Tomedes Blog

The practice of language translation is becoming increasingly popular as more companies discover the benefits of reaching out to potential customers in their native languages. But many companies find it difficult to locate a language translation service which is most suited to their needs. This is because we don’t have any tool yet which would help us measure or get an idea about the effectiveness of their services beforehand. However one can certainly employ specific strategies based upon language translation requirements.

It is important to get a clear idea about one’s own language translation requirements; this would include the volume of required translation, frequency of translation and the number of languages. If your translation requirements are less, in the sense, you need it for only one language pair and the frequency is low, then you can try out a reputed single language translation service provider. You can check them out by getting a translation test of a few hundred words and getting it proofread by another translator.

Complications arise when one is looking for more in terms of language pairs, amount of work and frequency of requirements. In such cases it is important to take stock of the language translation provider’s abilities to scale up production according to requirement, the existence of documented repeatable processes that would ensure consistent delivery of high quality work and the technological tools used by the vendor.

Of course all this requires some basic knowledge about the processes involved in language translation. Some time spent on the Internet would help one achieve that and anyway it’s a small effort considering the stakes involved especially in the case of high volume and critical translation projects.

Back Translation is the process of retranslating a document from the target language to source language, to verify how accurate the process of translation has been. The back translation is usually is carried out by a translator other than the original translator of the document, whose expertise is atleast on par with the translator of the original document.

Back translation is commonly used by organizations that are heavily involved in research and product testing; typically these are biotech companies, pharmaceutical companies and survey organizations and the like. Such companies frequently require to localize their content for their target users. However the risk of losing/mutilating information in translation is ever present. A literal translation never works, because there are many factors that need to be accommodated in the process of translation, for example an original document may have words for which there are no literal equivalents in the target language, sometimes it may happen that a so-called literal equivalent might have a completely different meaning/effect than originally intended.

This is where the experience and personal judgment of a translator comes into play where he/she fashions a translation to make it as authentic and consistent with the original as possible. However this also means that when the document is back translated it would not match with the original. But that does not mean that the original translation was inaccurate. This is why back translation is not always reliable especially for verifications between source and target languages which have significantly different characteristics.

So are there alternative ways of verifying language translation? The answer is yes. We need to understand first that back translation as a method to verify translations was started long back. We have come a long way since then and it’s prudent to have an effective translation strategy in order to ensure accurate and high quality translation.

• Firstly, all translations should be managed effectively right from the beginning i.e. the creation of the original content. Any errors in the source content can cause heavy losses further down the localization process.

• Use of translation memories is fairly common nowadays, they are an asset and should be managed intelligently. The organization should develop a proper strategy for managing translation memory which should include collection, maintenance and scrubbing. Apart from translation memories, it’s also important to maintain terminology guides for various target languages.

• A quality control process should be established to ensure that all the steps required for the translation management strategy are properly implemented, including the selection of translators who are not only skilled in the required languages and knowledge areas, but also have a solid understanding of the target market.

Incorporating all the above elements effectively into a translation strategy would ensure much better results than any amount of back translations.