I am taking a particular view of the relationship of theory to practice, one which has two aspects that should be kept separate: (i) that that relationship works best when theory describes practice, but does not dictate it and, (ii) that knowledge of theory may alert the practitioner to phenomena s/he would not otherwise have noticed, and especially to subtle aspects of the style of the source text. Stockwell (2002a:7) worries that accepting this second point carries with it the suggestion that only reading which is based on such knowledge or awareness is valid, but that is not a worry I share. Theories, approaches and views, whether one аccepts them or rejects them, are just aspects of knowledge and knowledge of one's subject is likely to have an influence on practice. While this might not be true of the general reader about whom Stockwell speculates, it certainly applies to a professional translator: there is no reason to avoid gaining knowledge of theory, and it is likely to be a useful tool. 

I hope to reflect these two aspects of the relationship by, on the one hand, examining other people's translations and offering suggestions as to how they can be understood from a stylistic point of view and, on the other, giving examples from my own translations, which, to a greater or lesser extent, were informed by awareness of many of the issues. In particular, those issues will shape the discussion. To recap briefly: these issues are to do with how we read and how we translate what goes beyond the actual words on the page, and how we ensure literary translations preserve the mind-altering qualities of the original. I shall also, where appropriate, refer to the interplay of universal and specific and the difference between literary and non-literary texts. And we will need to keep at the back of our minds, while considering the examples here, the notion that translators have to pretend (based on evidence) to know what the original author meant.

In discussing both reading of the source text for translation and reading of the target text as a translation, the underlying assumption is that a stylistically-aware analysis can help explain how readings are arrived at more easily than if analysis is impressionistic. This view of stylistic analysis does not of itself suggest it is an advantage for an actual translator to read the source text in this way; it merely suggests that stylistic awareness makes it easier to present and discuss examples. But by making the claim in (ii), I am in fact unequivocally arguing for the advantage to the translator of at least a stylistic sensitivity. And I am making an even stronger claim than this: I am suggesting that literary translation is, in a very basic and important sense, the translation of style, because style conveys attitude and not just information, because style is the expression of mind, and literature is a reflection of mind. This applies much less to non-literary texts (except journalism), because they are not primarily reflections of mind,though there is certainly a sense in which the influence of mind, attitude and point of view of the originator of a non-literaгy text will always play a role; this is something which Critical Discourse Analysis, in particular, aims to bring to light. Stylistically-aware reading is, then, the only possible way to read a literary (source or target) text as a literary text. In order to set out the parameters of stylistically-aware reading for translation, I want at this point to revisit the six issues in cognitive stylistics would be likely to prove relevant for translation. We can now attempt a fuller view of these issues, to be examined: