Interpreting between SVO and SOV languages

I don't hear a lot of people talking about this, but there is definitely something to be said about the true wonder that is the ability to translate from SVO (subject-verb-object) into SOV (subject-object-verb) languages, and vice-versa.

Let's start easy and take languages from the same family: English, German, Dutch, Swedish and so on. For the sake of clarity while trying to reach a broader audience, let's pick the most common ones, English and German.

Germanic Languages

It is well known that word order can easily differ between the two of them. Let's take, for example, the following sentence in German:

Kennedy hat einen Apfel gegessen.

This literaly translates as "Kennedy has an Apple eaten" -- which, rendered in proper English, would be "Kennedy has eaten an apple". Notice the change in word order?

Now suppose you're hired to do a simultaneous interpretation from German into English. As your client speaks, your task is to translate orally, at the same time, what's being said (remember I'm talking about simultaneous translation, not consecutive translation). The problem is, if you have only heard the sentence up to the point, say, "Kennedy has an apple" in German (and the whole sentence is "Kennedy has an apple eaten"), how can you switch that to English "Kennedy has eaten [an apple]"? In other words, given that word order is different between these two languages, how can you use a verb that hasn't even been uttered yet? Guesswork?

Another example: one of the recurring elements in the German language is the penchant for placing the negative particle "nicht" precisely at the end of a sentence. Now, imagine you have just interpreted (from German into English) a sentence like "new market demands are to be met the upcoming semester with reasonable expectations by vigor-powered industry tycoons" -- and all of a sudden you find yourself face-to-face with the word "nicht" right at the end of the German sentence. This one word has the power to actually negate everything you just said. How do you fix that, given that English should have placed a negative particle right at the beginning (new market demands are not to be met...)?

Think that's tough? Not quite yet. 

Altaic Languages

Let's up the stakes. Leaving languages of the same family behind, let's get into relatively uncharted territory: Altaic languages. How can one successfully interpret, say, something from Turkish, or Mongolian, into English?

Tim Drayton, a cypriot translator working with Turkish and English, gives us a neat example:

Şirket fabrıkayı kapattı.

This translates as (the company) + (the factory + [glide y] + [accusative ı]) + (close + [third person past tense tı] ). I.e.: "the company closed the factory". See below the Turkish word order:

[Şirket] [fabrıkayı] [kapattı].
[The company] [the factory] [closed].

"Here a simultaneous interpreter would have to retain the factory in his/her mind", says Drayton, "and first translate the verb closed before adding the object".

In Turkish the placement of the verb at the end of the sentence is an "iron rule", to use Drayton's own term. And whereas in a sentence this simplistic it would be easy for anyone to retain just one or two words, what happens when the sentence gets bigger? Let's add geçen hafta ("last week") to the sentence:

Şirket geçen hafta fabrikayı kapattı.
(The company closed the factory last week.)

"Now the interpreter has to retain last week the factory", Drayton rightfully points out, "before hearing the main verb and then slightly alter the word order in the predicate to boot."

We can see how this might pose a problem. The larger the sentence, the more the interpreter will have to retain before using verbs. That would be fine otherwise, except verbs are generally the very linchpin of language. So then, for example, if we imagine the gap between subject-object and verb is not "the factory" or even "the factory last week", but instead "...the factory last week following an intensive campaign led by local union officials that involved a sit-down protest in the main square", how can the interpreter -- in fact, anyone -- possibly retain so much before a verb comes along? Especially when SVO languages such as English demand that a verb be used at the beginning, rather than at the end of a sentence? "Even if they do", Drayton says, "surely in a sense what is happening is no longer simultaneous."

Mongolian interpreter Tsogt Gombosuren agrees that working with languages of different word order poses a problem sometimes unsurmountable. "When I am asked by clients to provide simultaneous interpreting service", explains Gombosuren, "I often tell them that there is no such thing as simultaneous interpreting between English and Mongolian, and I can do only consecutive interpreting if they want me to be accurate."

While that would probably be the best, or easiest, choice, we know for a fact that simultaneous interpreting does happen. And the whole point of this article is precisely this: if they are happening, how are they being carried out? What are interpreters doing in order to avoid the immense pitfalls surrounding this line of work?

How are they doing it?

Swiss translator Jonathan Sanders gives us a few insights into the way to solve the problem.

"A key element is anticipating what is going to be said. Not so much making a subject up, but if you hear someone say Jonathan Sanders I..., you assume they are introducing themselves and say I am Jonathan Sanders, for example."

True, but not all sentences are that short or simple. What if you have a really long sentence like the ones I used here?

"When a nicht comes at the end of a [German] sentence that they don't expect, [interpreters can] say something like or rather, it would be if that were true or Or is it? Actually no, or (as somone jokingly said) Like hell it is!"

As we can see, one of the options is to anticipate what the speaker is going to say, promptly correcting your sentence if it turns out you were wrong. However, it's not difficult to see how this might be problematic. What manner of interpreting is this that every sentence needs correction? Even half the number would be a potential invitation to disaster. Guesswork is risky business, and it's very likely you might put yourself in a corner with no way out at some point. As a result, listeners would at least wonder why someone so "unprepared" (for that is what they'd think) had to be employed for the job. "Where did they find this guy?", I can already hear them ask.

A better way to do it is trying to wait as much as possible, so as to stay not only a few words, but a few sentences behind what the speaker is actually saying, a technique called decalage. This can be a very difficult process, seeing as it involves excellent short term memory. But that's not all.

"Of course this is very very difficult if the speaker is talking very slow", says Australian interpreter Roberta Bazzoni. "Some interpreters like to take notes (using consecutive symbols and such) if the sentences are really long and there is no hint of the verb as yet!"

"It can become a nightmare", Roberta explains, "if the speaker puts in a lot of extra information before the verb. If possibile the interpreter will try to break down the different 'blocks of information' and make several shorter sentences: sometimes the result isn't very elegant, but you might not have a choice."

It goes without saying that this is not for the feeble of heart. It demands that one's attention span and skills like short-term memory are top notch and in great shape. Apart from that, where possible, interpreters prefer to avoid simultaneous translation, choosing consecutive instead. When that is not convenient, the only way out, as usual, is for the professional interpreter to make the most he can out of his abilities and employ techniques that have been tested and proved successful in their line of work.

So that's how they're doing it.