This guest post is contributed by Patricia Walling, a healthcare blogger who can be contacted at patwalling85 ~at~ gmail.com.
Technology is becoming more and more prevalent in the field of translation as computer linguists become better at programming good grammar recognition. With the advent of technologies like Phrazer poised to replace translators in a clinical setting, and CAP tools becoming a reasonable replacement for human editors, it can be quite easy to think that in the future people won't be necessary for translations.
I will probably make a lot of people happy by submitting that this could never happen. Languages are defined by their usage rather than by any algorithm, and every language is constantly evolving. Perhaps this problem could be solved if computers used in translation were made to speak regularly with human beings out loud rather than in a text based context.
Yet even then, as humans do, they would need to be edited by people who speak regularly themselves. There is a multiplicitousness in any language, which varies by age, formality, locale and a number of other variables that can only be really understood by fluency, and we know that everyone has their own perceptions of what is correct. A spoken language is not as easily standardized as procedures like in medical coding, and it is the human element that is required for good translation. Without that, we'll only continue to get embarrassments like Engrish and "spun" articles.
Of course, there are arguments against this, and many people in medical professions won't hardly care but as long as we're still arguing about how many editors a translation needs and playing games like telephone, it will be quite difficult to entirely replace human translators.
ForeignExchange's METRiQ quality system provides medical device and pharmaceutical companies with known translation quality - on every technical translation assignment.