The aim of this paper is to investigate some of the problems involved in translating foregrounded lexical items within the domain of literary discourse. By way of illustration, it focuses on the semantic field of gustatory terms to be found in the Gulestan, a celebrated Persian prose text composed by Sádi in 1258. The field incorporates five basic members: — / sirin / ‘sweet’, — / sur / ‘salty’, — / tælx / ‘bitter’, — / tors / ‘sour’, — / tond / ‘hot’. These terms enter into a contrastive or oppositional relationship (explicit or implicit) with one another along the syntagmatic axis, their modes of combination being determined by contextual requirements. In addition, they are foregrounded through phonological, morphological and / or semantic motivation. Underlying this study is the assumption that the foregrounding process which the gustatory terms at issue undergo brings the translator face to face with insuperable difficulties. The reason is that three distinct, but closely interrelated, categories of equivalence (i.e. cognitive, functional and aesthetic) must be established simultaneously between the source and target language texts. Whereas the first two types can be tackled with relative success, the third one is virtually impossible to set up. This claim is corroborated by the structural analysis of the empirical data drawn from five English translations of the Persian literary work cited above.