This paper begins with a description of the origins of the speech act theory, and the classifications of speech acts. Then, it reviews different camps of thought which had a bearing on our current understanding of speech acts, and specifically focuses on Halliday’s metafunctions, Isaacs and Clark’s ostensible and genuine acts, Sperber and Wilson’s relevance theory, Mey’s conception of pragmemes, and Wittgenstein’s concept of language games. These are then put together in a general discussion to present a view of speech acts seen as language micro- and macro-games. It is suggested that any act of language use (be it semiotic, kinesthetic, proxemic, verbal, orthographical, or otherwise) is essentially a language micro- or macro-game. Finally, a model of ‘language game constellations’ is described which comprises language game sets, hierarchies, chains, and networks.