Monday, December 1, 2014

Final: Contrastive Grammar

1. The verb to cast (to throw something forcefully in a specified direction) appearing in the ST opening sentence "Who cast the first fateful tomato…" is rendered as:

TT1: lanzó
TT2: lanzó
TT3: tiro

The lexical unit lanzar is much more likely to have an equivalent effect in the target reader than tirar, because it denotes more strength or violence applied to the action than the broader and vaguer term tirar, which can also be to get rid of, to ditch or to dump. Besides lanzó, another popular equivalent in other translations was arrojó.   


2. Determiner that in “that first fateful tomato”:

TT1: ese
TT2: aquel
TT3: el


The determiner that is a category-one type correspondence because it matches up well with ese. Both these units express in this case temporal distance. However, in Spanish we can express further temporal distance with aquel. It is not a mistake per se because both words convey almost the same meaning (I’m being a bit picky here), but since that first fateful tomato was cast over 70 years ago, I believe aquel does a better job painting a picture of how that situation developed or how long ago that situation took place. As far as the TT3 translation el, it represents a word-class shift (from adjective to article), and it fails to convey the semantic (temporal distance) that aquel or even ese does.

3. “first fateful tomato”

TT1: fatídico primer tomate
TT2: primigenio y profético tomate
TT3primer tomate catastrófico






















The TT1 translation is a faithful translation because it stays within the constraints of the ST and expresses the exact same meaning intended by the author. The TT2 translation is a free translation because it focuses on the content rather than on the form.

As far as the TT3 is concerned, the syntactic arrangement of words, or in more technical terms the immediate constituents, is misleading to say the least. The mistake in this translation can be explained citing the structuralist or taxonimic grammar model.
               
TT1: [fatídico] [[primer] [tomate]]
TT3: [[primer] [tomate]] [catastrofico]

The error in TT3 translation is that it implies that that tomato was only the first of a series of fateful tomatoes. Much to my surprise, this was a mistake that more than 50% of professional translators committed. Below are some examples:

·         primer venturoso tomate.
·         primer tomate fatídico.
·         primer tomate preñado de destino.
·         primer tomate crucial.
·         primer trascendental tomate.

4. “a carnival that got out of hand”


TT1: un carnaval del cual se perdió el control.
TT2: un festejo que se volvió incontrolable.
TT3: un carnaval incontrolable.

Krzeszowski speaks about the universal semantic inputs and the language specific surface structure outputs his Contrastive Generative Grammar is based on. In other words, he breaks down into five stages the linguistic process that operates between the universal semantic inputs, and the language specific outputs. This is, of course, a model designed to compare languages, but –if we think about it– it makes a lot of sense for analyzing translations as well.

A few things can be said about the translations of this phrase in the TT1 and TT2, but all things considered they both get the job done as far as transferring the meaning from the source into the target text. That is, they both convey –mind you, in different ways– the idea that the carnival went from being peaceful, to being a mess. But once again, the TT3 translation has some issues that keeps it from reaching that equivalent effect status. It makes it out to be like the carnival was always out of control, and that is far from what the original text says.

So back to Krzeszowski’s five stages, the TT3 translation only goes past the three first stages: 1) conceptual input (It’s like a framework with different slots: agent, patient, time, space); 2) framing the semantics into categories (like unit, structure, etc.); 3) the syntactic level (minor lexicalizations). However, the 4th stage concerns inserting dictionary words  (major lexicalizations), and TT3 lacks that. It basically lack words like “se volvió,”  “se convirtió,” and what not.

5. “a giant paper maché puppet parade”


TT1: un desfile gigante de marionetas de papel maché.
TT2: un desfile de gigantescos muñecos de papel maché.
TT3: desfile de marionetas de papel maché.

This leads me to the question: Is the parade giant? Or is it the maché puppets that actually are? From an immediate constituents analysis point of view, it appear as if the parade was giant.



