relationship between CA and translation is bidirectional. On the one hand, the translation of specific pieces of text may provide the data for CA, as in Gleason (1965), Krzeszowski (1990) and James (1980). On the other, CA may provide explanations of difficulties encountered in translation (eg Nida 1964; Beekman and Callow 1974; Yebra 1982; Enkwist 1978; Baker 1992). [16] as a source of data for CA is strictly unavoidable. The crucial factors here are what size of language sample has been chosen for translation, whether it is naturally occurring or fabricated for the purpose, and whether the translation is the analyst s own. Though the focus of CA may continue to shift towards pragmatics and discourse analysis, its use in translation is not inevitable. It is however unlikely that it can be dispensed with completely either in the training of translators or in the assessment of translations, even in its more traditional lexico-grammatical manifestations; Halliday (1985) notes that a discourse analysis that is not based on grammar is not analysis at all, but simply a running commentary on a text . He adds that a text is a semantic unit, not a grammatical one, meanings are realized through wordings; and without a theory of wording - that is, a grammar - there is no way of making explicit one s interpretation of the meaning of the text . Baker (1992) cites the latter comment with approval in a book that is itself an indication of the continued vitality of CA as an aid to translation.
In one respect, however, Halliday s association of wordings with grammar is too narrow. An important future function of CA is likely to be in the area of ​​collocation, where parallel concordancing based on comparable corpora permits the possibility of contrastive analysis of the collocational properties of semanyically related lexis from the source and target languages. For example, translations in six languages ​​(English, French, German, Italian, Danish and Greek) are the data for a six-way collocational and grammatical comparison making use of parallel concordancing which is currently being undertaken with Lingua funding by a number of European universities led by the university of Nancy ІІ. As noted above, much CA has arisen as a result of the needs of the language teaching profession and this project is no exception in that one of its major objectives is to provide teachers with assistance in the use of parallel concordancing in the classrooms. However, the use as data of the diverse range of translations means that the project is certain to provide valuable evidence for translators on the transferability of certain collocations and colligations from one language to another. The future of...
