It was a PowerPoint presentation, written in English in the U.S. and translated into Portuguese in Brazil. The client had just called to say that the translation was unacceptable and they would not pay for the job. Too literal, the secretary had said.

too many people are overeager to interfere with the original in wanton and unwarranted ways

Now, I don't think literalness in itself is a problem. A translation is unacceptable if it violates the rules of the target language, if it belies the meaning of the original, or if it introduces needless changes in the style of the original. These three capital sins are as common in freer translations as they are in more literal ones.

Once I wrote an article on translating gobbledygook for the Translation Journal, and I do believe technical translators who often deal with unrevised and carelessly written texts have the right to simplify or clarify a text that would otherwise be tenebrifically befuddling to its intended audience. That specific freedom I call our license to kill.

However a license to kill means you may kill if you must, not that you have to mow down everybody in sight. In addition, I believe the license applies only to style, not to content: the translation should say what the author said, no more, no less. This is an ideal, an impossibility, for there is always some element of distortion, but the less distortion, the better the translation. Regrettably, too many people are overeager to interfere with the original in wanton and unwarranted ways.

Respect for the author's choices

Let me give you an example. A black dress is um vestido preto in Portuguese. However, I know several translators who would never accept such a pedestrian rendering. They would say something like um vestido de cor preta, for instance.

Now, I firmly believe that any English-language author who wanted um vestido de cor preta in Portuguese would have chosen to write a dress black in color in English, for the choice was there all the time and the two English forms correspond to their Portuguese counterparts both in style and frequency (the de cor / in color is far less frequent). The translation is latent in the original as the statue is latent in the stone, as Signor Buonarotti is said to have claimed on some occasion or the other. So, if the text says a black dress¸ it is up to me to translate it as um vestido preto, thus respecting the stylistic choice made by the author.

However, sometimes the target language offers a choice that was not present in the source language. For instance, black can be translated both as preto and negro in Portuguese and, if you know Portuguese, you will know that there is a world of difference between um vestido preto and um vestido negro, although both translate a black dress. So, here the translator has to make a choice and that is dictated by context, not by whim or a concern with a hypothetical need to prettify the text.

In short, a black dress can be either um vestido preto or um vestido negro¸ but not um vestido de cor preta or um vestido de cor negra.

The same goes for word order. I saw a movie yesterday is vi um filme ontem or eu vi um filme ontem since both formal and colloquial Portuguese allow me the choice of using a pronoun or leaving it out, a choice that does not exist in English under the same conditions. On the other hand, it cannot be ontem (eu) vi um filme. If the author had wanted the latter, (s)he would certainly have said yesterday I saw a movie, for the choice exists in English. Contrary to what some logicians may say, I saw a movie yesterday and yesterday I saw a movie mean different things and the rhetorical difference should be respected in translation whenever possible. It is a question of emphasis and if the author chooses to emphasize something, who am I to emphasize something else just because, in my opinion, it looks nicer?

The other day, I edited a job by a translator who methodically inverted word order when the original order was perfectly correct in Portuguese. It so happens that there was a short series of sentences that had been carefully planned to place the most important bit of information in the beginning. In addition, the importance of the terms was further emphasized by setting the first few words in bold. A text that carefully integrated form and content, I'd say. Inverting the word order had simply destroyed that integration, and I had to straighten out all sentences. The "straight" translations made perfect sense in Portuguese and were grammatically and stylistically correct. Why change word order then? To make the translator's voice heard?

Who wants to hear the voice of the translator, anyway?

After all, people read translations because they cannot read the original, not because they want to know what the translator thinks the original should have said in the first place. Some translators—principally those who would be writers but cannot find publishers and/or readers—insist they must make their voices heard and firmly believe they are entitled to transmogrify perfectly good originals into translations that have no relationship therewith and even expect applause for the carnage perpetrated.

I firmly believe people engage me as a translator because they need help to hear the voice of the original author. Had they wanted to hear what I have to say, they would be reading the Translation Journal.

Translations vs. trashlations

But I am straying from the original aim of this article, which was to tell you the Case of the Literal Translation. The client service manager showed me the translation and asked for an opinion. Now, some translations are so obviously bad that a cursory look will tell the experienced eye they are mere trashlations.

On the other hand, it is never possible to tell that a translation is good before comparing it against the original. Some translations are so well written that they earn applause from critics too lazy or too ignorant to compare them with the original. Careful comparison, however, will show they are all wrong: the guy who wrote the original simply had not said that. Or not quite that. The French call them belles infidèles: the unfaithful beauties, and a great name it is.

This, contrary to general belief, is more often the case when the translation is done by an expert, meaning someone who works in the area, not a professional translator. For instance, a translation of a medical text done by an MD who feels entitled to tamper with the original just a little bit. In many cases, those experts do not do the translation themselves but nullify other people's efforts in their capacity as consultants or revisors.

