Mobile menu

What does your back-translation look like?
Thread poster: Samuel Murray

Samuel Murray  Identity Verified
Netherlands
Local time: 02:58
Member (2006)
English to Afrikaans
+ ...
Aug 31, 2008

In a recent thread the issue of back-translation was raised again. In this thread I would like translators who regularly do or have done back-translation to show us what their back-translations look like.

I've taken a few lines from Wikipedia for us to translate. To participate in this thread, do a forward translation of reasonable quality, and then do a back-translation of your forward translation, in the way that you normally do it.

http://en.wikipedia.org/wiki/Myocardial_infarction

Original English:
The patient will receive a number of diagnostic tests, such as an electrocardiogram (ECG, EKG), a chest X-ray and blood tests to detect elevations in cardiac markers (blood tests to detect heart muscle damage). The most often used markers are the creatine kinase-MB (CK-MB) fraction and the troponin I (TnI) or troponin T (TnT) levels. On the basis of the ECG, a distinction is made between ST elevation MI (STEMI) or non-ST elevation MI (NSTEMI).

I'm specifically hoping for responses from translators who regularly do or did back-translation. I want to see what each of us mean by the term.

The purpose of the thread is not to evaluate each other's forward translations based on the back-translations, but only to show what our back-translations look like.

[Edited at 2008-08-31 17:06]


Direct link Reply with quote
 

Samuel Murray  Identity Verified
Netherlands
Local time: 02:58
Member (2006)
English to Afrikaans
+ ...
TOPIC STARTER
Afrikaans Aug 31, 2008

Samuel Murray wrote:
Original English:
The patient will receive a number of diagnostic tests, such as an electrocardiogram (ECG, EKG), a chest X-ray and blood tests to detect elevations in cardiac markers (blood tests to detect heart muscle damage). The most often used markers are the creatine kinase-MB (CK-MB) fraction and the troponin I (TnI) or troponin T (TnT) levels. On the basis of the ECG, a distinction is made between ST elevation MI (STEMI) or non-ST elevation MI (NSTEMI).


Here's my example, from Afrikaans:

First forward translation:
Die pasiënt ondergaan 'n aantal diagnostiese toetse, soos 'n elektrokariogram (EKG), 'n x-straal van die bors, en bloedtoetse om verhogings te bespeur in kardiale merkers, d.i. bloedtoetse wat skade aan hartspiere kan aandui. Die algemeenste merkers is kreatienkinase-MB-fraksie (ook genoem CK-MB-fraksie) en die vlak van troponien-I (TnI) of troponien-T (TnT). Uit die EKG kan onderskeid getref word tussen 'n ST-verhoging- (STEMI) en 'n nie-ST-verhoging- (NSTEMI) hartaanval.

First back-translation:
The patient undergoes a number of diagnostic tests, such as an electro cardiogram (EKG), an X-ray of the chest, and blood tests to detect increases in cardial markers, i.e. blood tests that can indicate damage to heart muscles. The most general markers are creatine kinase MB fraction (also called CK-MB fraction) and the level of troponine I (TnI) or troponine T (TnT). From the EKG a distinction can be made between an ST increase (STEMI) and a non ST increase (NSTEMI) heart attack.


[Edited at 2008-08-31 17:07]


Direct link Reply with quote
 

Lawyer-Linguist  Identity Verified
Portugal
Local time: 01:58
Dutch to English
+ ...
Don't see the point Samuel ... Aug 31, 2008

The exercise is flawed from the outset, in my opinion:

(1) The forward and back translator can't be one and the same person.

(2) The back translator is also not supposed to see the original source. As he/she will obviously be influenced by the original source, the results of this exercise will be skewed.

Not trying to rain on anyone's parade, just honestly don't see the point.

Perhaps it would be better to do it with a colleague, use this thread for now to find someone to do it with in your language pair and report back on the results - the forward translator should just choose a different source to the one posted here.


[Edited at 2008-08-31 21:04]


Direct link Reply with quote
 

Katalin Horváth McClure  Identity Verified
United States
Local time: 21:58
Member (2002)
English to Hungarian
+ ...
Precisely - this is not a valid experiment Aug 31, 2008

If you do the forward translation you know the original. You WILL be influenced by it when you do the back translation. You cannot shut it out from your mind, the only thing that would make a difference between the original and your backtranslation is:
A) intentionally introduced differences (for example synonyms)
B) phrases/grammar in the source that does not come naturally to you, in other words, you know them passively, but normally don't use them actively.

Either way, your understanding of the source is given, because you read it, that was the basis of your forward translation. When you read your forward translation your interpretation of the content will be the same.

I think the purpose of the backtranslation is to check whether people that are native in the target language would get the same meaning that was intended in the original. That's why the forward and backtranslator should be a different person, the latter should not be influenced by the original document.


