Unit of translation
University of Artois
Professional translators instinctively think in terms of translation units when they want to assess or describe their work and it proves to be equally necessary and useful to theoreticians when they want to analyse the translator’s work. The concept of translation unit has been variously defined according to the theoretical framework within which it has been addressed. Vinay & Darbelnet devote several pages to the subject but fail to integrate it within a theory or at least a comprehensive vision of translation; their definition may seem attractive because it includes the notion of cohesion – “the smallest segment of utterance whose signs are linked in such a way that they should not be translated individually” (V&D 1995: 21) – but the trouble is that they mainly deal with lexical problems in that section and, more questionably, foster the idea that one can isolate translation units from the source text. Seleskovitch & Lederer (1984), in accordance with their theory, describe the translation unit as a unit of sense gathered from a segment of discourse (a small number of words) that will contribute to the construction of the deverbalized sense of the whole message to be re-expressed in the target language. Christiane Nord (1997) offers a functionalist approach of the matter: to the classical view of translation unit as an ‘horizontal’ segment she adds the notion of ‘vertical’ unit, which gathers a collection of elements scattered in the text and contributing to its function.