Building a Joke Engine

Let’s go through the build process of a mental engine, specifically one that many people would find useful — a joke engine.

What is a mental engine? It’s the program you run to compute complex functions subconsciously, like to create something new, or come up with something funny, or identify a pattern, or tell a good story, or spot an insight, or draw a connection.

You already run these programs, and you already have libraries installed that the output of these programs depend on.

If you’ve ever come up with an excuse or a compliment in real time, then you have a mental compliment or excuse engine that helped you do that, using compliment, excuse, pattern, and social libraries that your experiences installed in your brain.

What I want to accomplish by the end of this post is:

  1. breaking down a built-in mental engine into components we can understand
  2. building a brand new mental engine using adjusted versions of those components

So by the end of this post, you’ll be able to build engines whenever you want to add more complex processing to your brain.

Disclaimer: if you still want to find every joke funny, click out of this story and into another story right away. Your standards for comedy will be much higher after you read this and you will find less jokes funny, although you will also be funnier, unless of course you’re already the funniest person on Earth. Choose your path with care; I cannot and would not choose for you.

First we’re going to examine a joke (the conceptual object we are building an engine to output) to gain some insight about comedy.

Then we’re going to establish some attributes that apply to all jokes, so we can build an engine to produce them.

Lastly, we’ll build the joke engine using engine attributes & our newfound insights about comedy.

  1. Joke Analysis

Let’s look at an example of something that is funny, and reduce it into insights. Maybe you’ve seen Melissa McCarthy’s impression of Sean Spicer:

Why is this funny? Well, first of all, it’s absurd. The contrast between Spicer’s display of ego and his lack of ability to deliver on that ego is ridiculous. Also, the obvious lie inherent in complimenting himself while insulting everyone else, given his lack of ability, is amusing because it’s so obviously false. Idiocy makes us laugh because it makes us feel better about our mistakes. That suggests an element of psychological safety is important for humor. The fact that Spicer cares so much about persuading people that he’s right is funny because those are high stakes for something that doesn’t really matter to the balanced person. The fact that he assumed no one would be able to point out that he was legitimately wrong is a funny assumption.

Funny speech mannerisms (a Southern accent, a grown woman calling her father her ‘daddy’, a maniacal reference to silly words like hot dogs, a grown man using childish mocking to mask his inability to come up with a good point) introduce a silly element that undercuts the otherwise serious subject (a representative of democracy being better suited to another job). We also see an element of surprise at work; it’s shocking that a public servant would behave this way so our expectations of public servants are blown to smithereens, and that’s a contributing factor to the humor.

So by describing and cataloging the qualities of this comedy bit, we’ve already amassed some useful observations that might become insights. Let’s organize those into a list.

Already we’ve noticed that humor has qualities like silliness, imbalance, absurdity, surprise, obviousness, falsehood, safety, high stakes, assumptions, and has behaviors such as A, discrediting egotists and liars, which serves to make us feel safer being ourselves, and B, surprising us by subverting our expectations.

If you read up on Wiki articles on humor, you’ll spot similar patterns people have identified:

Of course you don’t have to rely on articles to tell you the mechanics of concepts like humor; look at how much we noticed & figured out for ourselves, just by watching a single video.

Armed with these insights about how comedy works, we can move on to the steps for building a mental engine to produce complex output. Then we can implement these insights with a brand new mental engine so our social circles will think we’re funny.

2. Mental Engine

Step 1: Describe the System

Identify the rules that the system follows, using patterns you’ve noticed about the relationship between the system’s input & output to reverse-engineer the function producing that output. This translates into listing the rules that your concept follows as well as attributes that allow you to define the system.

For a concept like comedy, we’ve already defined some rules & attributes above. By analyzing other jokes & thinking about it more, we can define the following list of comedy rules:

  • make the joke concise, cut out anything that isn’t surprising or particularly funny, doesn’t fit with the tone, pushes an agenda too obviously, humors our ego too much, etc
  • adjust for your audience; if a joke isn’t timely or insightful it won’t make some people laugh because they’ve heard similar jokes before or about the same topic, or they may not get the reference at all.
  • do it with love: if you make a joke about a group of people and some people in the audience are offended, your joke came off too mean. Comedy is supposed to make people feel safe so they can evolve and grow out of whatever behavior was just ridiculed. As Monty Python vet John Cleese suggests here, comedy can’t be too mean if it’s going to fix anything.
  • make sure the joke has some or all of the following attributes of humor: silliness, imbalance, absurdity, surprise, obviousness, falsehood, safety, high stakes, and assumptions.
  • timing is key; we’ve all heard about a comedian’s perfect comedic timing — to see how important this is, you can repeat a comedy bit with pauses in different places and see if it makes you laugh as much as the original. Pauses can be used to wait for the audience to make an assumption and realize they’re making an assumption, which can be funny if done right.
  • make sure your comedy has a useful or new point to make; the point of the clip above is that our egos can be our downfall. This isn’t a new insight but it bears repeating, especially given the behavior of some public servants throughout history. If you tell a joke that doesn’t have a point, no one may get it or find it funny.

Step 2: Build a Function

Once you’ve identified the system’s attributes and rules of interaction, assemble them into a function. This function should follow existing thought processes when thinking of a joke, unless you want to install additional libraries (recommended but not required) — what any humans in your social circles might call learning.

Now we can assemble our comedy insights and our tribute to oversimplification of a joke function into an engine that we can call whenever we want.

3. Building the Joke Engine

Just kidding; you already did this step. By analyzing what makes a joke work, and identifying the rules & attributes that characterize the concept of comedy, and translating this into a function, you’ve already built a new engine that you can call whenever you’re in a tough spot and need to crack a joke. You can modify the function however you want, add attributes, add extra rules and additional processing functions called within it — whatever makes your joke engine output just right for you and your audience.

As you start consciously using engines that you’ve built, you’ll start noticing other people using their own engines to produce output, and will be able to gather new attributes & rules to build in to your function by examining their output. Sometimes you’ll be able to spot their mistakes; other times you’ll be amazed at what they can do with their own situational context and your engine will evolve as you learn from them.

As you use a particular engine more and more, it will get easier & will become second nature to you, so coming up with a new conceptual object (a joke or an insight) will be next to effortless for you. At that point, when you’ve achieved computational time approaching zero, it may be time to teach what you know, pick up some tips from another engineer, and build a new version.

Build on, Engine-Builders

By the way, the process we just worked through will work for any other concept you want to build an engine for: creativity, predictions, insights, perspectives, whatever conceptual object you’d like to be able to generate on demand. You can take an example of that object, analyze it, describe its system, and build a function. You already did it consciously at least once, and subconsciously many other times, so we know for a fact that you can do it again. Think it through, and trust your brain. You and your brain can do great things when you give it thinking time to play.

One clap, two clap, three clap, forty?

By clapping more or less, you can signal to us which stories really stand out.