Translate Differently and Don’t Fear

Translators always say their translations are “better” than machine translation, but what does “better” mean for your clients? How do they judge?

Early this year, a post popped up on my Facebook feed: a Russian professor of philosophy at a German university was very excited about DeepL, a machine translation (MT) service he recently discovered. He wrote: “It’s hard to beat, especially in comparison to Google Translate. Extraordinary!” Most comments under his post were just as euphoric: “I moonlight as a translator, wrote one philosophy student, but now I guess I can pack up and go home.” Well, I personally have been making a decent living off my translations for more than two decades. I’ve also been familiar with DeepL for quite a while now and haven’t felt threatened.

I usually don’t let myself get dragged into discussions concerning MT with “laypeople,” but this time I felt differently. So, I went ahead and replied to the philosophers on that Facebook thread.

A Short Trip through Time

Whenever someone compares human translation to MT, another comparison occurs to me: drawing or painting versus taking photographs.

The birth of photography must have been a revolution for its day. Suddenly, it became possible to get a photo of yourself without having to sit in a particular posture for hours, spending a fortune while you waited for someone to paint a portrait. In addition, the new photographic technology promised a more accurate, lifelike representation, free from an artist’s fancies and whims.

Visual representation is always secondary; it requires and assumes some primary reality. For a painter, a draftsman, or a photographer, this might be a landscape, a street scene, a model, or objects for still life. In the same way, translation can only be secondary, meaning that a translator’s primary reality is the source text. Like drawing and painting, human translation takes effort and time. The outcome is largely contingent on the translator’s mastery of the subject matter and linguistic competence, but also, at least to a certain extent, on their preferences and writing style.

Machine translation, on the other hand, is not only cheaper or even free of charge, but delivers a quick on-the-spot-result. What’s more, it’s devoid of the individual subjectivity of human translators, which, from the perspective of the author of the source document, may be a good thing. But is it really?

What Is “Neutral” and “Objective”?

Fast forward through 180 years of photography and now anyone anywhere can take a photo at any time. To capture our primary reality through a photo, you don’t have to go to a professional photographer or be one. Nonetheless, corporate clients still rely on professional photography services (e.g., in the area of product photography), although it might make more sense economically for them to invest in pro-grade photo equipment and do it themselves rather than pay higher fees for a service each time. So, why do these clients spend a lot of money on something they could potentially do themselves?

In my opinion, there are two reasons. Let’s assume that technical skills are not a differentiator, that this factor is actually a given, and safely set it aside. If technology matters at all, it’s because of what it isn’t capable of. In other words, there are no cameras or gadgets that can read your client’s mind. Only human beings have the ability to listen to and accurately understand a client’s needs. I believe that every professional who interacts directly with their clients must possess this skill if they want to be successful. In this sense, product photographers and translators are very much alike.

Taking photos of products (or writing or translating the copy pertaining to them) is, in the broadest sense, producing materials for customer communication. Therefore, translators, especially in the field of verbal corporate communication, can learn a lot from the effective strategies used in visual communication and design, especially when it comes to customer orientation. Employing techniques derived from design theory and practice will help us recognize our clients’ needs, deliver noticeably better translations, and, ultimately, become more visible and successful as translators.

Professionalism Alone Is Not Enough

One of the main reasons product photographers are still used is their talent for listening to their customers’ needs and “translating” them into visual results. This is what makes them professionals. Another main reason why people still hire product photographers is that they’re seeking a particular creative solution or “personal touch.” Stick 20 photographers in a room with one product and you get 20 different perspectives, some of which might be real attention-grabbers.

Ming Thein, one of the most influential commercial photographers (and writers on photography) today states:

The expectation of a photograph is that it is an unmodified accurate and complete representation of what we’d see if we were there. There’s a problem with this, however: what two observers see might be the same, but what they notice is probably going to be very different.1

In my opinion, this also applies to our profession. And that’s a good thing because it’s how we can stand out, both from our human competitors and from MT.

