Monday, January 29, 2007
Unsuppressed bitterness
EUObserver, like EurActiv, is a site dedicated to EU news stories. And like EurActiv, the Economist, and so many others it too seems incapable of getting the facts straight on the EU's working languages (see recent posts):
Because what is meant (Italian original: non posso tacere la mia amarezza) is not of course bitterness - that most unstatesmanlike of sentiments – but disappointment, regret etc.
As in the BBC report of the Zinedine Zidane interview last summer (earlier post), this ‘literal’ translation strikes the wrong note and misrepresents the speaker.
I am surprised to see however that the practice of literal translation is not without its eminent defenders.
In his otherwise excellent book on translating the French legal system, Martin Weston, following Peter Newmark apparently, identifies literal or word-for-word translation as the basic method of translation. Newmark himself is quoted as saying:
How can word for word (a.k.a. ‘formal lexical equivalence’) be described as a method, anyway? In fact, what exactly does it mean? How do you know which (target) word to use for any given (source) word taken in isolation?
Is it the one you find in the dictionary? The one that looks most like it? The one that first springs to mind??
Let’s assume it’s the dictionary option, this being on the face of it less absurd than the others. If you look up amarezza in an Italian-English dictionary (e.g. the Oxford/Paravia, self-described as “the world’s most trusted dictionary”), bitterness is the first and more than likely the only translation you will find. So in that sense Frattini’s bad translator was perfectly faithful to this method while producing a perfectly unfaithful translation.
But apologists for the literal method explain that if it produces the wrong result a different and better method is then applied to get the right result. But in that case why not use that other method from the outset?
I suppose word-for-word can be understood as the basic method in the sense that if there is no alternative available it may well be better than nothing.
For example, people taking their first steps in a foreign language tend to piece together phrases consisting of word-by-word ‘translations’ from their native language. That doesn't take them very far though and they only really get going once they start building up an autonomous stock of expressions and patterns in the new language.
Similarly, generations of oppressed and apathetic schoolchildren have no doubt used the word-for-word method to grind through their thème and version exercises.
And novice translators typically resort to it before they realize that word-for-word is not how translation works - because it’s not how languages work.
Interpreters probably realize it faster. Simultaneous interpreting one word at a time can be ‘a terrible strain’…
From The Simpsons (The Itchy & Scratchy & Poochie Show):
On the website of the commission's External Relations directorate-general, an explanatory text on the March anniversary celebrations is available in the three working languages of the EU - French, English and German - as well as Spanish, but not available in ItalianThe story itself, dated 18 January 2007, is this:
Italian EU commissioner Franco Frattini has in an unusual step criticized his own institution for not translating a website promoting the 50th anniversary of the Treaty of Rome in Italian.He cannot suppress his bitterness? Well he shouldn’t have to – that’s the translator’s job, surely.
"I cannot suppress my bitterness and dissatisfaction over this decision that hits the Italian language", Mr Frattini said in a statement circulated among journalists.
Because what is meant (Italian original: non posso tacere la mia amarezza) is not of course bitterness - that most unstatesmanlike of sentiments – but disappointment, regret etc.
As in the BBC report of the Zinedine Zidane interview last summer (earlier post), this ‘literal’ translation strikes the wrong note and misrepresents the speaker.
I am surprised to see however that the practice of literal translation is not without its eminent defenders.
In his otherwise excellent book on translating the French legal system, Martin Weston, following Peter Newmark apparently, identifies literal or word-for-word translation as the basic method of translation. Newmark himself is quoted as saying:
‘a good translator abandons a literal version only when it is plainly inexact or […] badly written. A bad translator will always do his best to avoid translating word for word.’If the bad translator who translated Frattini’s 'amarezza' as 'bitterness' was doing his best to avoid translating word for word, then his best, it has to be said, was simply not good enough.
How can word for word (a.k.a. ‘formal lexical equivalence’) be described as a method, anyway? In fact, what exactly does it mean? How do you know which (target) word to use for any given (source) word taken in isolation?
Is it the one you find in the dictionary? The one that looks most like it? The one that first springs to mind??
Let’s assume it’s the dictionary option, this being on the face of it less absurd than the others. If you look up amarezza in an Italian-English dictionary (e.g. the Oxford/Paravia, self-described as “the world’s most trusted dictionary”), bitterness is the first and more than likely the only translation you will find. So in that sense Frattini’s bad translator was perfectly faithful to this method while producing a perfectly unfaithful translation.
But apologists for the literal method explain that if it produces the wrong result a different and better method is then applied to get the right result. But in that case why not use that other method from the outset?
I suppose word-for-word can be understood as the basic method in the sense that if there is no alternative available it may well be better than nothing.
For example, people taking their first steps in a foreign language tend to piece together phrases consisting of word-by-word ‘translations’ from their native language. That doesn't take them very far though and they only really get going once they start building up an autonomous stock of expressions and patterns in the new language.