Is the author an ass?

Early in my years as a translator, I worked for a publisher and one of my translations was revised by a guy who taught college Economics. He made many changes, some of which I agreed with, and inserted a negative in a sentence that had none in the original. When I complained, he claimed the author was an ass and had got it all wrong. Notice that he did not say there had been a slip in the original, that the author had committed that most common and feared of all mistakes, the error of failing to add a negative where one was required; he did not claim there was a revision or proofreading error: he claimed the author was a jerk and he could not accept such stupidity in a text revised by him.

This same revisor later on published his own treatise on Economics and, hopefully, said things the way he thought they were. His book seems to have achieved a certain degree of success and he seems to be a competent economist—although I am no authority on that: I just translate what those guys write. As a revisor, however, he was a failure. For people who read the book he revised got the wrong impression of what the author had in mind. They bought a book by Professor X but got one from Professor Y instead. A counterfeit.

Shading

Sometimes, things are not nearly as drastic, but equally misleading. I recently compared several published translations with the respective originals and detected a practice I call shading in English and matização in Portuguese, for want of a better term. Shading takes place when the translator plays a bit with modifiers to give the text a different slant.

For instance, when the original says many and the translator uses the equivalent to most. Or when the original says any and the translation says a large number of. Small things, but small in the appearance only, for they deeply distort meanings. Again, here, the expert in the area did not quite agree with the original and thought it was his (or her) job to "correct" it a bit, which it was not. Of course it is perfectly correct and ethical to add a translator's note calling attention to an error in the original—or rejecting the job altogether—but if you deign to do the job, by all means, try to provide a true translation.

Translators sometimes engage in other—but equally objectionable—types of shading, such as using polite and politically correct expressions where the author was very rude and politically incorrect. In doing so, they may transform a rabid racist pamphlet into a placid comment on current affairs and believe that, in hiding the appearance of venom, they have eliminated the poison itself it and thus done society a service. You cannot contribute to the progress of humankind with bad translations.

Back to business once more

But I stray again. As I said, sometimes you can tell a translation is a mess even before you compare it against the original. So I had a look at the text. As far as I could see, it was all right: not the clumsy translationese intermixed with false friends and straight garbage that plagues so many translations done when the translator works on autopilot, is too pressed for time, or should try a career selling hot dogs on a street corner.

Then I asked to see the original. I already suspected a comparison would prove the translation was good, for it had been done by an experienced professional who knows her own limitations and blows the whistle when assigned the wrong kind of job or when the original has some kind of problem. She had completed the job on schedule and had not blown the whistle, ergo, there was a good chance she was in control. But I had to see. So I compared the translation against the original.

It was a powerpoint presentation, as I said, and powerpoint presentations are usually very difficult to translate, for at least two reasons: the first and more obvious is that Portuguese often needs more room than English, and if the source transparencies are overcrowded, cramming the Portuguese into the space available requires rigid discipline and a lot of creativity, two qualities that don't usually come together; the second is that powerpoint presentation are often meant to make sense only in conjunction with an oral presentation and it is often next to impossible to make head and/or tail of them if you're not present at the talk they were intended to support.

However, my friend had performed the task with flying colors. Her translation reflected exactly what the guy had said, the way he had said it and—the icing on the cake—was written in clear, correct and idiomatic Portuguese, as I said above.

Case closed?

I wish the case could be declared closed now. Unfortunately, it wasn't. The client revised the translation and produced what she expected a professional translator should have done; show a model of translatorial competence. So I asked the client service manager to ask the secretary to ask her boss for a copy of the revised translation, which I duly received by e-mail.

I must say that the revised text was a lot better than the translation my colleague has provided. Clearer, more detailed and more energetic, probably a better support for the talk. In addition, it included much information that my colleague had not included in her translation—for the very simple reason it was not in the original. In other words, the client wanted the translator to improve on and add to the contents of the original, something that is far beyond the task of any of us.

As a proof that the revised translation could not issue from that original, I translated several of the transparencies back into English and a simple comparison between my back-translation and the original would show that they said different things—even with a very generous allowance for the fact that English is not my first language. Very possibly it would be a very good idea to back-translate the whole revised Portuguese into English and use it as a basis for future presentations into that language, since it was obviously a better text. However, it was equally obvious that no translator could have start with the original and arrive at the revised text, for they said different things.

End of case

Unfortunately, I don't know how the case ended. I am afraid it ended badly for the agency. The client may have paid for the translation but I don't think they entrusted that particular agency with another job. They probably gave the next translation to somebody else and were equally dissatisfied with the results, for such clients cannot understand that our mission is to reproduce what the guys said, the way they said it, as best we can.