What’s Best: Fluency or Accuracy?

How many times have you heard—or maybe even asked—the question “is it better to be fluent or to be accurate?”. In language learning, this is a common debate and it basically places in opposition language production that is formally correct (grammatically, syntactically, lexically, phonetically) but perhaps not very successful at negotiating social function, with language production that may well be technically at fault but is nevertheless very good at achieving its communicative purpose. The debate is an old one, and there are opinions to suit all tastes.

In the Translation Unit, of course we’re primarily concerned with the written word. So the sort of language that we produce is obviously different in character to the spoken form. But the fluency/accuracy debate is relevant to our work, too, and I’d like to consider just a few things in order to try and give some insight into this question, at least from the perspective of translation and text correction.

It sounds reasonable to say that an informal e-mail doesn’t need the same sort of accuracy as an academic paper, a contract or an instructions manual. On the one hand, we have a text that’s probably aimed at some sort of pragmatic objective (arranging to meet up; asking a quick question; inviting a friend, etc.); on the other, we have texts with on obvious need for linguistic precision. This tends to make people think that the only feasible answer to the question we’re asking here is “it depends”! If your objective is informal communication, you don’t need to worry too much about accuracy. Just let your language flow freely (which is the lexical idea that underpins the word ’fluency’). For more formal contexts, we’re likely to say that we need precise, unambiguous language, that is, ’accuracy’ (from Latin accuratus, “prepared with care”). This sounds perfectly rational, doesn’t it?

Except—unfortunately—that it’s not exactly like that. Why not? Because if we see fluency in opposition to accuracy, then logically they can never occur simultaneously. Or, to put it another way, if we insist on seeing language as a continuum, like a straight line, with accuracy at one end and fluency at the other, then absolute fluency (such as that characterising native speakers, for instance) would imply a total absence of accuracy; and absolute accuracy (such as that required by an instructions manual, for example) would imply a total absence of fluency, and nothing at all would get communicated in any meaningful way. Clearly, that’s nonsense.
Language is not a fluency-accuracy continuum; instead, it makes use of both elements simultaneously, though each in distinct proportions depending on the circumstances. A legal contract has to be highly accurate or else it will fail in its most essential purpose; but to be a good pragmatic document (one that is easy to follow, coherent, comprehensible and yet precise in its objectives) it needs a solid degree of written fluency; it needs to “read well”.

And, actually, this is true whether we’re talking about written or spoken language: genuine, effective communication depends on both fluency and accuracy. So, in a sense, the answer to our question really is that “it depends”. But if we think that, by focussing on accuracy we can move away from fluency, and vice versa, then inevitably the sort of language we’ll end up producing won’t be much use to us, or to anyone else!

Unitat de Traducció i Revisió de Textos

Deixa un comentari

L'adreça electrònica no es publicarà.

Aquest lloc utilitza Akismet per reduir els comentaris brossa. Apreneu com es processen les dades dels comentaris.