Philosophical writing on what it is for several agents to act together—in other words, what it is for them to engage in “joint intentional action”—often start off by a short list of some paradigm cases of the phenomenon in question, such as two people walking together, two people painting a house together or carrying a sofa together. Another recurring alledged paradigm example, mentioned by Michael Bratman, Margaret Gilbert and Christopher Kutz for example, is that of two (or more) people having a conversation. Philosophers working on joint action rarely elaborate on this particular example though, it from reading this literature, it is unclear in what sense having a conversation is a joint intentional action. Furthermore, while many linguists and sociologists take conversation to be a kind of joint activity, they are rarely explicit about in what sense it is a joint activity (but there are certainly exceptions, see the work of the cognitive psychologist Herbert Clark for example). Their claim is surely not merely that it is an activity in which several agents are present.
As I’m now teaching a course entitled “Social Cognition” which largely focuses on theoretical and empirical work in pragmatics, the branch of linguistics is concerned with how speakers and addresses coordinate their social actions and interpret each other’s utterances, I’ve been thinking a bit about this mentioned but rarely discussed (by philosophers) example of joint intentional action, and thought I would share some of my (not very original) thoughts.
The example of two people walking together—made prominent in philosophical discussion after Gilbert’s paper “Walking Together: A Paradigmatic Social Phenomenon” (Midwest Studies in Philosophy XV, 1990)—can be contrasted with a behaviourally similar case of two people walking in parallel down a street, each trying to keep an appropriate distance to each other in order not to collide as well as in order not to violate social norms of appropriateness. Roughly speaking, what seems to distinguish the joint from the parallel case is that in the former, the walkers have a common goal—that they walk—around which their actions are coordinated. There is at most a common type of goal in the parallel case—each might intend to walk to the tube station, say.
Is there a similar contrast case for conversation? Consider a brief exchange between two strangers S and H: S asks H, “Do you know the time?”, and after a brief pause H responds, “No, I don’t have my mobile phone on me,” and S might then in turn respond, “Ok, no worries.” At first blush, there seems to be no common goal here. S’s goal is to be informed of the time and H’s goal might just be continue doing whatever it is that he is doing when interrupted (perhaps he in fact has his mobile phone with him, but he cannot be bothered to pick it out of his pocket, so he lies). Just like in the case of the strangers walking in parallel, there is nevertheless much coordination with respect to various social norms though. H responds to S’s request for information with a relevant answer, and he signals that he will have to decline the request by pausing briefly before answering (it is a widespread norm that declining responses to requests are prefaced by brief pauses and followed by an explanation of why the request cannot be accepted). Now, we could contrast this with the case where S and H are close friend who agree to meet a café, each with the intention that they converse together and hang out. The former case is like the strangers walking in parallel, while the latter seems to be like the case of two friends walking together.
But the first case is arguably nevertheless a case of two people conversing with each other (even if very briefly)! This seems to suggest that, contrary to what some claim, conversing together isn’t, not as such at least, a joint intentional activity. Indeed, it suggests that conversing is a useful way of achieving coordination of activities, whether or not these are parts of one and the same joint intentional activity or not. If this were the case, then it would not be crucial to understand joint intentional action in order to understand the nature of conversation as such. After all, to understand conversation, we could also study conversation in cases where they occur in the context of a “parallel activity”.
However, if we again consider S and H’s brief exchange about the time a bit closer, we will see that there is actually is a common goal around which they coordinate their social actions, at least insofar as their exchange really is a case of conversing. This common goal around which they coordinate is the following: that H understands what S intends to communicate. We can see that they have this common goal if we consider what would happen if this goal wasn’t reached, or if S didn’t know it had been reached (of if H didn’t know that S knew it had been reached, or if S didn’t know that H knew that S knew that it had been reached, etc). Suppose that H did not hear what S said, then H would probably respond with something like “Sorry, what was that?”, or suppose that H for some reason took S’s request for information literally as a question about whether or not he knew what the time was (this is of course very unlikely to actually happen, but run with it for the sake of the argument). If he then just responded “Yes”, without actually providing the information, then S would clarify that he is making a request for information (e.g. “Right, could you please also tell me what the time is?”). This suggests that the S and H indeed have a common goal during their brief conversation about the time. This in turn suggests that perhaps understanding joint intentional action will provide important insights about the activity of conversation after all.