In translation, a recurring question is “What is your area of specialisation?”. The translation agencies you work with want to know, your Proz or Translator’s Cafe profile has a dedicated section for it, and your colleagues will probably be curious about it as well. As such, many translators define themselves as “technical translator”, “legal translator”, or “medical translator”, and the like. As a result, the idea that there is a difference between 'specialised' and 'general' translation has developed, with offering the former seen as more desirable a career move than the latter.

However, this has never seemed 100% correct to me. How do you unambiguously and completely define what ‘general’ versus ‘specialised’ means? And is being a “specialised translator” always better than a “general” one? In Specialization in Translation - myths and realities (Translation Journal, Vol. 16 No. 2, 2011), Charles Martin asks whether it would be better to see 'generalist' translators as those who “know their limits, don't take on work in areas they know little about and often offer a more varied background than the specialist, and above all a broader and deeper understanding of the source language, not to mention fine writing skills?”.  

Martin also questions the idea that calling yourself a specialist in legal translation necessarily has any real meaning, because of the vagueness of the term. It “would naturally include documents that are used by lawyers and judges in criminal and civil proceedings and which require a good knowledge of legal principles, systems and institutions, documents that require familiarity with a given field of law, such as commercial or intellectual property law, and also contracts and other legal instruments that may require very little or even no real knowledge of the law”. Can many ‘legal’ translators really claim they successfully specialise in all of these separate sub-fields? Can translation agencies claim to offer true specialisation in “financial translation”, “business translation”, AND “legal translation”, as many do?

Finally, Martin dispels the idea that in order to be a ‘specialist’ translator in a particular field, one must be an ‘expert’ in that field. Rather, “translators need a more basic level of knowledge that enables them to understand underlying principles, do the research necessary to figure out what they don't understand, and find the right term in the target language.” I couldn't agree more. I’m neither fully trained as a patent attorney nor as a physicist, but I have studied the fundamentals of physics and patent law sufficiently to be able to translate documents in these fields, and translate them well. In addition (and perhaps more importantly), I know exactly when my research skills will enable me to deal with a topic I’m not as familiar with, and when something goes completely above my head. Here, honesty is key...

Despite agreeing that the terms ‘specialism’ and ‘expert’ are misleading, it's nevertheless undeniable that the industry still wants to know what you ‘specialise’ in. I understand that translation buyers need to be able to make a judgement on whether or not a particular text is beyond the scope of their supplier's knowledge or not. In certain fields, including patent translation, not being certain of when to choose one term over another can have serious consequences. I would personally never trust someone who had a background in contract law to translate a patent, but I might give someone with a background in physics a chance as patent law is easier to research than physics! I therefore do think that, when talking about 'specialised' translation, it makes a lot more sense to say "I'm looking for a translator with experience in electronic engineering" than "I need a technical translator", for example. 

Lately there has been a lot of talk among freelancers on the blog and twitter spheres about rates. In relation to translation: Per hour or per word? Source (original) language or target (translated) language? To offer volume discounts or not? More generally: To publish on your site or not to publish on your site?

The last question in particular has been extensively debated (See this recent article by the Freelancery and the response by Thoughts on Translation). I’ve thought it over too, and done the “put them up, take them down” thing several times. For now I've concluded that I’m sitting on the fence, with leanings towards the no-camp. Hence, my site doesn’t have a “rates” or “fees” tab. I do mention cost at the top of my FAQ page, where I explain that prices depend on the complexity of the text, the formatting, etc. At the moment I also have a “prices start from X” type sentence, but I’m still pondering the usefulness of it.

What might be more useful is to explain what factors influence the cost of a (human) translation. While I understand potential clients might prefer to see immediately what they can expect to pay, it simply is true that cost depends on many things, and that giving a standard ‘price per word’ isn’t particularly helpful. Not to the translator, at least - all it seems to do is leave me out of pocket when said standard rate is applied to a document that is decidedly un-standard.

So what is it, then, that makes translation jobs non-standard? When you pay for a translation, what do actually get for your money? Or, conversely, what do you need to think about before you quote your price as a translator? Thinking about that might not answer the publish/don’t publish question, but at least everyone will know what to expect (ish).

‘Reading’ the thing
In an ideal world, I’d get everything sent to me as a word file. It takes second to feed into a CAT (Computer-Assisted Translation) tool and if I need to look something up I simply copy-paste the term(s) into an online dictionary.
However, often the reality involves reams of un-editable PDFs. If I want to use my CAT tool (which I usually do) these PDFs have to be converted into readable files using OCR (optical character recognition) software. OCR processing can be time-consuming, especially with Asian characters or poor quality scans. The same goes for image files, with colour images being particularly difficult to process.
If I don’t do the OCR processing to use a CAT tool I’m slightly slower too, so either way: un-editable file = extra time = more expensive. 

Research
Most translators specialise in a limited number of fields. The theory behind this is that the best translations are produced by those who have an understanding of the broader context of the original document. Nevertheless, even so-called ‘experts’ don’t know everything about their field, particularly if that field is fast-moving. I see this with patents; a 15-year-old physics patent may be straightforward and involve only well-known technology, but a recent one could throw up concepts so new that no equivalent term has yet been thought of in my target language. Coming up with one requires significant research.
Even in less extreme situations, any translator worth their salt will have homework to do. Who is the target readership? Does the text need adapting to suit the conventions of this readership? Are there multiple ways of saying the same thing in the target language, and if so, which is most commonly used whilst still conveying the original meaning? Answering these questions takes time, which should be reflected in the final price.

Format fiddling
Japanese text takes up less space than English text. As a result, I often spend hours trying to squash my translations back into their little powerpoint text boxes. Dealing with tables and graphics also takes time. Indeed, formatting the translated text as a whole to look the same as the original can be quite a job, especially if that original was not an editable file (see above). Hence, if you want to preserve formatting, you can expect to pay for more than just the word-for-word translation.

I’m sure there are other factors that influence the cost of a translation. What do you think?