Published on LexisPSL Arbitration on 27 May 2014
Arbitration analysis: The issue of translation is seldom at the forefront of arbitration lawyers’ minds. They often treat it as a simple logistics task, such as photocopying and collation of binders, which somebody has to do, and inevitably somebody, either internally or externally, will do. Peter Nikitin of Arnold & Porter explains the importance of taking translation seriously in international disputes.
In almost any international arbitration, whether investor-state or contractual, the greatest part of the evidence presented to the tribunal consists of foreign language documents. These documents may contain text of all kinds, from highly specialised technical reports to simple business letters. Inevitably, many of them will be excerpts from relevant national legislation or court judgments.
Before these documents can be presented to the arbitral tribunal, they must be translated into the language of arbitration, usually English. The clarity and accuracy of the translation determines the probative value of the documents.
Despite this, the issue of translation is seldom at the forefront of arbitration lawyers’ minds. They often treat it as a simple logistics task like photocopying or collation of binders, which somebody has to do, and somebody, either internally or externally, inevitably will eventually take care of while they focus on seemingly more important issues.
In this context, the only question that plays a role is how much the actual translation of this or that many pages will cost in dollars and cents. The cost of bad translation for the case as a whole is almost never considered, and when it is considered, it is usually treated as an inevitable–and routinely underestimated–overhead. It is a common myth that the most trouble a bad translation can cause is a little embarrassment. It rarely crosses people’s minds that something as trivial as translation, if not properly managed, may actually lead to losing–or at least substantially complicating–their entire case.
The importance of translation
Occidental Petroleum v Ecuador (ICSID Case No. ARB/06/11)
Occidental Petroleum v. Ecuador, the largest investor-state award in history, has been severely criticised on many grounds. One ground for criticism was that the Occidental tribunal admitted and relied on poor English translations of Ecuadorian Supreme Court judgments originally rendered in Spanish. As Professor Brigitte Stern pointed out in her powerful dissenting opinion to the majority’s award, had the translations been correct, the conclusions arrived at by the majority on a key issue of Ecuadorian law relevant to quantum ‘would have been impossible to sustain’ (Occidental Petroleum v. Ecuador, Dissenting Opinion, paragraph 78). If she is right, it is quite possible that Ecuador and its counsel lost their case–and over two billion U.S. dollars–in translation.
In my ten years of work as intern, paralegal, trainee solicitor and junior associate in various dispute resolution practices, as well as a freelance multilingual translator for such practices, I have seen many awful translations that have made me laugh and cry at the same time. I am still amazed how much money is being spent on poor quality translation and how much effort is needlessly wasted on their account. I see the reason for this in a fundamental lack of understanding of both the translation process and the translation industry on the part of many arbitration lawyers. Understanding translation, including its evidentiary significance, the problems that can arise as a result of its poor quality and the ways to prevent these problems is essential for a better and more efficient prosecution of arbitral cases.
Anyone who has had to translate will know that, contrary to the assumption of many monolinguals, translation is not simply retyping text in another language; nor is speaking two languages enough to be able to competently translate from one into the other.
The point of any translation is to convey the meaning of the text to a foreign language reader. That meaning can be compared to a liquid filling an intricately shaped vessel–the source language. The job of the translator is to delicately transfer that meaning into a very differently shaped vessel–the target language–without spilling or contaminating the precious liquid in the process. If you don’t see the meaning clearly, or are unable to capture it in its entirety, you cannot translate it. Put in simple terms, you cannot translate what you do not understand. An attempt to do so will translate the words, but not their meaning: a pointless exercise.
Legal translation adds another level of complexity to this because, like languages, legal systems have many ‘false friends’, i.e. pairs of words or phrases that appear similar, but differ significantly in meaning. For instance, ‘statement of defence’ is sometimes erroneously translated as ‘counter-claim’, the word ‘security’ in legal English is not a synonym of ‘safety’, the word ‘equity’ is not a synonym of ‘fairness’, and the words ‘pleading’ or ‘motion’ may have entirely different legal meanings depending on what procedural system you are operating in (to say nothing of the different meaning these same words have when used in common parlance). So unless you are translating within a bilingual legal system like Belgium, when dealing with legal texts you will really be translating twice: from one language into another, and from one legal system into another.
One more common myth is that so-called ‘professional translators’ do not require specialist legal training to be able to translate law. People who think this forget that legal training is the only reason why they are able to read and understand legal texts whose meaning escapes the vast majority of people. In addition to proper language skills therefore, it is essential that your translator understands the legal implications of the words before him and is able to ensure that the same legal consequences follow from the words he uses in the translated text.
Occidental Petroleum v Ecuador – Dissenting opinion
Using a legally illiterate translator can have grave consequences. For instance, in Occidental v. Ecuador, the Spanish legal term ‘solemnidad,’ which means ‘legally prescribed form‘, was translated as ‘legal requirement‘ (see Occidental Petroleum v. Ecuador, Dissenting Opinion, paragraphs 93-98). The legal consequences of not respecting the prescribed form of a transaction can, of course, be very different from the legal consequences of non-compliance with more general ‘legal requirements’, such as prior government approval. According to Professor Stern, the Occidental tribunal’s holding that such legal consequences were the same in both cases–an important premise of their award–was based on a mistranslation. (see Occidental Petroleum v. Ecuador, Dissenting Opinion, paragraphs 98) This demonstrates clearly how using translators who do not properly understand the source text can distort important evidence and lead your client to losing the case.
