In a previous post, I discussed the various translation services that a company might purchase. In that post, I mentioned back translation, which is the process of translating a target text back into the source language in order to verify that it was translated correctly. In this post, I would like to offer a bit more information about the process and a real-world example where back translation ensured that a client’s message was properly conveyed.
The best article I have ever read on back translation was published in the ATA Chronicle nearly a decade ago (August 2008). You may view a copy of that article here. In many instances, back translation can be an incredibly useful verification step. The article talks specifically about insurance, where things like the list of covered items should be back translated to ensure that nothing was missed during the translation process. This could save the company a significant amount of money by catching a mistake that could force it to pay out for an item that it normally would not cover.
Other areas can benefit from back translations, too. One is surveys (marketing, customer feedback, medical, etc.) where the data gathered will be used to make strategic business decisions or decisions about patient care. Companies will want to make sure they are asking the right questions the right way in the target language before spending what might be millions of dollars to deploy the survey, analyze the findings, and then use those findings. Yet another may be the user interface of a new software or online platform, particularly one that gives directions that users must follow.
Whatever the reason, back translation can save a lot of money and headaches for clients who are staking a lot on a certain translation. If an error is found, in most cases the back translation will largely pay for itself.
A word of caution to clients who order a back translation: because they are meant to find ambiguities and errors in wording and terminology, back translations will not be word-for-word recreations of the original source texts and may read in a slightly awkward way. If there is any confusion, a reconciliation should be done to analyze any significant differences and adjust both the original source and target texts as necessary. The ATA article recommends that clients and the translators performing the reconciliation focus on “differences that matter,” such as incorrectly translated terminology or ambiguities, and not “differences that do not matter,” such as the difference between “doctor” and “physician,” which have identical meanings. This is a good rule of thumb to go by.
So how does back translation work in practice?
Last year, I was contracted to perform a back translation for a Jewish organization of materials that had been translated into Chinese. The majority of the materials focused on the history of Judaism, Jewish customs, Jewish religious texts, holidays, traditions, etc. One particular passage had been translated like this:
|摩西所著的这五部书与口授非书面律法及解释统称为《旧约》，是“教谕”的意思。||These five books that Moses wrote and the oral, unwritten laws and explanations are collectively known as the Old Testament, which means “teachings.”|
The Chinese translator made a critical error here and translated the original word Torah, which does mean teachings, as the Old Testament (旧约). This is problematic for several reasons, least of all because Jews do not use the term Old Testament to describe their religious texts. Whether due to a misunderstanding, lack of research, or some other reason, the translator and editor made a mistake. Ironically, the most popular online Chinese encyclopedia states that 犹太人不称旧约 (Jewish people do not say Old Testament). The term the translator should have used was 妥拉 or 托拉, which are accepted Chinese transliterations of the word Torah. By doing the back translation, the client was able to find the mistake and correct it.
Do you have any entertaining or interesting back translation stories? Share them in the comments or on Twitter by tweeting me at @Bentranslates.