Up the garden path

Up the garden path

…and into the lion’s den

Productivity is a controversial issue among translators. We all know what we can do and we often don’t like the idea that other people can work faster than we can. If they can, we have a tendency to think they can’t be doing things properly. Now I’m quite tolerant of other people’s productivity, maybe because I’m quite a productive translator myself. I can translate 4,000 words a day to a quality standard that I am happy with, and more in an emergency. So I’m used to people saying “I don’t know how you can do that” and I’m prepared to believe that others may be able to do more than I can. If someone said they could do 6,000 words a day, for instance, I’d be surprised but I wouldn’t be calling anyone a liar.

Some productivity claims, however, go too far. One such appeared a couple of weeks ago, when Wolf Steinhauer offered a webinar claiming to teach translators how to translate 10,000 words a day. It caused quite a stir on Facebook, with many colleagues agreeing with me that such a level of productivity was impossible if quality was to be maintained and others saying they would prefer to see the webinar before making a judgement. The discussions even became quite heated, with one translator accusing me of being “incompetent”, an insult I couldn’t quite understand in the context. Another colleague said she couldn’t understand the apparent anger the advertising of the webinar had aroused.

The point of this post is to explain exactly why this sort of unbelievable claim is objectionable, particularly when combined with the offer supposed translator training, as in this case. To begin with, why do I say it is impossible to translate 10,000 words in a day? After all, I’ve already admitted that everyone has a tendency to disbelieve higher productivity than their own. The problem with a figure like 10,000, though, is that it is very close to the number of words I can edit in a day, and, remember, I’m quite productive. This must mean that either the whole translation is being done in barely an hour so that there’s time to fit in the editing process, or it simply isn’t being edited at all. Either way, the resulting quality is hardly going to be acceptable.


Added to this is the fact that if you do manage to translate and edit 10,000 in a day you are going to be very tired indeed. That will almost certainly put you out of action for the next working day, nullifying any gains your extra productivity has achieved for you. In the unlikely event that you do manage to keep up the pace for more that two or three days at a time, you’ll have had little time to do anything other than work and, as all of us who have done such daft things from time to time know, your health is bound to suffer. So not only will the quality of your work will be down, the quality of your life will also be negatively affected. Where’s the sense in that?

And yet there was Wolf happily promoting his webinar as if he was about to reveal some sort of important secret. What could it be? Maybe I should have attended to find out. There might indeed have been something to learn, as some productivity tricks are useful. My best one, as I think I have explained before, is touch-typing. Others swear by voice recognition software. Some translators even use codes or hotkeys to reduce the amount they have to type. And, of course, there are CAT tools, although if the 10,000 words a day that were being claimed all came out of a translation memory that would be cheating, as no translator is going to get that lucky on a regular basis.

But I didn’t attend the webinar because, as I have already explained, what Wolf was claiming to be able to do is simply not possible day in, day out to a decent quality standard, and that makes anything he was trying to say simply misleading. Because who wouldn’t want to be able to achieve that sort of productivity, especially translators who feel forced, for whatever reason, to accept low rates? So is it fair to tempt these vulnerable translators with a promise of something that can’t realistically be done? I would say it’s not.


But it doesn’t end there. Not only is the idea of translating that much that quickly unrealistic, it is also precisely the wrong approach in a market increasingly threatened by machine translation. No-one can be any doubt that the very first humans to lose their place in the market to machines will be those who translate like machines. Yet this is exactly the way people translating 10,000 words a day are going to be working. And they won’t beat the machines for speed, that’s for sure. Nor will they beat them on price.

Where they might score as human translators is on quality, but Wolf doesn’t set much store by quality. In an e-book he has published, he claims that (the book is in German, but the translation is by a German-speaking colleague) that: “The competence and abilities of a translator or interpreter don’t count for anything. The clients just don’t have any idea how to assess these.” Now, while he may have a point about the inability of some clients to judge quality, the basic idea that translators’ skills are irrelevant is utterly wrong-headed as he is disparaging the only strategy likely to save members of this profession from losing out to those machines.

It’s true that, even at low rates, translating 10,000 words a day could bring in a reasonable living, but for how long? The only way this end of the market is going is down, in both price and quality terms. So what happens when those rates are cut again? Will people be trying to translate 20,000 words in a day? Or 30,000? In the end it won’t matter because, as we all know, Google Translate is free.