A translator who is not proficient in the target language or legal system can cause just as much damage by distorting evidence at the point of entry into the ‘second vessel’. For example, I was once involved in the enforcement proceedings of a case where a relatively simple Russian sentence saying that the parties waived their claims against each other ‘as at’ the date of signing was translated as saying that they waived their claims ‘as from’ the date of signing. This simple language error (and one commonly made by Russian speakers) led to the client’s entire case being quite literally ‘lost in translation’.
In addition to misrepresenting your evidence, there is another, less obvious way, in which you can lose your case in translation. Here the word ‘lose’ does not necessarily relate to the final outcome of the case, but to the time and effort needed to see the case through. In other words: efficiency.
Imagine that you are fact-developing a case and you don’t speak the language of the evidence, say Russian. Your local counsel identifies a number of key documents for you and sends you translations or summaries. Your task is to determine which of these documents will eventually be submitted to the tribunal as evidence. As you are reviewing, your eyelids become heavy. Your eyes slip through the text, you are scrolling though the document mechanically. You are skipping chunks of it. Why? Because what you are reading is gibberish. Despite the fact that the words you are looking at are English words, the text is really still in Russian. You do not understand it. Eventually you decide not to refer to or submit the document because you know that your busy arbitrator will not spend time trying to decipher it. He will simply turn the tab and move to the next document in the bundle.
Now what if that document was hiding a key quote, a killer point, which if translated properly, would have saved you days and weeks of further–perhaps unsuccessful–digging? I had some experiences dealing with foreign language documents when none of them seemed to contain anything relevant until we replaced the translator. After that, it turned out that the same documents were littered with important points. Unfortunately, at that stage it was already too late to use many of them in evidence and most of them had to be abandoned in favour of more complicated and intricate arguments which, with hindsight, were entirely unnecessary.
Wasting time and money
There is a third way in which your client can suffer serious losses from bad translation. This is less about the quality of the evidence, and more about its actual financial cost.
Suppose you are engaged in the same task as before but this time you do speak the relevant language. This should enable you to spot the evidence that bad translation would otherwise obscure. But how much time will it take you to review and analyse a clearly translated document as compared to deciphering–and often re-doing–a bad translation?
None of us like to admit how inefficient they are, but I once timed myself at this task. For a professionally translated document, I need between 10 and 15 minutes per thousand words. Reviewing a badly translated document takes me between 30 and 40 minutes per thousand words. If this badly translated document contains highly relevant points, I will need to re-translate it so my arbitrator, who does not speak Russian, will see those points. That takes another hour per thousand words.
Now multiply that one hour, plus thirty minutes of preliminary review, by my billable rate. Don’t forget to add the cost of the bad translation itself (the minimum you would pay for any translation nowadays is 10 Eurocents per word, so a 100 Euros for the thousand words). The figure you come up with in my case will be 550 Euros plus VAT per 1000 words of badly translated text against 160 Euros plus VAT per 1000 words of well translated text. That represents an increase of at least 244% for my client. And, of course, there is also opportunity cost to think of: what else could have been done in that extra hour and fifteen minutes?
How to get the most from translation
To summarise, an arbitration case can get ‘lost in translation’ in at least three ways:
- time and money wasted for your client
- key evidence slipping through your fingers or past the arbitrator’s attention
- losing the entire case if bad translation ends up distorting key evidence beyond all recognition
Translation therefore deserves much more attention and respect than we are accustomed to giving it. Selection and testing of translators should be on top of our task list when a case takes off. My advice here would be to avoid ‘quote fishing’ from various language companies and going with the cheapest one. Language companies and translation agencies are intermediaries, who will employ freelancers at the cheapest possible rate, often without even verifying their credentials. Unless you have positive experience with a specific agency, it is far better and cheaper to identify an individual translator, test that translator yourself for language proficiency and understanding of the relevant legal or technical subject matter, and then use that translator throughout that case. There are many online resources where freelance translators can be contacted, and while translator selection may take a little time and effort at the outset, it will save vastly more time and effort at later stages.
The topic of translation also deserves a better place in inter-partes negotiations leading up to the first procedural order. For instance, it is worth considering whether joint appointment of a single independent translator or agency to translate both parties’ evidence is an option that may lead to significant cost saving and streamlining on the case, not least because of greater terminological consistency and avoidance of ‘massaging’ of translations by counsel, with resulting disputes.
Furthermore, there is much to be said for developing closer co-operation between the translation industry and the arbitration industry in general. For instance, major arbitral insitutions could ally with those few translation agencies that employ vetted lawyer-linguists and provide document translation as part of their administrative services. The EU employs a large body of lawyer-linguists to ensure the consistency, accuracy and clarity of European legislation. It is difficult to see why consistency, accuracy and clarity of written evidence used in arbitrations should not be equally important in dispute resolution, where parties’ rights and obligations largely depend on it.
When dealing with translators, we must always remember that they are the people who will ultimately present most of our evidence to the tribunal. Therefore it stands to reason that we should select them at least as carefully as we select our witnesses or experts. As respondent’s lawyers in Occidental v. Ecuador have probably come to realise, saving on translation is – to put it mildly – a false economy.
This talk was first delivered at the ICC YAF event on 6 May 2014.