Direct link Reply with quote
 

Samuel Murray  Identity Verified
Netherlands
Local time: 02:58
Member (2006)
English to Afrikaans
+ ...
TOPIC STARTER
Why not? Aug 31, 2008

Lawyer-Linguist wrote:
(1) The forward and back translator can't be one and the same person.


Why not? After all, why can't a translator himself use back-translation as a technique to verify that his own translation is accurate? Forward and back-translation simply use different principles.

(2) The back translator is also not supposed to see the original source. As he/she will obviously be influenced by the original source, the results of this exercise will be skewed.


I don't think there is any danger if the back-translator sees the original. If you take a look at how my back-translation differs from the original, you'll see that there is hardly any "influencing" going on. My forward translation is idiomatic. My back-translation is direct. Remember, when you translate back, you simply don't look at the original, and you ensure during your translation that your translation is accurate according to the current source text (which happens to be a translation itself).

If a translator can be influenced by the original source, then I suspect many other things could also influence him. As translators, we don't translate stuff under the influence of some other texts. We translate the current source text.

Perhaps it would be better to do it with a colleague, use this thread for now to find someone to do it with in your language pair and report back on the results.


Well, firstly I'm not interested in the "results". I just want to know what other people's back-translations look like.

Secondly, I did consider making this exercise one where an existing translation is back-translated, but we all speak different languages and it would be very difficult to find translations for everyone to use.

If I had said "go to the Wikipedia, do a search for some medical term that is likely to have the same text in your language as well as in another language that you speak, and do the back-translation accordingly, into English", I don't think many ProZians would have taken the trouble. What do you think?


Direct link Reply with quote
 

Sergei Leshchinsky  Identity Verified
Ukraine
Local time: 03:58
Member (2008)
English to Russian
+ ...
It's an old class trick... Aug 31, 2008

Many teachers give such tasks to the students (the would-be translators) to illustrate the fact that translation cannot be "exact" or "precise", it can only be _adequate_. Something is always lost in translation, as the language means differ. (Otherwise, there would have been one common language for all the people in the world.) When I was a student we used to pass the same text through 10-15 students (it took several days). The funniest results are obtained when the next student has never seen the previous translations and does not know about the whole process, so to say, the fresh eye.

The texts will be different.

I used to translate backwards, because sometimes, in joint projects, source documents are provided in Russian, while the final report is drafted in English. And then I have to translate the final rep into Russian... It is not always possible to cut and paste from the Russian source (even if you have one), because the texts are usually processed and reworded... The problems arise when Russian-speaking officials try comparing the source documents against the translated final reps and find discrepancies... it is very difficult to explain that the text was not just copied from the source and there might be (and are) some differences...

[Редактировалось 2008-08-31 21:37]


Direct link Reply with quote
 

Samuel Murray  Identity Verified
Netherlands
Local time: 02:58
Member (2006)
English to Afrikaans
+ ...
TOPIC STARTER
Why not II Aug 31, 2008

Katalin Horvath McClure wrote:
If you do the forward translation you know the original. You WILL be influenced by it when you do the back translation. You cannot shut it out from your mind...


I suspect this depends on your translation style and the way you check your own translations while you translate. If you do translations by reading an entire paragraph into memory and then typing the translation from memory, sure, then having been the forward translator will affect your ability to abstract yourself from the original. But if you translate one sentence at a time, and you compare the individual items within the sentence with its translation, I think the influence of the original source will be minimal.

Say, if you had to translate several paragraphs that were all 80% similar, would you find it impossible to translate it, because the previous translations "influence" the current source text to such a degree that you might make a mistake? Surely not.

...the only thing that would make a difference between the original and your backtranslation is:
A) intentionally introduced differences (for example synonyms)
B) phrases/grammar in the source that does not come naturally to you, in other words, you know them passively, but normally don't use them actively.


It sounds like you're speaking from the perspective of two languages that are cognates, i.e. they are so similar that a forward translation and a back-translation will necessarily be the same if both translators are given a glossary for it.

Either way, your understanding of the source is given, because you read it, that was the basis of your forward translation. When you read your [back-translation] your interpretation of the content will be the same.


I don't think so. This would only be true if I could remember the entire source text, or... or if I could recognise the source text in the forward translation, but that would mean the forward translation is extremely direct and possibly not very idiomatic.

I think the purpose of the backtranslation is to check whether people that are native in the target language would get the same meaning that was intended in the original.


You're saying that one should do a back-translation into one's native language, is that right?

If I did this exercise by translating only into my native language, then the exercise will be of no use to any other ProZians because very few of them understand my native language and therefore they won't be able to see what my back-translations look like.

Besides, most of my back-translation work has been into my second language. One could also argue that it is better that the back-translator is a native speaker of the forward translation language.


Direct link Reply with quote
 

Samuel Murray  Identity Verified
Netherlands
Local time: 02:58
Member (2006)
English to Afrikaans
+ ...
TOPIC STARTER
Not quite the same, though Aug 31, 2008

Sergei Leshchinsky wrote:
When I was a student we used to pass the same text through 10-15 students (it took several days). The funniest results are obtained when the next student has never seen the previous translations and does not know about the whole process, so to say, the fresh eye.


I don't think such an exercise is the same as the type of back-translation we're talking about here. In the class exercise, every successive student would create an idiomatic translation of the then source text, and it is when you create an idiomatic translation that sudden changes in meaning can occur if you translate back and forth.

But back-translation as used by clients to check the validity of the original is not an idiomatic translation. The back-translation is deliberately direct... although the text must still be grammatically correct, and fixed expressions shouldn't be translated directly unless they are confusing.

For example, if the forward translation reads "he did not mind the joke", I'm not going to back-translate it as "he did not take care of the joke", even though this would be the most direct, most literal translation of it, because "to mind something" is an established fixed expression in English for "not have a problem with it".

I suspect that if the students had to engage in back-translation of the type we're dealing with here, it would take many, many more iterations before the text became funny.


Direct link Reply with quote
 
xxxmediamatrix
Local time: 22:58
Spanish to English
+ ...
Something's not working right... Aug 31, 2008

[quote]Samuel Murrary wrote:
...
If you take a look at how my back-translation differs from the original, you'll see that there is hardly any "influencing" going on.
...
[quote]

I don't know how much time elapsed between:
- your selecting a text from wikipedia and doing the forward translation
and
- your doing the back-translation.

If the gap was anything less than 48 hours then I am very surprised at some of the differences between the wikipedia text and your back-translation. After all, your source text is only 75 words - and there's quite a lot of repetition in that.

No disrespect meant, Samuel, but it suggests to me that maybe you have some kind of short-term memory deficiency!

MediaMatrix



[Edited at 2008-08-31 22:02]


Direct link Reply with quote
 
xxxmediamatrix
Local time: 22:58
Spanish to English
+ ...
Something's wrong here, too Aug 31, 2008

Samuel Murray wrote:
...
As translators, we don't translate stuff under the influence of some other texts. We translate the current source text.
...


Every time I translate (and, indeed, when write something original) I am influenced by other texts - and especially those written by the author's (or my) peers in the same area of specialization.

Were that not the case, no-one would ever understand their peers' writings since there would be no commonality of terminology.

MediaMatrix


Direct link Reply with quote
 

Heinrich Pesch  Identity Verified
Finland
Local time: 03:58
Member (2003)
Finnish to German
+ ...
Translator should be native in both directions Sep 1, 2008

Otherwise the result cannot be considered trustworthy. If I'm not native when doing the forward translation I cannot "see" the possible mistakes or misrepresentations of the target text.

Direct link Reply with quote
 

Katalin Horváth McClure  Identity Verified
United States
Local time: 21:58
Member (2002)
English to Hungarian
+ ...
Article on back translation Sep 3, 2008

There is an article on back translation in the current issue of the ATA Chronicle magazine,
page 22-25.

[Edited at 2008-09-03 00:05]


Direct link Reply with quote
 

Samuel Murray  Identity Verified
Netherlands
Local time: 02:58
Member (2006)
English to Afrikaans
+ ...
TOPIC STARTER
The deficiency is selective and deliberate Sep 3, 2008

mediamatrix wrote:
No disrespect meant, Samuel, but it suggests to me that maybe you have some kind of short-term memory deficiency!


Yes, but the deficiency is selective and deliberate. As a translator you can choose to abstract yourself from certain influences, if you are aware of them. It also helps if you are able to focus on the sentence.

It is not a bad thing to allow yourself to be influenced. In any text where internal consistency is important, we allow ourselves to be influenced by that which we had translated a few sentences ago. This is a good thing.

But if you have three texts about the same topic, from three clients, you should be able to translate one paragraph from one text, then move on to the next text, translate one paragraph, then move on again, etc, and maintain three different styles in the three texts, based on what is most appropriate for the then current source text.

Do you not think this is a skill all translators should have? Or perhaps it is a skill one picks up in environments such as newspaper offices where multiple texts come in simultaneously and the different levels of functional management keeps changing their minds about which translation is more urgent at that particular minute.


Direct link Reply with quote
 


To report site rules violations or get help, contact a site moderator:

Moderator(s) of this forum
Maria Castro[Call to this topic]

You can also contact site staff by submitting a support request »

What does your back-translation look like?

Advanced search


Translation news





CafeTran Espresso
You've never met a CAT tool this clever!

Translate faster & easier, using a sophisticated CAT tool built by a translator / developer. Accept jobs from clients who use SDL Trados, MemoQ, Wordfast & major CAT tools. Download and start using CafeTran Espresso -- for free

More info »
Déjà Vu X3
Try it, Love it

Find out why Déjà Vu is today the most flexible, customizable and user-friendly tool on the market. See the brand new features in action: *Completely redesigned user interface *Live Preview *Inline spell checking *Inline

More info »



All of ProZ.com
  • All of ProZ.com
  • Term search
  • Jobs