As we have seen in previous chapters, there is a significant gap between what we encode grammatically and the conveyed meaning we actually intend to communicate: the total take-home message (much more, usually). Accounting for how we fill this gap is the task of pragmatic theories. All pragmatic theories assume that the bridging of this gap is achieved by inferencing, and this is why the grammar/pragmatics division of labor is the one between codes and inferences. A theory of pragmatics is then a theory about pragmatic inferencing in the service of linguistic communication. This chapter presents three such theories: Grice (1975, 1989) (5.1), neo-Gricean pragmatics (Horn, 1984; Levinson, 2000) (5.2) and Relevance Theory (Sperber and Wilson, 1986/1995) (5.3). We should note that all three theories similarly define grammar as a set of codes and pragmatics as inference. They are also all Gricean in that they view the role of pragmatic inferencing as prominent in interpreting natural language utterances. For the most part, they equally account for the same set of inferences (albeit somewhat differently), and they mostly draw the grammar/pragmatics divide along the same lines (but for differences between the theories regarding specific analyses, see Ariel, 2008: chapter 3; Carston, 1990, 2002). The goal of this chapter is to introduce these inferential pragmatics theories, so we can then assume that any use/interpretation we can explain by reference to a rational inference can be relegated to pragmatics, where it will receive a proper account.