The possibility that the practical interests of those engaged in second-languagerelated activities may afford a distinctive insight into language theory has fairly recently been endorsed by Stephen Levinson, who, in his influential Presumptive Meanings, specifically commends the concern of students of translation and second language learning with ‘a great body of language lore beyond knowledge of grammar and semantics’ (2000: 23). Appropriately encouraged, I embark here on discussion of some applications of a particular area of discourse pragmatics – that of argumentation – to a field to which the dedicatee of this volume has made important contributions: the interface of humour studies and translation (Round 1995, 1996, 1998, 2005). On the assumption that linguistic theory can make useful contributions to practical activity by providing guidelines for reflexive monitoring of intuitive or ad hoc procedures, it is proposed here that greater familiarity with argumentational structures and strategies can both help identify certain ways in which verbal humour is produced and offer guidance in the search for equivalent effects. If, as someone once observed, one of the problems with jokes is that they are not taken seriously, another is certainly the fact that individual responses to humorous presentations are notoriously diverse. What amuses and provokes laughter in one person can leave another unmoved and even irritated. In terms of intentionality or function, however, it is feasible to categorize certain textgenres – such as cartoons appearing in newspapers – as expressly oriented towards the creation of humour, and it is from material of this type that virtually all the examples analysed here are drawn. One of the obstacles often faced by translators is that verbal humour does not, on the whole, travel well. When the humour is derived from linguistic forms themselves the difficulties are immediately obvious. Puns and doublemeanings, for example, often oblige the translator to opt for a ‘second best’ solution by finding a lexical item in the co‑text more amenable to translation as a polysemous term or a homonym. So called ‘garden path’ sentences (such as ‘we saw her duck’, or ‘es gato y araña’) are even more intractable, as this sort of ambivalence cannot be transferred when the different syntactical categories concerned are clearly differentiated in the morphology of the target language.