Last week I asked for input about where translation fits into the plain language process. Editor, translator, and plain language specialist Dominique Joseph contributed such a well-thought-out response—so long she put it on a Google Doc rather than directly in the comments—that I felt leaving it buried would not do it justice. She’s given me permission to reproduce it in full here.
Do you have more thoughts to contribute to this discussion? How does translation fit into your plain language workflow? Let’s keep the conversation going, either here or in the original post.
Your post touches on so many fascinating topics, Iva!
I’ve chosen to focus, rather quickly, on these 5 aspects:
- How language-specific is plain language?
- If the original is in plain language, will the translation be, too?
- Where does translation fit in the writing process?
- Examples: translation in plain language (reader-focused) projects
- Short conclusion: the writing process
1. How language-specific is plain language
Surprisingly, it’s really not that language-specific. Apart from a few tiny details, what applies to English also applies to French (and to German, and Spanish, etc.).
An important distinction: I’m not talking here from a “narrow” plain language perspective, which focuses only on words and sentences. Instead, I’m talking from the wider (I’m tempted to say, more modern) “big plain language” or clear communication perspective.
It’s bigger than just words and sentences. The key elements:
- focusing on the reader, the reader’s needs, the purpose of the document and the context of use (to produce a reader-focused, usable document);
- then, deciding what to say, how to organize it, how to say it, how to present it visually.
Almost all of this applies to all Western languages. Although some potential problems or solutions may be language-specific, mostly at the sentence and word level (think of “on” in French, or noun chains in English), these peculiarities are but a tiny, tiny bit of the whole picture.
The most important parts of clear communication apply across languages. That’s something we actually discussed a fair bit as part of the IC Clear project. We wondered whether it made sense to teach clear communication and clear writing modules in English to a multilingual public, whether the contents would actually be useful to them. And we concluded that yes, it would be relevant, it would work.
To go back to the Government of Canada guides [Plain Language: Clear and Simple and Pour un style clair et simple]… The French and English could have been almost identical. It’s just that the teams decided to work independently and each produce their own guide.
For example, the section on testing is a lot more detailed in English than in French. But it didn’t have to be that way. Also, the French team decided to talk about punctuation, but the English team didn’t. Again, the difference comes from the authors’ choices, not from differences between both languages.
Since we’re talking about plain language across multiple languages, here are a few interesting links. These two are from the European Union:
- How to write clearly (available in 23 languages; scroll down to download a PDF)
- Clear writing newsletter (from the EU Translation section)
Also interesting: a discussion we had on the LinkedIn group “Plain Language Advocates.” Lots of interesting comments!
2. If the original is in plain language, will the translation be, too?
Many people assume that if you have a plain language original, the translation will automatically be in plain language. That’s wrong (sorry guys!).
It depends a lot on the translator. If you give the same text to 5 translators, you’ll get 5 different versions, some of which will be easier to read and understand than others. Some translators—like some writers—are better at producing clear, understandable texts.
What can help: choosing the translator wisely; giving him or her good info about the context, the users, the goals, etc.
3. Where does translation fit in the writing process
That’s a huge question! First, here are a few ideas. You’ll see examples in the next section.
Instead of seeing translation as one box to fit into the existing chain, I’d like to suggest another way to look at it. (It’s something I discovered around 2001, while working at Clarica. I learned later that Michel Gauthier, from the federal government, was also a fan of that same approach.)
This idea came from examining the writing process for a typical text.
Let’s assume we’re talking about an English text which will be translated into French. Think about all the skills, knowledge and experience that often go into creating the original: you’ll have subject matter experts, writers, people with experience in the field… all working together to produce the document.
Then it’s sent to a translator. One person. Who’s typically a language expert, not a subject matter expert, not an expert in the field (never had contact with the clients or the intended audience), and probably not an expert in clear writing either. Just one person, one set of skills, one pair of eyes.
Do you see the imbalance? We end up with many steps (and people and sets of skills) coming together for the original text, and only one step for the French version.
What I’ve found works very well is to “re-balance” that process so that the French version gets as much input as the English.
Basically, we’re restarting another writing process, but focused on the French.
That means using the translation not as a final version, but as a starting point. Then you bring together your French experts, writers and communicators (subject matter experts, people with experience in the field, etc.). You look at all aspects of the text—choice of information, structure, wording, design—and see whether they’ll fit your audience and context, and what can improved. And you test your document.
It’s not just a translation. It’s a “translation + adaptation and feedback” approach.
I’ve used it with excellent results. Michel Gauthier, too.
As for deciding, from the start, to write completely different documents, it’s not usually practical. And I think it’s rarely necessary.
Yes, there are some differences between anglophone and francophone audiences; but there are also huge differences between an East Coast dentist and a West Coast fisherman, even if they’re both anglophones. Yet, they have enough in common that one document is usually enough.
About when to send a text to translation:
What often works best, I think: waiting until the original text is basically finished… but making sure it’s still possible to make changes.
That’s because the translator will often ask questions that will help you clarify the original. And if you go through a “translation + adaptation/feedback” process, then any changes made to the French could inspire changes to the English, too.
4. Examples: translation in plain language projects
Here are 3 different scenarios.
1. Customer service letters at Clarica
We applied the “translation + adaptation” process quite a few times. So when an important letter came back from translation, I would gather a few good people together, and we’d make the text more reader friendly. It worked beautifully.
Note: By the way… I’m sure this idea is making a few people cringe. We’ve all had baaad experiences with bilingual employees “improving” a translation… But it does work very well if you have strong, knowledgeable writers and experts working on the revision.
Note 2: I had management approval to rework customer service documents that way. I was in a good position to do that, since I was working at improving the quality of French in the company. (Plus, by that time, I had spent 6 years in the translation department and worked on the company’s “complex complaints” team, all valuable experience.)
2. Simplifying a huge insurance contract at Desjardins Financial Security
Four people (2 writers, 1 lawyer, 1 actuary) worked on this project on top of their other responsibilities for 2 years. They were all knew to plain language and learning as they went. They focused on rewriting the original French; and as they were learning, they would make decisions that impacted sections they had already worked on. So they had to go back again and again.
When the texts were ready for English translation, they were sent to an external translator. He worked very closely with a member of the team (one of the writers, also a translator), to create the English version.
3. Ville de Montréal, “Charter of rights and responsibilities”
Montréal is very multicultural. The charter is all about how residents can live well together, be good neighbours to each other.
If I remember well, the city first created the English and French versions of the charter.
Then, to create the other versions (Arabic, Italian, etc.), the translation team would meet with cultural groups. They would discuss the values mentioned in the charter, see how they fit with people in that community, learn about similar values in their culture, etc. These conversations helped create a text that would make sense to people in that community.
It was a collaborative process.
(I must say, the charter isn’t exactly in plain language… But the efforts the team made to talk with the audience and find out what would make sense to them, that’s very much a “plain language” approach.)
5. Short conclusion: the writing process
Something you’ve probably noticed…
Representations of the writing process tend to be all neat and orderly: first this step, then this one, then this other one…
It’s a lot messier in real life! Instead of a straight line, the real visual would have a lot of back and forth, running in circles, zigzagging, waiting, jumping back… 😉
The “translation + adaptation” I mentioned would probably look like an offshoot from the main process. And it would probably link back to the writing process of the original text at some point, when changes made to the translation inspire changes in the original.
That real picture would be rather messy… but fascinating!
(translator, clear communication specialist, fan of good processes)