Re: reversible translation

From: Paul Zellmer (
Date: Mon May 19 1997 - 19:15:40 EDT

Larry & Beth Hartman wrote:
> <...snip>
> This would be my practical experience. My feeling is, if it isn't
> practical, nor probable, should I waste any time with it? The uses of
> reverse translations in field work or study doesn't seem to have many
> practical applications, especially since we have the originating texts
> (or close to it). If we didn't have the originating texts, how could we
> begin to guage the accuracy of reverse translation? In the end, without
> the text, all our translation turns out to be is mere speculation. And
> if I had the originating text, what need do I have of the reverse
> translation, since the original is my goal to begin with?


Your [limited] presuppositions for the reasons for translation are
showing. Much of the translation work going on today is not just to
make the text more understandable for people who already have a
multitude of translations. Many are done for groups that do not have
the Bible in their languages, and these tend to be done, not by large
groups of scholars, but by small groups of Christians with relatively
limited education. Back-translation is an important tool for these
works, not because it recovers the original text, but because it reveals
the interpretations that the translators made in their translation from
the original. These interpretations can be examined to check the
accuracy of the translation.

> <...snip>
> At times it is difficult enough to translate forward from the target
> language, taking into account differing grammars, words that may be
> synonyms, but not totally conveying the original idea, or differing
> cultural slang and jargon. With all of this even the forward
> translation is unlikely to be totally accurate. With every generation
> of translation you will find that accuracy will be lost. Think of
> taking a photograph of a photograph. Can anyone imagine the mishmash
> that would result if we translated the NIV or Living Bible into Arabic?
> May I be so bold as to ask, who would trust such a translation? I am
> speaking now in practical terms, not theoretical.

And this is why, in practical terms, much of the output of groups like
SIL are technically classified as "paraphrases."

> <...snip>
> If anyone, without knowledge of the original text, has succeeded in
> making a reverse tranlsation which is accurate, he should look up our
> friends who are making machine/computer translation programs. His very
> unique skills I believe would be invaluable to that industry and
> research. [He may even become very wealthy. :) ] The difficulties
> presented in computer/machine translation and reverse translation are
> not too dissimilar from one another.
While I would agree that the "reversibility" is an unrealizable
characteristic of a translation, a back-translation of a translation
would result in something closer to the original than would the
back-"translation" of a paraphrase. And, theoretically, even a
translation using dynamic equivalence should be reversible IF the
infamous "deep structure" is marked with enough of the nuances of the
surface structure. The loss of these nuances seems to be the biggest
drawback to the better machine translations that I have seen.

Apparently you must think that these difficulties are not
insurmountable, or you wouldn't be "wasting your time" with machine
translation. :^>


This archive was generated by hypermail 2.1.4 : Sat Apr 20 2002 - 15:38:16 EDT