Signalling games select Horn strategies

Abstract

In this paper I will discuss why (un) marked expressions typically get an (un)marked interpretation: Horn’s division of pragmatic labor. It is argued that it is a conventional fact that we use language this way. This convention will be explained in terms of equilibria of signalling games introduced by Lewis (1969) but now in an evolutionary setting. I will also relate this signalling game analysis with Blutner’s (2000) bi-directional optimality theory and with Parikh’s (1991, 2000) game-theoretical analysis of successful communication.