Items in Context: Their Bearing on Translation Theory

Abstract

One essential characteristic of language is the capacity to combine a finite number of lexical items occurring in a finite number of grammatical constructions in an all but infinite number of combinations. By the shaping and reshaping of lexical items to particular contexts a certain suppleness and adequacy are achieved, whatever the apparent limitations of the lexical inventory.1 In this respect language may be considered to be a descriptive calculus of a very adequate sort and capable, for all we know, of indefinite refinement. The living interplay of lexical item and context is increasingly recognized as crucial to translation theory. Note for example the articles on machine translation by Mickelsen and Garvin in the volume For Roman Jakobson. Mickelsen mentions several times the 'role of mutual pinpointers in dealing with multiple non-grammatical and multiple grammatical meanings of individual forms'.2 He has several excellent illustrations involving such 'mutual pinpointing' in Russian. Garvin comments similarly:3 'What is found in a particular text to be translated is thus not the system-derived meaning as a whole, but that part of it which is included in the contextually and situationally derived meaning proper to the text itself.' Pertinent also in this connection are Harris's emphasis on lexical selections and Firth's similar emphasis on collocations. Note Firth's comment:4 'One of the meanings of night is its collocability with dark, and of dark, of course, its collocability with night.' This interplay of item and context in the content structure of a text may be regarded from two aspects: (1) the distribution of the same lexical item in various contextual situations within the text; and (2) the distribution of different lexical items in identical or similar contextual situations within the text. Procedurally, the first consideration entails tracing a given lexical item through the various contexts in which it occurs in a text or discourse, while the second entails tracing a series or chain of lexical items assumed to be related in the content structure of a text by virtue of their distribution in associated contexts. The first procedure is straightforward and simple; it has been applied to various bodies of discourse,