Un-Cooperative Principle: Conversation Design and Humor

Comedy

Why is the following conversation kind of funny?

Billy: “Hey Dave, you know what day it is?”
Dave: “Yes.”

Let’s figure it out. But first…

What is Conversation Design?

I work in conversation design, which is a fancy way of saying that I help build chatbots. In the past, I’ve helped numerous Fortune 500 companies and brands turn their brand into conversational experiences by teaching them the principles of conversation design and how to apply them.

I am also a comedian. It’s my passion and joy to make people laugh. I’ve found that these principles of conversation design that I work with, can be used as a lens through which we can analyze what makes certain types of humor funny.

So, why was it kind of funny when Dave was a jerk?
I’m going to explain this with the combination of two ideas. First, from conversation design, and second from comedy theory.

The Cooperative Principle

People have been talking for a long time and there have been people who have been researching how that works. Paul Grice, is one of them. Back in the 1970s, he came up with what’s called “The Cooperative Principle”. A set of four rules (or maxims) describe how people achieve conversational communication in normal social situations.

The four maxims are as follows: “Maxim of Quality”, “Maxim of Quantity”, “Maxim of Relevance”, and “Maxim of Manner”.

The important thing to remember here is that when people talk, they usually are cooperative – they all try to help each other have a successful conversation.

In my line of work, we usually ask our clients to have the chatbot they are designing to honor the Cooperative Principle. But I think we shouldn’t always. More on that later.

The Subversion of Expectation Theory of Comedy

There are countless ways people think how comedy works. Some say it’s about superiority, some say it’s about relief, and some others say it’s about incongruities. My view is that these are all labels describing the same thing from different angles – like how you might describe the Pyramids of Giza as squares if you looked at them from above, or I might describe it them as triangles if I saw them from the side. In this case, I’d like to take on the idea that humor is about subverting expectations. An example:

You see a man about to walk through a door. The door turns out to be a painting, and he runs into a wall!

Badum-tss. Haha! So funny! Your expectation was subverted!

Yeah, it’s not laugh-out-loud funny. But it’s a funny idea nonetheless. This is because it’s an illustration of an expectation being subverted.

Adding Both Together

Let’s take a look at the conversation again:

Billy: “Hey Dave, you know what day it is?”
Dave: “Yes.”

Given the idea that humans expect others to be cooperative in conversation as described in “The Cooperative Principle”, anything that is uncooperative is going to be unexpected. So…

Dave was kind of funny because Dave broke the Cooperative Principle, which subverted expectation.

(Of course, if Dave is always uncooperative, he is just going to come across as a massive jerk. But that’s a whole another issue.)

So what?

Technology is currently at a nascent stage in terms of computers beings able to carry out natural sounding conversation. Yes, we are getting closer than ever before, but if we were talking cars, we are in the Ford Model-T stages.

We are still far away from being able to have hours of deeply engaging conversation with computers like we might with our best buddy over drinks. But if we want to get there, as part of the effort, it will definitely be worth investigating how we can implement uncooperativeness in our chatbots, like real humans do in their conversation.

Because, hey, humans aren’t perfect. But that’s what makes us perfect.