|The cup is on the table. Which table? Image © | Dreamstime.com|
Translation is often divided into literal and free, making a distinction between translating the words of the original and conveying its meaning. In most applications, the former is frowned upon. But is it even theoretically possible?
While literal translation is usually deprecated as too mechanical, violating target language conventions and potentially losing the meaning, it does have its use. Even professional translators routinely fall back to literal translation in situations where they have exhausted all other possibilities. This happens mainly in two situations:
- When the translator is uncertain about the meaning.
- When the translator is in awe of the status of the original and wants to “stay true to the source”.
So the point of falling back to literal is to avoid making decisions that would require competence and bring along responsibility. Here we could paraphrase the myth in the heading as “It is possible to avoid decision-making in translation by sticking to the original”. Let’s see why this can’t possibly be true.
Basically, this is an extension of the previous two myths: if words don’t exist and don’t have meanings, then consequently they can’t have equivalents in other languages either.
Suppose you would have to translate this sentence to a language of your choice, literally:
“The cup is on the table.”
I’m using German as the example target language here. Other differences aside, most people would translate the last word as “Tisch”.
Why not “Tabelle”? Where does the information come from that enables the translator to consider “Tisch” the literal translation here? Maybe I as the writer of that sentence have a printed spreadsheet in front of me and have inadvertently placed my coffee cup so that it obscures my view of the figures in the table? How are translators to know? The fact is, they don’t. The information simply isn’t available anywhere except my (the source writer’s) head. The only thing translators can do is guess. To paraphrase Abraham Lincoln, some of them guess correctly most of the time, but it just doesn’t happen that all of them guess correctly all of the time.
By guessing, the translator adds information. Selecting one of the two alternatives contains more information than the original ambiguous state. In a direct, measurable sense this information doesn’t come from the source text, or the context, or the background reading the translator has done, or her university training, or any other external source that translators usually refer to. It comes from the translator’s mind only.
It is well known that if you give the same non-trivial text to several translators to be translated literally, each translator turns in a different target text, many of which can be equally acceptable literal translations. Likewise, if you show a particular target text to a bunch of translation evaluators, each of them will have a different opinion about its acceptability. Of course, both are also a function of experience, so the same translator or evaluator would make different decisions at different points in time.
The property of being a literal translation is not a property of the two texts. Neither “Tisch” nor “Tabelle” is objectively the literal translation of “table” in the sentence above. The literalness is subjective, decided by the particular translator based on their prior experience and their understanding of the source text, the translation task, the client’s needs and other background factors.
So the decision is still made by the translator even after supposedly falling back to the mythical literal translation. There is no way to delegate that responsibility to the source text or any other external source. If you look close enough, translators are only ever true to themselves.