'A form of chorusing, almost singing together'

Three guys stranded on a desert island find a magic lantern containing a genie, who grants them each one wish. The first guy wishes he were off the island and back home, and it's granted. The second guy wishes the same. The third guy says: "I’m lonely. I wish my friends were back here."

Researchers at Oxford University gave a group of undergraduates from the London School of Economics 65 jokes, most of them from successful stand-up comedians. Some jokes were short, others more complex.

Professor Robin Dunbar, who headed a study on the students' reactions, shares his insights into what makes a good laugh line.

DW: What makes a successful joke?

Robin Dunbar: It seems that the more mindsets there are in a joke, the funnier it becomes. But there is a limit on how complex they can be. We can't handle more than about three characters in a joke whose mental states create the joke: their misunderstandings of what's going on in the world or their misinterpretation of word meanings. Those kinds of jokes come across as particularly funny for us.

Of course, that's not the only thing that makes jokes - it's not the final answer to all human humor (laughs); other things are also important in humor, for instance the surprise element.

Why do people tell jokes at all?

This has to do with the bonding mechanisms we use to create both our own relationships with other people and the level of cohesion and bonding within communities. The very primitive mechanism we inherited from our primate ancestors is effectively social grooming, and, at some early state in human evolution, we needed something that was more efficient to get more people groomed simultaneously than is possible by actual grooming. Laughter allows us to do that.

In our study, we were interested in what constrains the length and complexity of jokes given that our abilities to manage other people's mind states is fairly limited. We can only cope with so many minds, that is characters, at any one time. In the context of understanding the constraints on literature in general, storytelling and novels, etc., jokes seemed particularly interesting because they're very simple and clear.

Are there differences between the cultures?

A lot of jokes are about confusion of word meanings, where, in any language, we use a word metaphorically, or have two words that sound the same but mean something different. Confusions create jokes - that's what makes us laugh, a sort of verbal equivalent of Charlie Chaplin-type slapstick comedy.

This works across cultural boundaries, but, of course, when it's a word-based joke, it will depend on the particular language.

Would you say humor is at all subjective?

Not as such. It seems to be very primitive, very visceral, in that we can't help laughing at something when other people laugh, even when we don't understand the joke. Let's say we were in China, sitting round a table, and somebody told a joke in Chinese and everybody laughed - we would find it very difficult not to laugh.

It's very intensely social. It really is a form of chorusing, almost singing together. In fact, we think it's the precursor to singing. Singing without words emerged out of laughter later in human evolution. We share laughter with the great apes. They also have a version of laughter, a call they give when they are inviting play. We just use it more intensely. Eventually in the course of human evolution, when language evolved, we developed the capacity to use language to trigger laughter by telling jokes. It creates a sense of bonding and allows us to groom. The problem with grooming - which we call cuddling or patting and which is the natural mechanism to create our social bonds and group cohesion - is that it's a one-on-one activity. Laughter allows us to trigger the same mechanism with several people simultaneously. It's a form of grooming at a distance.

Women and men laugh at different things, surely?

Not necessarily. What seems to be the case is that women are more likely to laugh at a joke by men than men to laugh at jokes by women. Both laugh at jokes - very bad jokes usually - by their bosses.

Robin Dunbar is a professor of evolutionary psychology in the Department of Experimental Psychology at the University of Oxford. His research is focused on understanding the "behavioral, cognitive and neuroendocrinological mechanisms that underpin social bonding in primates" - humans in particular. The new study is published in the scientific journal Human Nature.