This thesis focuses on the role in communication of 'natural' signs and natural behaviours - facial expressions, spontaneous expression of emotion, interjections etc. - and considers how they might be accommodated within a pragmatic theory. Linguists generally abstract away from such behaviours and focus on the grammar, or linguistic code. However, there are two reasons why the pragmatist should cast a broader net. Firstly, it is increasingly recognised in pragmatics that verbal communication is not simply a coding-decoding process but an intelligent activity involving the expression and recognition of intentions. Secondly, sentences are rarely uttered in a behavioural vacuum; the aim of a pragmatic theory is to explain how utterances - with all their linguistic and non-linguistic properties - are understood. The analysis I propose has implications for theories of utterance interpretation. Firstly, verbal communication often involves a mixture of (what Paul Grice called) natural (N) and non-natural (NN) meaning, and there is a continuum of cases between 'showing' and meaning NN; I show that natural communicative phenomena may be located at different points along the showing-meaning NN continuum. This has clear implications for the domain of pragmatic principles or maxims, for it suggests that they are best seen as applying to the domain of intentional communication as a whole, rather than to the domain of meaning NN. Secondly, the wider question of how 'natural' communicative behaviours are to be accommodated within a pragmatic theory breaks down into several further, more specific, questions - How are natural phenomena interpreted? Do they all work in the same way? I argue that some natural behaviours have a signalling function: they are, in effect, natural codes. Such behaviours do not fit easily into Grice's distinction between natural and non-natural meaning, which suggests that it is not exhaustive, and that it does not provide the best starting point for constructing theories of communication.