Many writers have argued that dialogue should be regarded as a joint a
ctivity (see for example (Clark and Wilkes-Gibbs, 1986; Grosz and Sidn
er, 1990; Schegloff, 1981; Suchman, 1987)), something that agents do t
ogether, rather than simply as a product of the interaction of plan ge
nerators and recognizers working in synchrony and harmony, as plan-bas
ed theories propose. Such plan-based approaches do not explain why add
resses ask clarification questions, why they confirm, or even, why the
y do not walk away. Rather, the joint action model claims that both pa
rties to a dialogue are responsible for sustaining it. Participating i
n a dialogue requires the conversants to have at least a joint commitm
ent to understand one another. The key questions to be answered includ
e how to formalize such general commitments precisely, and to show how
they predict the fine-grained synchrony so apparent in ordinary conve
rsation. To begin to answer these questions, we sketch here how a form
al theory of joint action explains confirmations that arise in task-or
iented telephone dialogues. A more formal account is given in (Cohen a
nd Levesque, 1991a). Then we argue that extensions of this analysis to
dialogue more generally will be difficult. In particular, it will for
ce us to give up our simplistic analyses of propositional content and
literal meaning.