General Remarks on Translation

Re the choice of texts for exams: We avoid fictional (literary) texts, as well as technical subjects. We are only concerned with texts whose primary purpose is to convey information to the general public (journalism).

Re clarity: I would improve on the clarity of the original in the translation if necessary. This doesn't mean go to extremes and add a lot of information. Clear is clear. But if it is not clear, yes, do what is necessary to make it clear in the translation. Clarity is the highest priority. If the translation is not clear, there is no point in doing it at all. If the original is not clear (which is more often the case than one might think), we must clarify it first, before we translate. If we cannot understand the original, we cannot translate it.

Re different "types of translation": How, then, one might ask, do computers do it? Well, I don't think they do--not yet, anyway. They do not "understand" the sentences they process. They react to and produce code equivalences, but not messages. That's why they are usually "bad translations"--because they are not really translations at all. Translation means going from the code to the message in the source language (here, German), and then from that same message to the code that expresses it in the target language (here, English).

So if we can say that computers translate at all, they can translate only word for word. To take an example from English to German, a word-for-word "translation" of It's raining cats and dogs would be *Es ist regnend Katzen und Hunden. A more sophisticated computer program might produce what is called a literal translation. This means the word-for-word "translation" is adjusted to conform to the grammatical structure of the target language: Es regnet Katzen und Hunden. A free translation ignores the linguistic structure of the source text (source code) and expresses its meaning (the message) in the target language (target code): Es schüttet.

Free translation is not only what we are always striving for but is really the only kind of translation that is possible. Word-for-word or literal "translations," inasmuch as they fail to communicate meaning, are not translations but something else, for which we seem to need a new term, so I will suggest one: transposition. This would mean transposing, or exchanging, certain code units (words) in one (source)language with certain code units (words) in another (target) language, which is what word-for-word and literal "translations" are.

<!-- Start of StatCounter Code -->
<script type="text/javascript">
sc_project=3589817;
sc_invisible=1;
sc_partition=42;
sc_security="d9ca9042";
</script>

<script type="text/javascript"
src="http://www.statcounter.com/counter/counter_xhtml.js"></script><noscript><div
class="statcounter"><a href="http://www.statcounter.com/"
target="_blank"><img class="statcounter"
src="http://c43.statcounter.com/3589817/0/d9ca9042/1/" alt="web page hit
counter" ></a></div></noscript>
<!-- End of StatCounter Code -->

Hosted by www.Geocities.ws

1