The book, Translation and Translating: Theory and Practice, derives from a feeling of considerable unease and puzzlement about the way translation has been treated, over a substantial period, by translation theorists on the one hand and linguists on the other. The translation theorists, almost without exception, have made little systematic use of the techniques and insights of contemporary linguistics (the linguistics of the last twenty years or so) and the linguists, for their part, have been at best neutral and at worst actually hostile to the notion of a theory of translation.
This state of affairs seems particularly paradoxical when one recognizes the stated goal of translation: the transformation of a text originally in one language into an equivalent text in a different language retaining, as far as is possible, the content of the message and the formal features and functional roles of the original text. It does seem strange that such a process should, apparently, be of no interest to linguistics, since the explanation of the phenomenon would present an enormous challenge to linguistic theories and provide an ideal testing ground for them.
Equally, it is difficult to see how translation theorists can move beyond the subjective and normative evaluation of texts without drawing heavily on linguistics. The need for access to and familiarity with the accumulated knowledge about the nature and function of language and the methodology of linguistic enquiry must become more and more pressing and less and less deniable if translation theory is to shake off individualist anecdotalism and the tendency to issue arbitrary lists of 'rules' for the creation of 'correct' translations and set about providing systematic and objective descriptions of the process of translation.
The essential argument of this book rests on the following assumptions:
(a) that the paradox we have been describing has arisen as a result of a fundamental misunderstanding, by both translation theorists and linguists, of what is involved in translation;
(b) that this misunderstanding has led, inevitably, to the failure to build a theory of translation which is at all satisfactory in a theoretical or an applied sense;
(c) that the co-occurrence of exciting advances in cognitive science, artificial intelligence and text-linguistics with the emergence of a genuinely socially and semantically based functional theory of linguistics - Systemic linguistics - makes this an ideal moment to attempt to resolve the paradox and develop an adequate theory of translation.
In 1960 Halliday wrote a paper on linguistics and machine translation, in which he made the remark:
It might be of interest to set up a linguistic model of the translation process, starting not from any preconceived notions from outside the field of language study, but on the basis of linguistic concepts such as are relevant to the description of languages as modes of activity in their own right.
It is precisely this task that the author, Roger T. Bell, has set himself: to model the process of translating, setting it particularly within a Systemic model of language.