Love letters aren’t really your standard fare when working for a language service provider. At the Translation Service however, we often have to translate personal correspondence, in the form of love letters, postcards, online chats, SMS exchanges, or even transcribed phone conversations. The vast majority of these translations are done to provide evidence of a long-standing relationship to New Zealand immigration authorities. At times, translations may be required in the context of a court case or a police investigation.

Unfortunately for us translators, rare are those who can write in the style of a John Keats writing to his beloved neighbour, or have a romantic inclination similar to that of a Napoleon Bonaparte professing his love to Joséphine. Let’s be honest: most letters we receive for translation are poorly structured and badly written. Does that necessarily mean that our translations should be poorly written and badly structured?

The answer to that question depends on the purpose of the translation. You’ll be able to decide which strategy to use, and which aspects to emphasise after completing a preliminary analysis of the context of the source text and the translation. (Our e-learning module on the basics of revision gives your some guidance on how to do that. We will soon publish a module specifically dedicated to this key step of the translation process.)

Let’s consider a hypothetical example: a man’s love letter to his girlfriend. The content indicates that he’s not inclined towards grand declarations and prefers the prosaic of everyday life, that he’s not particularly good at spelling and grammatical constructions, and takes no issue at repetitions.

Let’s imagine that he needs his letters translated for immigration purposes, in order to prove that his relationship is genuine. He might in fact get a number of them translated to show that their relationship is long-standing. Immigration authorities aren’t particularly interested in his ability to write in his mother tongue. What they want to know is if the recipient of the letters is truly his girlfriend.

As such, style is not a priority. What matters is the immigration officer’s ability to understand the meaning expressed in the letters. If some of the stylistic choices, spelling errors, or grammatical mishaps impede comprehension once translated, then you need to improve the text to convey the meaning. The only times where the translation would be incomprehensible are the times where you simply couldn’t make out what was meant in the original in the first place. The point is not to make him look good or not – it’s to make sure that the meaning is conveyed.

Now, let’s hypothesise that the translation was requested by a government agency wanting to assess the man’s literacy in his mother tongue. Our translation strategy is likely to be radically different, as the emphasis is put on his writing skills. Consequently, all errors, ungrammatical structures, repetitions etc. should be reproduced in the translation.

Both strategies are valid – they simply serve radically different purposes. The only way for us to make informed translation choices is through that preliminary step of analysing the source text, its context and that of the translation. Unfortunately we – myself included – often skip it and rush to the production step of the translation. This may result in a translation of poor quality.

Let’s take the odds out of the equation, and make it a habit to complete a preliminary analysis before starting a translation. The outcome will be all the better for it.