How to do things with (recorded) words.

The aim of this paper is to evaluate which context determines the illocutionary force of written or recorded utterances-those involved in written texts, films and images, conceived as recordings that can be seen or heard in different occasions. More precisely, my paper deals with the 'metaphysical' or constitutive role of context-as opposed to its epistemic or evidential role: my goal is to determine which context is semantically relevant in order to fix the illocutionary force of a speech act, as distinct from the information the addressee uses to ascertain the semantically relevant context. In particular I will try to assess two different perspectives on this problem, a Conventionalist Perspective and an Intentionalist Perspective. Drawing on the literature on indexicals in written texts and recorded messages, I will argue in favor of the Intentionalist Perspective: the relevant context is the one intended by the speaker. Bringing intentions into the picture, however, requires qualification; in particular, I will distinguish my Weak Intentionalist proposal from a Strong Intentionalist one. I will show that the Weak Intentionalist Perspective is flexible enough to deal with cases of delayed communication, but not so unrestricted as to yield counter-intuitive consequences. [ABSTRACT FROM AUTHOR]