In contrastive analysis, it has been important to compare items that are comparable. The contrastive analyst James writes: ‘the first thing we do is make sure that we are comparing like with like: this means that the two (or more) entities to be compared, while differing in some respect, must share certain attributes. This requirement is especially strong when we are contrasting, i.e., looking for differences—since it is only against a background of sameness that differences are significant. We shall call this sameness the constant and the differences variables’ [4,p.169]. In translation theory this factor of sameness has been referred to as equivalence or tertium comparationis [2, p.192]. Tertia comparationis can be placed at any level of textual organisation, from microlinguistic levels (i.e., phonological, lexical and syntactic levels) to macrolinguistic levels (i.e., textual). A useful guide to how contrastivists have understood the notions of equivalence is the work of T.P. Krzeszowski . He adopts a taxonomic view to this concept and considers various types of equivalence: statistical equivalence, translation equivalence, system equivalence, semanto-syntactic equivalence, rule equivalence, substantive equivalence and pragmatic equivalence. It is beyond the scope of this paper to discuss these different conceptions and how they have evolved over time [2, p.194-195; 9].