And here’s the source of the anger perceived by my colleague in some people’s reactions to the webinar advert, including my own, I have to admit. Because although I’m not going to be fooled, where will those translators who have naively followed Wolf’s bash-it-out-and-sell-it-cheap approach be when machines are doing all the bottom-end translation? They will be entitled to feel they’ve been led up the garden path, but in fact they will have been guided into the lions’ den, as lambs to the slaughter.


  1. Kevin Lossner

    Simon, it’s quite possible to put out 10,000 words of good draft quality in a day; I have seen it done by a number of colleagues who know their subject areas very well (typically having worked in the field, like a semiconductor engineer I met who now translates materials in that field) and who use a well-tuned speech recognition system. An older colleague who used to work in Philips translation department back in the 1970s told me that the throughput of his dictating colleagues was as high as 15,000 words per day, though at that time it was necessary to send the tapes to a transcription typist and have a later round of editing.

    On many occasions I observed one legal translator translate large projects for short filing deadlines with output at that level. I doubted the result the first time I saw it done, but when I checked the work (about 11,000 words done from about 8 am to mid-afternoon with a long lunch break), it looked rather good.

    I don’t routinely put out that much using speech recognition, but take advantage of it for more relaxing, ergonomic working conditions and the ability to deal with long, ugly patent claims, where I need both hands free to point at the screen while I untangle 300 words or so of nested, knotted German clauses. However, given the right text, with speech recognition output of about 10k is entirely possible, with editing and reasonable quality, leaving one much less tired than doing a ithird of that on a keyboard.

    • Simon Berrill

      Thanks for your comment, Kevin. I’m still not quite clear whether the 10,000 word output you refer to is finished work or just a draft to be edited later but either way it’s impressive. You also mention, though, that at least in your own case it’s not something you would do every day. I also maintain that, although we all want to be as productive as possible, it isn’t mere productivity that’s going to put us ahead of the machines.

  2. Amy Stevens

    Interesting post! I often wonder about the productivity of other translators, having started as a freelance translator in the last couple of years.

    I agree that the quality of 10,000 words in a day is probably not very high. I personally translate around 3000 words per day. I have done up to 5000 when it’s been urgent but I don’t like doing this. I like to ensure myself enough time to give my best work.

    • Simon Berrill

      Thanks for your comment, Amy. I think you’re taking the right approach. We’ve all done daft things from time to time to meet deadlines, but it’s not good to make a habit of it.

  3. Miranda Joubioux

    As time goes by, I find I spend more time and not less perfecting my translations. When I do work fast it’s for very run of the mill jobs or for clients that pressure me to do so, but who are aware it will not be my best work.
    Productivity is all about producing a sufficient quantity to make your work profitable. So pricing is also an important variable.
    If you work at low rates then you need to produce more. In these conditions, if you produce a lot to maintain a good standard of living, then you are invariably making quick choices and producing unimaginative translations. As you say, you are churning out words like a machine.
    I don’t like working like that and can only think of a few cases in which it is even possible.

    • Simon Berrill

      I find the same, Miranda. My productivity is going down, not up, but I’m sure I’m producing better translations.

  4. Claire Cox

    I tend to agree with Kevin; I have translated that many words (and more) in a day using Dragon, with straight text and a CAT tool. Definitely first draft, not edited, but the quality is usually pretty good, precisely because when you dictate your translations you are already applying a filter in that it has to sound natural or it’s really noticeable when you say it – much more so than when typing! Yes, this only works in a field you’re really familiar with and after many years of experience, but it’s certainly not impossible. Do I do it every day? No, of course not. But it’s nice to know you can if push comes to shove – and you get to know the clients whose work fits into that category and schedule accordingly. Not in the sense of offering to translate at that rate for days on end, but knowing when you can squeeze other jobs in and what’s realistic when quoting for tight deadlines.

    • Simon Berrill

      If we’re talking about a first draft, and in exceptional circumstances, I find it much more credible, Claire. I may even have got close to it myself once or twice. But what I object to about the claims made for this webinar is that they would lead less experienced translators to think they can produce 10,000 polished words every day, and that just isn’t going to happen.


Submit Comment

Your email address will not be published. Required fields are marked *