Same but Different

A photographer has one product and takes different photos of it, each worth a thousand words. Likewise, if you take a thousand words of source text and compare a number of translations, I’ll bet each would be different. Why? Like photography, it’s about the framing.

Some might find it surprising that their choice of words can influence (frame) how they’re perceived, for better or worse. However, for translators like us, choosing words carefully is simply how we work. Of course, it’s not a matter of choosing just words, but choosing different ways to phrase and transform the client’s source text to get the message across in a more purposeful and targeted way. These differences, or frames, are not only black or white, positive or negative, but may include many other shades and connotations, depending on the subject and situation.

You can refer to this loosely as writing skills or, more precisely, the rhetorical competence of a translator. They say that translators are the most attentive readers of all. Therefore, good translators try to make a silk purse out of their source texts. They rectify the author’s minor errors, clarify and resolve ambiguities, and rework the message to make it more palatable to the target group. I know clients who occasionally even change their own source text after it’s translated to match the (better written) translation. Ultimately, it’s not the means (e.g., different phrasing, etc.) that matter, but the effect of the words that will determine whether or not the translation achieves the communicative purpose. And you can only achieve the right effect through an intimate understanding of the client’s needs, which is something MT can’t do.

Talking to the Client

I remember a conversation I had with one of my clients, an executive assistant to the chief executive officer of a large international corporation. Once, while having dinner after a conference, he told me his job seemed quite similar to mine. Like me, he gets his “source texts” from his boss and has to “translate” them for the target audience. The only difference is that he does all this within the same language.

I’m well aware that this type of translation is not applicable to every situation. There is a concept in translation studies called “Skopos Theory,” which emphasizes the purpose first and foremost. According to Wikipedia:

To translate means to produce a target text in a target setting for a target purpose to target addressees in target circumstances. In Skopos Theory, the status of the source text is lower than it is in equivalence-based theories of translation. The source is an ‘offer of information,’ which the translator in turn remodels into an ‘offer of information for the target audience.’2

Nothing against translatology, but I think that as a translator, you can almost learn situational awareness, to use a military term, and client orientation better from the viewpoint and experience of a designer than you can from academic theory.

Back to Photography

What does all this have to do with photography and DeepL, having meandered as far as we have from my introductory thoughts? Well, those who want their photos taken are not necessarily interested in an ultra-realistic, “no-filters” outcome. A true portrait will always be different from a neutral view stripped of involvement, or a casual snapshot taken without thinking. Also, a good photographer or designer doesn’t just blandly accept their client’s drafts and ideas as they are, but clarifies the client’s purpose and intent with their own vision.

My experience is that the majority of my clients are not necessarily interested in translations “for their own sake.” They expect me to transmute their source text into something different in another language. There are even situations where a conventional translator’s accuracy might be counterproductive.

Crisis or Opportunity

However, this kind of accuracy, or “foolish consistency,” as Ralph Waldo Emerson would have put it, is pretty much the best that MT can achieve. This is because software, as is the case with hardware or any technical device like a camera, can’t read the client’s mind, and this is where we have an edge over MT.

If we take the three dimensions of the so-called magical triangle (time, cost, and quality), the third factor is the only thing we humans can turn to our advantage, but only if we deliver different results. And on that fateful day when MT is declared to have lived up to the claims of software engineers and achieved a level of development and sophistication so that all doubts and embarrassments evaporate, we’ll have to offer something entirely new.

The product of a human translator will always be different—from MT output, from some other translator’s rendering, and possibly from the source text. This difference can be leveraged for better competitive advantage. We can avoid becoming obsolete by exploiting it. Last year, Heike Leinhäuser, president of the European Union of Associations of Translation Companies, wrote on LinkedIn:

For certain types of content, MT is an incredibly useful tool. But if companies buy into the popular belief that translation can be done by anyone (or any one machine), they run the risk of overlooking the very thing that sets them apart from their competition: the way they communicate. If a company aims to capture the hearts and minds of its customers, communicating with a voice that speaks to those customers is key.3

It’s not enough to be professional and good at what you do. Having that unique voice may be crucial. Product photographers still manage to get paying work despite the fact that the client’s own employees can take photos. This situation is not actually dissimilar to, for example, English translators in Germany. Much of the work they manage to pick up could have been done by the clients themselves. However, professional into-English translators still get jobs, primarily because their clients want them to deliver a translation with a difference.

