Department of Philosophy, University of Birmingham, 29 October 2007

If children acquire social and mental understanding by engaging in joint action, what could joint action be? Not as the leading account (Bratman 1992) describes it. For on this account, engaging in joint action presupposes -- and therefore cannot plausibly facilitate the development of -- rich interpersonal understanding. This talk offers an alternative, simpler account of joint action which is compatible with the developmental hypothesis. According to this account, joint action is (i) a plural activity where the agents (ii) have largely correct and largely shared views of who is involved and (iii) can distinguish behaviours which are contributions to the activity from other actions, or are at least disposed to form views about how those engaged in the activity could or should contribute. The function of the kind of joint action targeted by this account is not to coordinate subplans or to rationalise optimal solutions to group choices but to provide immediate knowledge of what others are doing and a framework for interpreting their behaviours.