For professional translations, visit timtranslates.com.

One of the most useful Internet tools for translators translating into English, and indeed any English writers, is the (link is no longer accessible) British National Corpus. The corpus is a collection of modern British English, containing over a million words. It consists of 90% written texts and 10% spoken texts.

Anyway, to give you a real example of how I used it, I was just revising a sentence in which I said “on the contrary”, and I was unsure what the best punctuation was to go after it: a colon, a comma or a full stop. I searched in the corpus, and as you will see if you click on the above link and do the same search, by far the most used punctuation mark after “on the contrary” is the comma.

For professional translations, visit timtranslates.com.

I left my mobile charger at my parents’ house at Christmas. No problem, I thought, as I’ll see them at Easter and until then I can just keep borrowing my flatmate’s charger. But yesterday he went off to Mallorca for a week, I forgot to charge it before he went, and I had no power left.

I got home to find a Skype message offering me a 60,000-word translation job. Of course, she had tried phoning me on my mobile before sending me the Skype message. I phoned her back, but since the Skype message had been sent several hours earlier, it was no surprise to hear that she’d assigned it to someone else. So, by not having my mobile with me, I quite possibly missed out on a contract worth several thousand euros. Now there’s a lesson to be learnt.
Nevermind. I probably wouldn’t have had time anyway. Right, better go and buy a mobile charger…

For professional translations, visit timtranslates.com.

I was first introduced to corpus analysis when taking the Spanish-English module at the UAB as part of my degree. It proved very useful, especially for medical texts, since up to 500 texts can be batch downloaded from Medline in txt format. I don’t want to go into more detail right now, but if you want to see how corpus analysis works, you can read an article of mine on the subject.

At the time we used WordSmith tools, and at the cheap price of ₤50 (approx €75) I quickly bought myself a copy. My copy was version 4.0, while at university we were using version 3.0. The new version had the advantage of not being limited to DOS names (8 characters only) and generally had a better layout and a few interesting new functions, in particular the WebGetter. Unfortunately it also became less stable, and certain functions stopped working, like searching for words in context and bilingual text allignment.

This summer, while attending Mediterranean Editors and Translators, I came across a poster for a similar tool called AntConc. Since this is open-source software, I downloaded it and quickly tried it out. Now I’ve been giving a lecturing job, I’m particularly interested in AntConc, since students are much more likely to use a tool if it’s free. I have not looked much into it, but it does seem to be more stable than WordSmith Tools 4.0. However, it is possibly not quite so user friendly, as it does not calculate everything (concordance, collocates, etc) in one go, but rather you have to run a search separately for each tool. So for the moment I’m sticking with WordSmith, but I’ll definitely show my students AntConc as well, and encourage them to download it at home.

I have to say that this is an area that has not been exploited to its full potential. A program like this can’t be particularly hard to make, and if only someone could come up with a really excellent corpus analysis program, I’m sure it would be really successful. Or has someone already come up with one, and I’ve just not found it yet? Please let me know if you know a better tool.

For professional translations, visit timtranslates.com.

As many of my readers know, I have just taken on a lecturing job, giving me the unenviable task of trying to help foreign learners of English translate into our language. I realised just how complicated our spelling system is yesterday, when I couldn’t figure out how to spell “conniving”.
The problem is that the vowel written as [ə] by the International Phonetic Alphabet, often called a neutral vowel, that most basic of sounds that exists in most languages, can be rendered by so many different vowels. In fact, any vowel can be pronounced that way if it is on an unstressed syllable. For example:

m[a]ternity
bant[e]r
penc[i]l
c[o]nniving
s[u]ccumb

So, “conniving” could be rendered “kaniving”, “keniving”, “kiniving”, “coniving” or “cuniving”, with a single or double n, and it would be pronounced exactly the same. This was my problem. I thought the vowel was an e, meaning the word would have to commence with a k, and so I just couldn’t find out how to spell it.

I think what I’ve outlined above explains one of the easiest ways of detecting a foreign speaker of English, particularly monolingual Spaniards — and bilingual Galicians — since the neutral vowel does not exist in these languages. To them, all vowels have to be pronounced [a], [e], [i], [o] or [u] (note that the phonetic symbols are identical to how the sounds are rendered in written Spanish and Galician). When they say, for example Casablanca, they will literally say [ka-sa-blan-ka], whereas an English-speaking person would actually have to make a fair bit of effort to pronounce it that way, and would find it much easier to say [ka-sə-blan-kə].

Of course, this could get me on to why I think English-speaking people are wrong when they think that Catalan is “too hard”. The above example shows why, for Anglophones, Catalan is actually easier to pronounce than Spanish, once you’ve learnt the rules. But that deserves a completely seperate entry, and this is getting too long, so I’ll end it here.