The most important linguistic operation in translation. This involves moving away from the individual words to extract the overall meaning of the sentence and being able to express it in the best way possible in the target language.

Deverbalisation is central to the interpretive theory of translation – interpreting in order to translate – developed by Danica Seleskovitch who was General de Gaulle’s interpreter and founder of ESIT.

Areas in which professional translators work.

Some specialisms can overlap with other areas of activity: marketing translations, copywriting and SEO translation, for instance. The most important factor is the impact of the translated marketing text, regardless of the sector in which the client operates (tourism, luxury, etc.).

Language service activities

Translation: working with the written word.

Interpreting: working with the spoken word.

Translation and interpreting are two separate professions which work differently, even if the aim is the same: conveying a message from one language to another.

Unlike freelance translators and interpreters, a translation agency acts as an intermediary between the end clients and the language professionals who are assigned the tasks.

Because of this, they can very often offer a large number of languages and specialisms.

A translation that has been certified by a sworn translator as a true and accurate translation of certain types of document (legal proceedings, notarised deeds, bailiff documents, administrative documents, etc.) that may be requested by government departments in France and other countries.

A sworn translator is an expert translator attached to a specific court. You can obtain the list of sworn translators from the Courts of Appeal.

It is important to bear in mind the end use of the sworn translation. Not all sworn translators are authorised to certify documents for all purposes. 

Cabinet Perilingua does not offer this service.

Languages for which translation demand is low and language specialists are in relatively short supply. 

Note that “rare” has nothing to do with the number of people in the world who speak the language. Thus, in France, Russian and Chinese are always considered to be “rare languages” despite the percentage of the world population that speaks these languages.

The Chinese government reformed the language in 1955 with the publication of the “Chinese Character Simplification Scheme”.

The Chinese that is taught and mainly spoken in Mainland China and Singapore, in daily life at least, is simplified Chinese.

China’s special territories (Macao, Hong Kong) and Taiwan continue to use traditional Chinese (pre-reform).

So you need to know which country a Chinese translation is intended for to determine which version of the characters you require.

CAT

Computer-Assisted Translation.

This is an IT tool that allows you to compile databases – translation memories – from previous translations. 

These tools are used to save segments of documents with repeated content (a typical example might be “click on the icon/click on the button” in IT manuals). 

A CAT tool is not machine translation. It only supports the translation process by automatically searching for a relevant equivalent in a database of saved translations.

Translation obtained using algorithms that are designed to be able to find a language match. 

Google and Bing are examples of machine translation that have been used for a number of years either for personal or conversational purposes (“getting by” in a country where you don’t speak the language, for example) or by companies to get the gist of documents without using the services of a professional translator. The latter is more problematic, even if only for confidentiality reasons, because some online tools are designed to save all input in their own database.

These algorithms translate the terms based on their “face value” with no regard for the cultural context or the intended impact (innuendos, cultural references, word play, humour, etc.). So a large part of the meaning is lost.

Machine translation is also referred to as “automatic” translation, which is misleading because the tools would not work without the human intelligence and the database creation that goes on in the background.
Think of it another way. Captchas have been successful in thwarting automated spamming because a thought process is required and only humans can do this.

This is the latest version of machine translation. The algorithms use artificial intelligence, neural networks and machine learning to feed into the databases and make more sense of the context of the terms and sentences. This is also how voice recognition software works.

The language abilities of translation algorithms have improved significantly, coming close to CAT tools, but they cannot yet extract the overall picture or place part of a text in a wider context. 

The subtleties of human communication lie in what the writer intended to say (the different layers of meaning) and how this is interpreted by the readers or listeners. 

When you translate you seek out the deeper meaning of a text. This is why neural machine translation can cope with simple instructions but cannot recreate all the nuances of an editorial text, let alone advertising slogans or baselines.

It is, however, an AI application worth keeping an eye on.