Beat the Machine: Putting Technical Translation Under the Microscope (Sort Of)


A vintage toy robot
Photo Credit: Unsplash


By Sam Mowry

Welcome to the February follow-up of our Beat the Machine challenge! In our January post, I suggested a sentence to be translated and asked FLD members to submit their own versions, presumably improving upon the machine translated option. Now it’s time to go over some of their particularly interesting solutions.

As a refresher, this is what we were working with:

La gestion des résultats hors spécification a été revue au travers du dossier suivant : n° XYXY relatif à la fiche n° 123 de maléate de trimébutine dont le point de fusion a été mesuré non conforme ; l’hypothèse d’un capillaire trop rempli pour l’analyse a été confirmée par les séries de mesure n° 2 et n° 3 qui ont donné des résultats conformes.

And here’s what Google Translate gave us:

The management of non-specification results was reviewed through the following file: No. XYXY relating to sheet No. 123 of trimebutine maleate whose melting point was measured as non-compliant; the hypothesis of a capillary too full for analysis was confirmed by series of measurements n ° 2 and n ° 3 which gave consistent results.

Isn’t that fun? No points awarded for guessing this month’s theme, which is clearly SUPER DUPER technical. If it weren’t patently obvious (see what I did there?), this sentence was supplied by our beloved FLD colleague and technical translator extraordinaire, Karen Tkacyzk. Thanks, Karen, for this fascinating glimpse into technical translation. While this sentence struck fear into many hearts this month, mine among them, it’s an excellent opportunity to reflect and appreciate how varied the world of translation is. Even within a single language pair (French into English), the range of materials to be translated runs the gamut from literary fiction to texts like this one and literally everything in between. From a marketing perspective, it’s a good reminder that it’s almost impossible to specialize too narrowly, because this kind of extremely specific text exists in the world and needs to be translated. From a competition perspective, it’s a delight to remember that the vast majority of FR>EN translators are your colleagues, not your competition. I’m just one example, but this text is so far from the kind of texts I work with, and more importantly, it’s even farther from the kinds of texts I have any desire at all to work with. There are more than enough topics for everyone—and on the rare chance that there are many translators specializing in your language, direction, and specific subject: what a gift! A community you can reach out to when you get stuck on a term!

 Karen, blessedly, provided two translations, in her words, “the first one fairly faithful and the second more me writing what they mean”:

Translation 1:

The management of out-of-specification results was reviewed through file No. XYXY regarding form No. 123 for trimebutine maleate, where the melting point was measured as nonconforming. The hypothesis given of testing having been done with a capillary that was too full was confirmed by second and third measurement series, which gave conforming results.

Translation 2:

The management of out-of-specification results was reviewed through file No. XYXY regarding Certificate of Analysis No. 123 for trimebutine maleate, where the melting point measurement did not comply. The hypothesis given, that this was caused by testing with a capillary that was too full, was confirmed by two more series of measurements, where the results complied.

 To translate this yet again into what the sentence actually means (for laypeople like myself): there was a result that didn’t fall in line with the numbers it was supposed to. It was used as a case study for how that kind of result is handled. In this case, specifically form 123 in file no. XYXY, the melting point of a specific chemical seemed wrong. The people testing hypothesized that there was too much of said chemical in the tube to get an accurate result, which they verified by doing it two more times. Then the results were good.

Due to the nature of this sentence, evaluating the submissions we received is more a case of pass/fail, “Is this correct?” than critiquing fun turns of phrase. If you submitted a translation for this sentence, thank you! I really appreciate it, and you did a great job. All the submissions we received were reasonably accurate. I wanted to highlight one that read as particularly smooth to me, as someone without a technical background:

Out-of-specification result management was reviewed using File No. XYXY relating to Sheet No. 123 for trimebutine maleate, whose melting point was found to be non-compliant. Test series No. 2 and No. 3 yielded compliant results, which confirmed the hypothesis that a capillary tube had been overfilled during testing.

I asked Karen for her professional opinion, and she noted that, “Whoever submitted it knows what’s going on and is a decent technical translator.” Congratulations, anonymous submitter! Karen said that the only thing she’d change is that “during testing” at the end of the sentence is ambiguous, but in the source, it does mean the first series. She suggested “…during initial testing,” or “…during the first series.”

Thanks again for all of your submissions! Stay tuned for next month, which I promise will be very different indeed!

Did you forget to submit a translation in time? Not to worry! Share your version on Twitter and tag the French Language Division (@ATA_FLD) and me, @SamTranslates.

If you would like to submit a sentence for a future slam, I would like that very much! You can contact me, Sam Mowry, directly at sam [at] or on Twitter at the handle listed above. You can also contact the À Propos Editor Ben Karl at ben [at]

If you’d like to help launch a similar slam but into French, please also reach out!