Similarly, generations of oppressed and apathetic schoolchildren have no doubt used the word-for-word method to grind through their thème and version exercises.
And novice translators typically resort to it before they realize that word-for-word is not how translation works - because it’s not how languages work.
Interpreters probably realize it faster. Simultaneous interpreting one word at a time can be ‘a terrible strain’…
From The Simpsons (The Itchy & Scratchy & Poochie Show):
Homer: How'd you get to be so good?
June: Oh, just experience I suppose. I started out as Roadrunner. [as Roadrunner] Meep!
Homer: You mean "meep-meep"?
June: No, they only paid me to say it once, then they doubled it up on the soundtrack. [to herself] Cheap bastards.
[Myers gives them their scripts.]
Homer: Is this cartoon going on the air live?
June: No, Homer. Very few cartoons are broadcast live, it's a terrible strain on the animators' wrists.
Wednesday, January 03, 2007
Language Log and the EU's languages
The Language Log does a great job taking the likes of the BBC to task for their lazy and ill-informed reporting on language-related matters. It is disappointing, therefore, to find Language Log itself displaying the same lackadaisical approach to fact-finding and checking in a post entitled Happy New Year unless you're an EU bureaucrat.
For one thing, it repeats, without citing a source, the canard about the EU having three working languages only - the subject of the previous post here:
- even though a Wikipedia article that it links to correctly states the true position:
1. Conference interpreters work out of not one but several languages each. Senior staff interpreters at EU institutions typically have 4 to 6 passive languages in their combinations.
2. The use of relay, especially for the less frequently used languages (e.g. Maltese interpreted direct into English and from there into the other languages).
(Much the same applies to translation).
LL concludes from its speculations:
As observed in the last-but-one post here, this is simply not the case. The EU's language regime may not work in LL's theory, but it certainly works in everyday practice.
On the question of expense, LL has a link to a second Wikipedia article according to which:
Who writes this stuff, I wonder. Both these figures are overstated by a factor of about 10, by my reckoning. They appear in fact to be aggregates for translators and interpreters across all the institutions.
Anyway, the overall cost of translation and interpretation services at the various EU institutions comes to around 1% of the EU budget and represents something like the price of a cup of coffee per citizen per annum.
This is regarded as an acceptable price to pay given the importance of direct access to laws and law-making and equal participation and representation in the democratic process.
A couple of facts to close with, taken from a European Commission press release.
On translation:
On interpretation:
For one thing, it repeats, without citing a source, the canard about the EU having three working languages only - the subject of the previous post here:
Now for the quiz answer: here is the full list of 23 official languages of the EU as of today (notice that of these, only English, French, and German have the status of working languages).
- even though a Wikipedia article that it links to correctly states the true position:
All languages of the EU are also working languagesLL also calculates, without for some reason attempting to find out what the actual situation is, that it takes 506 (or alternatively 256) interpreters to handle the full complement of languages. The true figure is in fact about 70. This sizeable discrepancy between LL's conjecture and the empirical reality can be explained by two factors not taken into account by LL:
1. Conference interpreters work out of not one but several languages each. Senior staff interpreters at EU institutions typically have 4 to 6 passive languages in their combinations.
2. The use of relay, especially for the less frequently used languages (e.g. Maltese interpreted direct into English and from there into the other languages).
(Much the same applies to translation).
LL concludes from its speculations:
By any of these measures, it is clear that the EU needs a massively expensive interpreting and translating bureaucracy, probably more cumbersome and expensive than can possibly be afforded.
As observed in the last-but-one post here, this is simply not the case. The EU's language regime may not work in LL's theory, but it certainly works in everyday practice.
On the question of expense, LL has a link to a second Wikipedia article according to which:
The European Parliament employs over 4,000 interpreters at an annual cost in excess of 1 billion euro.
Who writes this stuff, I wonder. Both these figures are overstated by a factor of about 10, by my reckoning. They appear in fact to be aggregates for translators and interpreters across all the institutions.
Anyway, the overall cost of translation and interpretation services at the various EU institutions comes to around 1% of the EU budget and represents something like the price of a cup of coffee per citizen per annum.
This is regarded as an acceptable price to pay given the importance of direct access to laws and law-making and equal participation and representation in the democratic process.
A couple of facts to close with, taken from a European Commission press release.
On translation:
For 2007, the cost of translation in the Commission is estimated to be around EUR 302 million. This represents a cost to each citizen of around EUR 0.63 per year. The introduction of three new official languages will not increase the cost for the public thanks to the new translation strategy. In 2006, the overall cost of translation in all EU institutions is estimated at EUR 800 million.
On interpretation:
The total cost of interpretation in the EU institutions was almost EUR 190 million in 2005, the equivalent of EUR 0.42 per citizen per year. The cost of interpretation per citizen is not expected to rise with the accession of Bulgaria and Romania and the addition of Irish.
