The translation of idioms takes us a stage further in considering the question of meaning and translation, for idioms, like puns, are culture bound. The Italian idiom menare il can per l’aia provides a good example of the kind of shift that takes place in the translation process. Translated literally, the sentence

Giovanni sta menando il can per I'aia.

becomes

John is leading his dog around the threshing floor.

The image conjured up by this sentence is somewhat startling and, unless the context referred quite specifically to such a location, the sentence would seem obscure and virtually meaningless. The English idiom that most closely corresponds to the Italian is to beat about the bush, also obscure unless used idiomatically, and hence the sentence correctly translated becomes John is beating about the bush.

Both English and Italian have corresponding idiomatic expressions that render the idea of prevarication, and so in the process of interlingual translation one idiom is substituted for another. That substitution is made not on the basis of the linguistic elements in the phrase, nor on the basis of a corresponding or similar image contained in the phrase, but on the function of the idiom. The SL phrase is replaced by a TL phrase that serves the same purpose in the TL culture, and the process here involves the substitution of SL sign for TL sign. Dagut’s remarks about the problems of translating metaphor are interesting when applied also to the problem of tackling idioms:

Since a metaphor in the SL is, by definition, a new piece of performance, a semantic novelty, it can clearly have no existing‘equivalence’in the TL: what is unique can have no counterpart. Here the translator's bilingual competence—‘le sens’, as Mallarmé put it‘de ce qui est dans la langue et de ce qui n'en est pas’—is of help to him only in the negative sense of telling him that any‘equivalence’in this case cannot be‘found’but will have to be‘created’. The crucial question that arises is thus whether a metaphor can, strictly speaking, be translated as such, or whether it can only be‘reproduced’ in some way.

But Dagut's distinction between‘translation’and‘reproduction’,like Catford's distinction between‘literal’and‘free’translation does not take into account the view that sees translation as semiotic transformation. In his definition of translation equivalence, Popovič distinguishes four types:

(1) Linguistic equivalence, where there is homogeneity on the linguistic level of both SL and TL texts, i.e. word for word translation.

(2) Paradigmatic equivalence, where there is equivalence of‘the elements of a paradigmatic expressive axis’, i.e. elements of grammar, which Popovič sees as being a higher category than lexical equivalence.

(3) Stylistic (translational) equivalence, where there is‘functional equivalence of elements in both original and translation aiming at an expressive identity with an invariant of identical meaning’.

(4) Textual (syntagmatic) equivalence, where there is equivalence of the syntagmatic structuring of a text, i.e. equivalence of form and shape.

The case of the translation of the Italian idiom, therefore, involves the determining of stylistic equivalence which results in the substitution of the SL idiom by an idiom with an equivalent function in the TL.

Translation involves far more than replacement of lexical and grammatical items between languages and, as can be seen in the translation of idioms and metaphors, the process may involve discarding the basic linguistic elements of the SL text so as to achieve Popovič’s goal of ‘expressive identity’ between the SL and TL texts. But once the translator moves away from close linguistic equivalence, the problems of determining the exact nature of the level of equivalence aimed for begin to emerge.

Albrecht Neubert, whose work on translation is unfortunately not available to English readers, distinguishes between the study of translation as a process and as a product. He states bluntly that: ‘the “missing link” between both components of a complete theory of translations appears to be the theory of equivalence relations that can be conceived for both the dynamic and the static model.’The problem of equivalence, a much-used and abused term in Translation Studies, is of central importance, and although Neubert is right when he stresses the need for a theory of equivalence relations, Raymond van den Broeck is also right when he challenges the excessive use of the term in Translation Studies and claims that the precise definition of equivalence in mathematics is a serious obstacle to its use in translation theory.