In Our Digital Photo Lab

Below is my response to the conversation started by my friend, the philosopher, on Facebook.

DeepL is certainly a good tool. However, it depends on who is operating it and how. For comparison, photos are taken with a camera, but not by the camera itself. To extrapolate the example a bit more, let’s say MT can, at best, shoot images in RAW (unprocessed) format or, in Adobe’s language, DNG (digital negative).

This kind of RAW translation is often sufficient “for information purposes only.” However, when it needs to do more than just inform—when it needs to do something like persuade, for instance—RAW certainly won’t be enough.

In photography, RAW images look flat, “off,” and very much in need of enhancement. It takes a specialist to create something more desirable out of them using RAW image editing software. This is also the case with translation. Image and language are the same that way. And in particular, when we’re dealing with human communication, without even mentioning the power of framing, a tool such as DeepL, or anything else, is better left to the hands of a professional. Well, just as long as that person has a mastery of the subject matter, an excellent level in both languages, rhetorical competence, and writing skills.

Don’t Be Afraid to Be Different

Interestingly, many agreed with my comment on Facebook. However, what I didn’t point out in that thread is the following.

We can’t translate faster than MT, and we can never be that cheap either. The only thing we have left is the quality of our writing (or re-writing). Translators always say their translations are “better” than MT, but what does “better” mean for your clients? How do they judge?

“If you don’t want to be replaced by a machine, don’t act like one,” as the adage goes. Growing to meet clients’ needs that cannot be met with MT, we don’t have to be afraid to translate differently. I find the advice of Chase Jarvis, another professional photographer, actually quite helpful: “Don’t aim for ‘better,’ aim for ‘different.’ It’s funny how related ‘better’ and ‘different’ are.”4 To put it bluntly: if you don’t want to be replaced by DeepL, translate differently. To slightly paraphrase Chase Jarvis, “add value, don’t be a monkey with a tool.”

As a sidenote, I originally wrote this article in German. To give DeepL a field test, I installed the DeepL Pro plugin and let it process the entire piece. Apart from a number of clear-cut mistakes and misinterpretations, I found the outcome readable overall, but only for the purposes of information. As an author, I could never live with such a translation. So, I had to retranslate this article from scratch myself. I had never felt too threatened by the arrival of DeepL before, but now, having actually tried it, I fear something else: my own overconfidence.

Notes
  1. Thein, Ming. “Crystal Ball Gazing: Predicting the Photographic Ecosystem in 10 years, Part II,” http://bit.ly/Ming-Thein.
  2. Skopos Theory, a niche theory in the field of translation studies, employs the prime principle of a purposeful action that determines a translation strategy (http://bit.ly/wikipedia-Skopos).
  3. Leinhäuser, Heike. “The Shift from Language Service Provider to Premium Language Partner,” http://bit.ly/Heike-Leinhäuser.
  4. Jarvis, Chase. “10 Things Every Creative Person (That’s You) Must Learn,” http://bit.ly/Chase-Jarvis.

Valerij Tomarenko is a full-time German>Russian and English>Russian translator and interpreter based in Hamburg, Germany. He is a member of the Bundesverband der Dolmetscher und Übersetzer (Federal Association of Interpreters and Translators) in Germany and the International Association of Conference Interpreters. He is the author of several publications on translation business practices and, since 2011, has also been writing on his blog Anmerkungen des Übersetzers (Translator’s Notes). His upcoming book, Through the Client’s Eyes: How to Make Your Translations Visible, will be published this year by the BDÜ Fachverlag in Germany. Contact: info@tomarenko.de.

The ATA Chronicle © 2019 All rights reserved.