Step away from the red pen…

Translation proofreading, checking, revising or whatever you chose to call it, seems to me to be one of those areas that sends some translators into a frenzy of indignation, rushing around and around in dizzying circles after their own tails. Now, there’s nothing like a frenzy of indignation to make the rest of us tune out and resolve to never again broach a subject so I, for one, was excited by the possibilities opened up to me when I heard a fresh perspective on the matter during Spencer Allman’s talk on Negotiating Translation Revision this weekend.

The intricacies of defining translation checking are frequently debated within the profession, both online and off (indeed, a more impertinent observer might say it’s been discussed ad nauseam…). This made Allman’s perspective all the more refreshing.

The basis of his talk was a quote from Brian Mossop: Do not ask whether a sentence can be improved, but whether it needs to be improved. The most groundbreaking suggestion was a point that I have very, very rarely heard proposed in this area: Once you have agreed to accept a translation checking job, start with the assumption that the first translator was as experienced, educated and competent as you.

How very refreshing.

It seems to me that working on this basis has two very useful benefits:

1. you don’t make changes unless you have a very good reason to do so
2. you take responsibility for your role the translation process. In other words, you are responsible for the part you have played in accepting the job in the first place, and the implied guarantee that you can do a good job within the client’s time, and therefore, cost constraints.

In an ideal world, of course, when it comes to translation checking, clients and translators would be always singing from the same hymn sheet. But in reality, it’s not always clear what is being asked of the checker.

==By way of a bit of background for the uninitiated: translation agencies typically contact their freelancers with a checking job, stating the number of source and target language words, a broad indication of the subject area and crucially (given this kind of work tends to be paid at an hourly rate), the number of hours they expect the job to take. Fine. But the problems start when the client and the checker just assume they have the same definition of what is meant by checking (or proofing, or revising, or whatever your preferred term). And as the occasionally perceptive Mr D says, “When you assume, you make an ASS out of U and ME”. Quite.==

So Allman proposed the following three steps to help navigate these murky waters:

1. Before accepting a checking job, always ask the following questions:

* Is the translator experienced?
* Are they a native speaker of English, or have equivalent language skills?
* Do they have domain experience?

The answers to these questions offer a quick and easy way to assess where on the checking – re-writing scale a particular job is likely to fall, the length of time it is likely to take and whether this is a job worth taking, in line with your personal job criteria. After all, there is little point in agreeing to a three hour checking job if it is more likely to involve eight hours of tortuous back translating and substantial re-writing (especially if your client is not prepared to pay for this).

2. Once a job has been accepted, use the following to establish what is required:

SAFE AREAS

* accuracy, reliability, consistency
* typos, errors, omissions
* enhancements to style

Unless instructed otherwise, he knows these are within his remit as a checker.

GREY AREAS

* terminology
* layout
* eliminating factual errors

Allman suggests that these kinds of changes are up for negotiation. Automatically assuming they are “safe” can lead to more errors being introduced. For example, maybe the client has preferred terminology, a particularly historical perspective or other style preferences that you don’t know about. Again, this stems from the assumption that your first translator has had a good reason for making their translation decisions.

3. Finally, beware of:

* under-revision: missing errors, typos or omissions. You have not been thorough enough in your checking.
* over-revision: messing about with elements that don’t need to be messed with. Resist at all costs.
* hyper-revision: making so many changes that you introduce new errors. Unforgivable.

Even if your experience of proofreading has been slightly different to Allman’s, there’s something useful to be gleaned from his suggestions. Personally, I’d include slightly different elements in my “Safe” and “Grey” categories, depending on the text type and/or client. For example, when I’m checking technical manuals, my clients specifically request that I not make changes to style without very good reason. These documents often have only minor updates from previous versions and the client is perfectly happy with the style – a checker on a mission upsets their whole document work flow process by introducing unnecessary inconsistencies between versions and throwing out their translation memory matches for future updates. These clients have usually provided me with a list of accepted terminology too, so I’m expected to ensure this is consistent and to make any changes as necessary.

I’d even go a step further and suggest another useful technique to curb the perfectly normal, but sometimes uncontrollable human instinct to meddle. Keep a separate document with a list of each change made, along a short justification – and “just because it sounds better” really doesn’t cut it. I like this simple tool because it forces me to really think about what I am doing. This means I’m clear in my own mind about the reasoning behind my decisions, which is also useful should the client comes back with a query. Sometimes, I use this document to make note of a particularly good translation technique too, and this to me is the biggest benefit of taking on a checking job. Incidentally, I rarely send this document to the client unless I’ve specifically factored additional time into the job. It’s a document for my personal use and as such, is not what I would consider “client-ready” 😉

To sum up, translation checking jobs provide freelance translators with plenty of opportunities to learn on the job, can be an excellent way to build up your reputation with translation agencies and if you’re pricing yourself correctly, can be the bread-and-butter work that pays your bills above and beyond that tricky start-up period.

Just make sure you resist the urge to use checking as a sneaky means of implying, intentionally or otherwise, that your colleagues are useless. It’s unprofessional, and does not by extension, lead clients to think that you are wonderful.

27.11.07: edited for clarity