However, the TT2 translation as well as almost all of the other translations from the contest totally missed this, changed the syntax, and made it look in the Spanish version as if the puppets were giant.  Below are some examples:

·         “un desfile de títeres gigantescos hechos en papel maché”
·         “un desfile de enormes muñecos fabricados con papel maché”
·         “un desfile de muñecos enormes de papel maché”
·         “un desfile de figuras de gran tamaño confeccionadas con papel maché”
·         “un desfile de marionetas gigantes de papel maché”
·         “un desfile de monigotes gigantes de papel maché”

Having said that, the puppets in La Tomatina festival are indeed gigantic, so it might be that the error is actually in the source text.

6. “They happened upon a vegetable cart nearby and started hurling ripe tomatoes.”


TT1: Al toparse en las cercanías con la presencia de una carreta de vegetales, empezaron a revolear tomates maduros.
TT2: Se toparon con una carreta de vegetales que estaba por allí cerca y comenzaron a arrojar tomates maduros.
TT3: Se encontraron cerca de un carrito de vegetales y empezaron a tirar tomates maduros.

Halliday suggests four fundamental categories of grammar: unit, structure, class and system. He says that these categories are universal, and that they're sufficient as a basis for the description of any language. In the category of unit languages are broken down -from largest to smallest- into ranks, which are sentence, clause, phrase, word and morpheme. The larger ranks consist of the smaller ranks, and this implies a scale that is called rank scale


With that into consideration, Halliday holds that in traditional linguistics any single sentence will always correspond on a one-to-one basis with any single sentence in another language –So was it really necessary for me to modify the ST sentence so much? In the translation of the source text sentence, this is a principle that the TT2 and TT3 translators did apply, as you can see in the diagram:
Both these translations correspond on a 1:1 basis the down to the sentence and clause rank with the ST. TT1, however, disrupts the syntactic features of the original. 
While TT2 and TT3 are faithful translations because they stay within the grammatical constraints of the ST, TT1 attempts to put more emphasis on the naturalness than on the syntactic features, regardless of whether or not it successfully does so.

7. “Innocent onlookers”

TT1: Espectadores inocentes.
TT2: Los inocentes espectadores.
TT3: Los inocentes espectadores.

So I’m going to bring up the ignorance hypothesis developed by Newmark and Reibel. It distinguishes an ignorance-without-interference –which is about structures that are not a problem for learners because they will hardly ever use them– and an interference-without-ignorance, where learners stumble upon the same errors time and time again even when they know that a specific grammar structure is not correct –the can’t help it but to use them. This ignorance hypothesis is generally a theory used to describe the mistakes one makes in the L2 -however, it can also be used to explain a mistake done in the L1 due to background interference form the L2.

That's why I made a mistake here. In my translation I omitted the determining article, just like it is done in English. This error is considered to be interference without ignorance because no native speaker can be said to be ignorant of the central structures of their own language –I do know it’s misguided to omit the article in Spanish, but I did it nonetheless. I paid so much attention to the source text that I neglected the target text–.

It’s also worth noting that this is an intralingual error and not an interlingual one. Part of the job CA has in relation to language pedagogy is to predict mistakes –however, there’s only so much CA can predict that it’s not possible to cover all the variables. An Intralingual mistake is something CA does not predict.

Quick mention:                                                  
Also, I wanted to quickly go over this sentence: "repay the tomato vendors" in which I translated "repay" as "pagarles." A doubt I had was about whether I should to put "pagar a los vendedores" or "pagerles a los vendedores." Most of the other translations used "pagar," but that sounded a bit odd to me, so I researched and found out "pagarles" with the object "les" attached to the verb is optional in some countries, so to say just "pagar" is fine.

TT2 translation said "compensar," which I like it as a translation, and the TT3 translation is "reembolsar," which I think it's okay as a translation, but my only hang-up is that "reembolsar" is a transitive verb, so if you say "reembolsar a los vendedores" it kind makes you think they're gonna put the vendors into a bag or something like that.

Then, with clause "locals who defied the law" I didn't make the same mistake and I put the determining article "los lugareños que desafiariaron la ley." So did TT2 with the determiner "algunos" in "algunos veciones que no acataron la ley..." but I don't like that one because it conveys the idea that they were a few, but we don't really know about that.

