I first met Izaskun Orkwis at the 1st International Conference of Economic, Business, Financial and Institutional Translation, when my only institutional client was the OECD. Since then, she’s given me some great advice on how to get my feet in the door of international institutions.
In an article for the latest edition of the ATA Chronicle, Izaskun draws on her experience with international institutions to explain some of the quirks of working for them and how to ensure your translation is fit for purpose. The article is packed with pearls of wisdom. Here, I highlight just a few, with some additional remarks based on my own experience:
“the different language versions must match exactly, including all nuances and formal structure…replacing parentheses with commas may be fine in other contexts, but not in institutional translation”
“There will be humongous databases and parallel corpora that will be both a blessing…and a curse (sometimes you must reuse previous language, unchanged).”
Sometimes I’ve translated 3,000 word documents in 15 minutes, as nearly the entire document is a recycled version of an old one in the translation memory. But as Izaskun says, sometimes the databases are a curse. This is especially true when you find inconsistencies in the memory and you spend far more time deciding which previous version to use than you would have spent just translating it from scratch.
“the institutional translator must…lose their individual voice…the translated document, just like the original, belongs to the institution, which is its sole author”
“the institutional translator must…adopt the institution’s working methods”
The latter is particularly challenging for those of us who do short-term contracts for various institutions, as the working methods vary from one institution to another. To give just one example, the World Trade Organization uses Trados Studio (with two different workflows, depending on the type of document), whereas the United Nations Office at Geneva has its own computer-assisted translation tool that runs in the browser. Some institutions require you to upload your translation to a portal; others ask you to save it to a shared network drive.
“the institutional translator must…adhere to vetted terminology”
Izaskun goes on to say that you should consult a terminologist if you feel the need to depart from the vetted terminology. In the work I’ve done, I tend only to have to consult one of the revisers, not a terminologist.
Sometimes the solution in the terminology database is not what the more senior translation staff actually use. In such cases, if I were working at home, with my own setup, I would quickly edit the entry in my termbase, but in an institutional setting, terminology vetting is a complex process and it can take a long time for the entry to be corrected. It gets even more complicated if you suggest a change to the editorial manual. Since it is used by all the UN offices, someone in Geneva can’t just change something without consulting colleagues in New York, Nairobi and elsewhere, which is perhaps why we’re still not allowed to make country names possessive, despite the fact that every UN translator I’ve ever spoken to hates the rule!
“whenever there’s a quote from or a reference to a previous document, no matter how long or short, you must assume there exists a previous translation that needs to be found and reused”
Izaskun goes on to provide some great examples of the kind of situations where you’d have to do this kind of research. It is something I already practised in my non-institutional work – sometimes I’ve ordered a book online just to check how one sentence was translated in the past and to check the page numbers for the citation – which is perhaps why I fit in well in an institutional setting.
The article goes on to say that a fit-for-purpose translation “requires little to no intervention by the in-house team”. One way you’re expected to minimise the reviser’s workload is by adding a comment whenever you quote or refer to previous material, otherwise the reviser will be forced to perform the same research as you, wasting precious time (and costing lots of money, as the revisers are more senior staff who earn a higher salary).
“An institutional translation must be linguistically flawless.”
Izaskun explains that one of the reasons why the pass rate for the entrance exams is so low is that many translators do not have a full command of the target language, citing dangling modifiers as an example of the kind of problem that trips up English translators. I read a lot of what my colleagues write and, while it would be unfair to judge a translator based on a single mistake in a Facebook post (possibly written on a phone) or a blog post (I’m sure that, like me, my colleagues don’t read their blog posts as thoroughly as they would read their translations), I do see certain mistakes crop up regularly enough to realise that there are certain grammatical aspects of the English language that many colleagues don’t master to the level required by international institutions.
Izaskun mentions dangling modifiers. A mistake I often see is the failure to use parenthetical commas properly. Another is the hypercorrection of saying “my sister and I” instead of “my sister and me”. I find hypercorrections particularly revealing because they are not the result of the writer not having time to think about the correct expression; they occur because the writer thought about what to write but applied a rule incorrectly. You can probably get away with such things if you’re translating a tourism brochure or even a financial report for a CAC 40 company, but it will not do for an international institution.
“you are expected to raise the issue [of an obvious error] with someone in a position to contact the author for clarification and correction”
In my experience, the above depends on the type of document and who authored it. There are certain document types where we might just fix an obvious factual error without contacting the author. Also, we would normally correct a factual error that is clearly the result of a typo (e.g. 2009 instead of 2019). But there are times when contacting the author is necessary.
At the WTO, there’s usually another translator working into the other target language (French, if the source language is Spanish; Spanish if the source language is French), so if you do get feedback from the author, you should share the information with the other translator.
“not all assignments are exciting or interesting”
I can vouch for that! I once spent two and a half days translating Costa Rican tariff subheadings. If you’re not sure what tariff subheadings are, open this, then go to page 16 and read until just before you lose the will to live, probably somewhere on page 17. Then think about the fact the list goes on and on and on like an Ariston advert from the late 80s and early 90s (for readers who didn’t live in the UK back then, click here to see what I’m referring to, then feel free to send me hate mail for making you watch it). The final heading in the list, used when a country exports “Antiques of an age exceeding one hundred years”, is on page 494! And spare a thought for the Spanish translators: every time they see the word “Other”, they have to carefully check whether the item being referred to is masculine (“Los demás”) or feminine (“Las demás”). Apart from a few interesting things I learned about Costa Rican Spanish, it wasn’t the most thrilling couple of days I’ve ever spent in the office.
Much of the time, though, you’re working with interesting documents and, as Izaskun puts it, you have “daily opportunities to be on top of world affairs and to work on assignments that make a difference”. If you’ve only read this blog post and not Izaskun’s article, I urge you to read it, especially if you’re keen on working for international institutions.
Finally, I leave you with one of the other benefits of being an institutional translator: the wonderful view over Lake Geneva from an office window in the English Translation Section at the United Nations Office at Geneva. Click on the picture to see the full-definition view.