredient of the editing process is in fact …… picking the right translator for a given job.

Once you do that, it is still important to proofread carefully every translation. But the proofreading is not nearly as important as making sure that you have the right person do the job in the first place, because all the proofreader then has to do is basically just look for typos, omissions, and inconsistencies, which is something that most educated and intelligent people can do, or learn how to do that with proper training.


  1. I love the video!

    I had a training on the same issue you raised here and it is exactly that. Editing and proofreading are not the same and they do have an important role in the whole process. This is absolutely right:

    ‘The most important ingredient of the editing process is in fact …… picking the right translator for a given job’.

    I have translated in the mining and geology field for years and I do have the knowledge to do that. However a month ago, one agency offered me a job En to Pt – Br, which is my area, and their proofreader would review and send the file back to me so I could accept or refuse the changes. As I opened the file I was shocked with that bluish my job had gained, and when questioned about the changes, the answer was he/she did not like the way it had been written. Is it such professional feedback?


  2. shocked by**


  3. Especially beginning translators sometime think that they need to justify their acumen when it comes to “proofreading” by making a lot of changes in a translation, and sometime they fear, not without good reason, that unless they find a lot of things to change, they may not be given a proofreading job next time.

    To ask the original translator to approve or reject a proofreader’s changes is in my opinion a very questionable practice for a number of reasons. First of all, the translator is supposed to do another proofreading for the agency for free because agencies will not pay for translator’s time in this case.

    This is wrong.

    Another problem is that translation agencies who require a translator to work for free, while stopping all other work because this is something that must be done right away, do this because, again, they are unable to determine on their own whether the changes that a proofreader made were in fact warranted.


  4. Agree with everything… plus:
    Reviewers are supposed to find ERRORS, not to make a better text style-wise.
    I did some proofreading, some time ago, but decided I would not do that anymore. To do that and not ‘deny’ a job, I just charge from 30% to 100% of my translation rates after I examine the translation to be proofed.
    I have received quite good translations to proof but also Google’s jobs…
    Some well known ‘LSPs’ hire translators for 20% my rate and expect me to proof their job paying 10% of my rate. Predators!

    Liked by 1 person

  5. If only it were this good. Agencies mess it up, because the PM goes over the proofread copy and then points the finger at the experienced proofreader when they, the PM, does not understand something (like text already translated they think is “missing”, the proper spelling of a medical term, a major typo that the proofreader “missed” or they did not understand, yet they do not know it is a typo in the original document, etc). The problem is the PMs. I sometimes question their education level, as the questions they ask are bizarre. I find that adding a lot of notes within the text so they can get it the first time and not come back with questions is a great way avoid such questions, because they really should know how to spell hydrocephalus.

    And, you should take a break between the translating and the proofreading. Your brain will work better that way- to be away from the document for a while and the come back to it. We learned this in school and it holds true and that is why you find it works.

    And yes, the proper formula is indeed: 1 highly competent translator + 1 highly competent proofreader = 1 really good translation

    Why not stick to it?


  6. “Why not stick to it?”

    Because on paper, it is cheaper to use an inexperienced translator (at 20% of the cost of an experienced one, as ribeirojc mentioned above), and then try to find a pro to fix the butchered job.

    In reality, however, this method usually does not work very well.


  7. Has anybody read that “down-to-earth view of revision” published at:

    It opens with “Revision is a deceptively simple notion.” and unfolds towards the idea that revision (editing) should ideally be done by a tandem working in a friendly atmosphere of non-competitive mutual feedback. That’s exactly what our family company does; I mean my “boss” and me, both translators married for 35 years. I can assure you, nothing works better. I can’t even imagine editing someone else’s translation without keeping in touch with them all the time, one of us reading the source text aloud, the other checking the translation on the screen, both freely commenting, sometimes arguing and sometimes joking: easy, fast and pleasant.

    Liked by 1 person

    • Rennie, that is excellent! It is ideal, for sure. That is how I worked with my associate until I was on my own and it was indeed fast, accurate and fun!

      I was thinking that we live in a society in which most people do not receive a higher education. The last time I checked, 30% of the population does. In this case, I need to work with those who are less educated and have less tact, so I need to train myself accordingly. I start by not taking it at all personally, no matter how much they finger point, yet it is their oversight and they are even being paid.


  8. Sounds like fun – but it would not work with my wife who is not a translator and who generally thinks that all translators are 変人 (henjin, weirdos).

    She should know – she has been married to one for 31 years.

    But she still helps me sometime with poorly legible originals, pronunciations of Japanese names, and things like that, even though I am clearly a henjin.


  9. “Bankoku …” was supposed to mean “Proletarians from all countries, unite!” or something like that. Doesn’t it?


  10. Although I don’t see the characters, bankoku literally means 10,000 countries.

    But since there are only about 180 countries in this world, some translators might be arguing about whether this is the right translation for months, especially if they are colleagues who have been cooperated with each other very closely for many years.


  11. 10,000 countries? So, it’s like myriads. The man who wrote “Bankoku…” in that forum had studied Japanese and Chinese at university and as long as I know had traveled and lived in China or Japan, I’m not quite sure. He joked about “Proletarians..” in Japanese about 3 years ago. How would YOU put this famous slogan in Japanese?


  12. “Bankoku” does mean “all countries”, I was just making a joke about the etymology of the first character of the word which literally means “10,000”.

    So it does mean something like “workers of all countries – unite”, which will of course never happen, although the banker of all countries have become united a long time ago.


  13. Ha, ha, ha, ha ….

    No, I am not surprised one bit.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s


%d bloggers like this: