Iconic Linkage & Textual Cognetics
The following is something I prepared for a
poster session in the School of Applied Language and Intercultural
Studies. It's a couple of years old but it describes the basic concept
behind my research at the moment.
Q: What is Iconic Linkage?
A: Repeating the same target language structure for
different non-isomorphic, but semantically identical source language
structures.
Q: So how does this improve the quality of user guides?
A: It improves consistency not just within entire
texts but across multiple texts; makes information more accessible; helps
users remember information and assimilate information quickly.
Q: Why User Guides?
A: Software user guides were chosen primarily
because of their factual nature and also because of the fact that a lot of
information is repeated or is common to various parts of text.
Q: Isn't Iconic Linkage just parallelism with a different name?
A: Not really, although Iconic Linkage is quite
similar to parallelism. Instead of parallel structures in lists or clauses
- essentially in small localised areas of text - Iconic Linkage spans
larger amounts of text, entire texts or even multiple texts.
Q: But doesn't using the same translation defeat the purpose of
translating?
A: That depends on how you view translating and
translations. If you view translation as a source-based process intended
to reproduce the source text, then yes. If you view translation as the
design and production of a product designed to do something for its
readers, then no. This approach serves to make the communicative act more
efficient.
Q: Can you give some examples of Iconic Linkage in practice?
A: The following are examples of full Iconic
Linkage. The sentences mean exactly the same thing.
| Original Sentence:
|
Maschine ist im Arbeits-Modus.
|
| Original Translation:
|
The machine is now in Work Mode.
|
| Second Sentence: |
Die Maschine befindet sich im
Arbeits-Modus.
|
| Second Translation:
|
The machine is now in Work Mode.
|
Using a CAT tool such as Trados, the second
sentence registers as a 67% match with the original sentence but it
actually means the same thing. Consequently, we can introduce Iconic
Linkage by reusing the original translation.
Another example:
Original Sentence:
Wollen Sie ohne Begasung arbeiten, stellen
Sie den Parameter-Wert Gas/Restvakuum auf den selben Wert ein,
wie den Parameter-Wert Vakuum.
Second Sentence:
Für das Arbeiten ohne Begasung
müssen Sie den Parameter-Wert Gas/Restvakuum auf den gleichen
Wert einstellen wie den Parameter-Wert Vakuum.
Translation:
If you intend on working without gassing,
use the same value for Gas/Residual Vacuum and Vacuum. |
Using a CAT tool such as Trados, the second
sentence registers as a 71% match with the original sentence but it
actually means the same thing. Consequently, we can introduce Iconic
Linkage by using the same translation for both ST sentences.
The following are examples of partial Iconic Linkage - the sentences are
almost identical except for certain factual changes (formatted like this).
Original
Sentence:
Die essentiellen
Aminosäuren
kann der Körper nicht selbst herstellen, sie müssen
daher mit der Nahrung aufgenommen werden.
|
Original
Translation:
The body cannot produce essential amino acids itself and as a result they must be obtained in
the diet. |
Second
Sentence:
Die ungesättigte
Linolsäure und GLS
kann vom Körper nicht selbst hergestellt werden und muss
daher mit der Nahrung zugeführt werden.
|
Second
Translation:
The body cannot produce unsaturated linoleic acid and GLA itself and as a result they must be
obtained in the diet. |
Using a CAT tool such as Trados,
this second sentence registers as a 60% match with the original
- but only the main words (object) is substantially different.
Otherwise, these sentences would be complete matches.
Q: OK, but isn't it difficult
to assess the quality of translations?
A: It can be very difficult. Part of the problem is that there
are so many definitions of what constitutes a good translation.
Another problem is that there is a shortage of useable translation
quality assessment (TQA) models. Those that do exist are often:
|
 highly
subjective |
very
difficult if not impossible to implement |
|
not
repeatable |
source-based |
Q: What's wrong with source-based approaches?
A: Instead of regarding translations as a product for a specific
audience, many TQA models (such as equivalence theories) insist
on a source-based approach. This means that the TT is always
regarded as a reflection of the ST and not as an independent
autonomous text, which is how the people who use the translations
regard them. Basically, source-based approaches mean we can't
tell how the text scores in the real world.
Q: So what's the alternative
to source-based assessment models?
A: In practice, translations are generally regarded by users
as "original" texts produced in the target language
and as such they are subject to the same requirements as "native"
TL texts. What we need to do is to develop a TQA model based
on target-orientated criteria.
These criteria include such things as language quality, terminology,
accuracy, style and register as well as publishing and functional
accuracy. These criteria are then used to create a checklist
for assessing the quality of translations.
The assessment model will be
implemented in two stages:
-
Establish whether the translation
actually conveys the information contained in the source text.
This is similar to pragmatic revision except no changes are made
to the translation.
-
Evaluate the translation using
a checklist containing the target-based criteria. This is similar
to "fresh-look" revision.
Q: How do you test whether
Iconic Linkage improves the quality of translations?
A: Two copies of a translation will be produced. The original
version will be left "as-is" while the copy will be
re-written to introduce Iconic Linkage. Both texts, as well as
the source text, will be analysed to establish the amount of
both natural and added Iconic Linkage.
-
The TQA model will then be used
on the original translation.
-
The TQA model will also be used
on the re-written translation into which Iconic Linkage has been
introduced.
The results of the two sets of
evaluations will then be compared.
|
