Why do jokes come in rounds, like drinks? And what makes them an interactional safe haven in a group therapy session? These are some of the unlikely questions you encounter if you’re reading Harvey Sacks’ Lectures on Conversation, a two volume compendium of his groundbreaking work on the structure of conversation. Originally given in the 1960s and early 1970s, these lectures are full of original insights and sparkling observations. Here’s the relevant bit on jokes from Lecture 12 in part I, Volume I (a lecture originally given in 1965):
The first thing that’s important about jokes is that to use one is something like buying a drink among a bunch of people: They come in rounds. And if some person tells a joke then every other person present has the right to tell a joke.
Sacks writes this after giving a lengthy extract from a a group therapy session in which people do exactly such a round of jokes. Why do they even happen in the context of such a session? He points out a number of structural features that may play a role in creating a need for such a device.
First, such a group therapy session typically goes on for about two hours. Second, in such a session, “occasions of silence are extremely dangerous to all persons concerned”, because when the conversation lapses, one thing people will do is “begin to look for a face that’s noticeable among themselves, and then pose for that person the problem of giving an account of why he is silent”. Third, silence is dangerous, but so is talk: “any given body of talk, starting at any given place, will, if allowed to go on, end up dangerous. That is, it will end up on some topic which is perhaps too important to be talked about except under real feelings of relaxation.”
Jokes as emergent solutions
In short, in a group therapy session, you are likely to find yourself in an interactionally fraught situation. And while this is in part the point of therapy, obviously it is also an interactional problem. One that, it turns out, people can sometimes collectively overcome with emergent solutions that alleviate some of these pressures. As he explains:
For someone to use a joke on that occasion is then to give each other a set place to talk, and also to give [them] something to say.
In other words, a round of jokes can provide a neat collective solution to the interactional and personal risks involved when silences happen, or when talk is perhaps all too free. It provides a kind of interactional shelter, giving people a thing to do and a place to go. This is connected, Sacks points out, to another useful feature of jokes:
when jokes are told they are things that are ‘going around’; they are quotes. So they’re unaffiliated remarks, and in that sense it’s hard to say about somebody that the fact that they told some particular joke has some special significance. They just heard it, and now they’re repeating it.
If jokes provide an interactional shelter by opening up opportunities for relatively innocuous talk, then their second-hand nature provides further insulation against risks, offering a form of plausible deniability. It’s not really you saying something; you’re merely passing on something you heard.
You could apply Goffman’s work on footing here and note that they’re merely animating something that they haven’t authored, though I like the simple clarity of Sacks’ observation that these are unaffiliated things. The brilliance of Sacks’ observational style —which relies on his having access to a lot of conversational materials— is that he is then able to connect this to other things that are not jokes but that still share this unaffiliated nature. There are things “used in similar circumstance to the jokes”, namely when there is a lapse in conversation:
What they do is, at points in the conversation when either nobody is talking or they haven’t talked for a while, insert slogans. They’ll just come out with a piece of an advertisement from the radio, or a jingle, or obvious quotations sarcastically said. Again, then, it seems that they go about monitoring when they ought to be talking or when silence seems to be preent, and flick out these things which, again, have this unaffiliated character.
Deep structure of interaction
A final note. I think the sequential and interactional nature of Sacks’ observations is what makes them stand out. A simple content analysis-style take on this material might find that some group therapy sessions feature jokes and might guess that they ‘add levity’ and perhaps ‘relieve pressure’ or whatever. Indeed I would not be surprised if there is some discourse analytic work arguing this, and some experimental work showing a ‘levity effect’ of jokes in therapy contexts. (I haven’t checked, but feel free to weigh in in the comments if you’re into this.)
But such analyses stay altogether too much on the surface. They would miss the very real interactional challenges that Sacks so sharply saw; the structural and sequential features of the resources that can be mobilised to counteract them; and the intricate interactional achievement of these solutions as they emerge in interaction. Only the study of interaction in vivo can deliver deeper insights, and can really hope to answer the question of why language in interaction works the way it does.
Sacks, Harvey. 1992. Lectures on Conversation. 2 vols. London: Blackwell.