Also, here, I made a shameful mistake. “Mock” is translated in both the TT2 and  the TT3 as “simulacro.” However, because I got overconfident I thought I didn’t need the dictionary here. I knew “mock” is “burla” in Spanish, and to make it fit in the context I translated it as “parodia sobre el funeral del tomate.” However, little did I know that mock also means “simulacro”. So I made one of the main mistakes translators have to keep themselves away from: assuming you know something, and not double-checking. This would be something like ignorance with interference LOL

8. "(it) decided to roll with the punches "

TT1: optó por adaptarse al cambio.
TT2: decidió amoldarse a la situación.
TT3: decidió ser flexible.

In this case we have the idiom roll with the punches, which literally it’s used in boxing, but figuratively can be used in any walk of life as well. It means “to adjust to difficult events as they happen.” All the three translations do a good job capturing the meaning of the original.

As it is the case with most idiomatic expressions, they are a category three-type correspondence. A category three-type correspondence is when a language A has a feature that B either lacks or can only be rendered in terms of B’s, which operates according to different principles. This expression has no direct translation in Spanish –We can’t find a translation that conveys the same boxing imaginary–, and therefore an equivalent has to be found. Whatever it is the equivalent chosen by the translator, it will be rendered as a phrase that operates according to different principles.  

9. “the tomatoes take the center stage”

TT1: los tomates toman el protagonismo.
TT2: los tomates son los protagonistas.
TT3: los tomates toman el centro del escenario.

The verb take as used in the ST sentence is translated as tomar in TT1 and TT3. This word is relatively easy to learn, so that is why it is safe to say it’s a category-one-type correspondence. At the same time, it is worth pointing out that even though there is certain correspondence between take and tomar, the first has a much higher functional load in English than its equivalent does in Spanish.

The same thing happens with the adjective epic in epic paella. We don’t use the word epic anywhere near as much as they use it in English. An epic paella was translated as una paella épica in TT3, but in the TT1 it was translated as una estupenda paella and in the TT2 as una colosal paella.

And another case of the same principle is the sentence modifier today. We could use hoy as a sentence modifier (though hoy en día would be more common), but we don’t use as often as we would use actualmente o en la actualidad.

I wanted to make a quick mention about the translation of unpalatable tomatoes. I translated it as tomates de mal sabor and in TT3 it was translated as tomates incomibles. However, I believe the translation in TT2 which is tomates no aptos para el consume is fundamentally wrong. A lot of people got confused over the difference between no comestible and incomible.

10. “with the firing of a water cannon, the main event begins

TT1: con el estallido de un cañón de agua, se da comienzo al evento principal.
TT2: con el disparo de un cañón de agua, comienza el evento principal.
TT3: con el tiro de un cañón de agua, el evento principal empieza.

Going back to Halliday's suggestion that any single sentence will always correspond on a one-to-one basis with any single sentence in another language, there's in this case a total one-to-one correspondence on the sentence rank between the ST sentence and the three TT sentences.

Holliday doesn't specify there has to be this same correspondence on the level of the phrase unit, but the three translations begin with the same syntactic configuration: a sentence modifier adverbial.

The most notable difference between these translations is than following the adverbial, TT1 makes use of the passive voice with “se” while the other two use the active voice. Therefore, it can be said that there’s not one-to-one correspondence on the phrase rank between the ST sentence and the TT1 sentence.

Because of this syntactic difference, TT1 is a semantic translation –it attempts to sound more natural to the TT reader–, and TT1 and TT2 are faithful translations.

As an aside note, I have seen that among the 20 best translations from the contest, only two of them used the passive voice while the other 18 used the active voice. This might be an indicator that it’s better to use the active voice in this kind of construction. Moreover, the main distinct feature between TT2 and TT3 is that in one the verb precedes the subject, and in the other the subject precedes the verb. This marks whether the translator places more emphasis on the action or on the subject. Again, out of the 20 best translations, 15 of them chose to put the verb first, and the other 5 did it the other way around.
verb + subject
subject + verb
·         empieza entonces el acontecimiento principal
·         comienza el evento principal
·         marca el comienzo del evento principal
·         empieza la actividad principal
·         inicia el acto principal
·         el evento principal comienza
·         El evento principal se inicia
·         el evento principal da inicio
·         El espectáculo principal empieza luego
·         el evento principal inicia

No comments:

Post a Comment