As several studies have shown, speakers from different countries, and even speakers from different regions of the same country, ‘vary their speech act production […], convey and perceive politeness differently, and display different interactional patterns during conversational interaction (e.g. greetings, opening and closing sequences, turn-taking, laughter)’ (Félix-Brasdefer 2007).1 Sociopragmatics focuses on the relationship between linguistic structure and social action (Leech 1983, Thomas 1983), more specifically on the influence of socio-contextual factors on language as social action (Martínez-Flor and Usó-Juan 2010: 6). To achieve this goal, the study of sociopragmatic variability ought not to limit itself to the ‘where’, the ‘who’ and the ‘how’ of speech act behaviour, but also crucially tackle the ‘why’. That is, it should move beyond description and attempt to address interlocutors’ reasons for opting for particular acts and for particular realizations of those acts in specific circumstances. Understanding the emic significance of speech act behaviour is important because it cannot be taken for granted that when people from different sociocultural backgrounds engage in the same speech act behaviour, they mean the same thing by it. As numerous examples testify, the possibility of pragmatic failure (Thomas 1983) looms large, and this is especially so for speakers using the same language, as is the case with L1 and L2 speakers of a language, or speakers of different varieties of the same language.