Steven Wright, the deadpan American comedian, has a line that runs like this: "I've been getting into astronomy, so I installed a skylight. The people who live above me are furious." Sit with that for a second. The first half of the joke quietly builds a picture in your head. A guy on the top floor, cutting a hole in his roof, lying back to watch the stars. You can see it. And then four words — "the people who live above me" — detonate the entire picture, because if there are people above him, he was never on the top floor. He cut a hole in someone else's living room.
That little explosion in your head, the half-second where your brain scrambles to rebuild the scene, is the engine of nearly every laugh you've ever had. This whole section is about that engine — what it is, why it fires, and why understanding it turns "I'm just not funny" into a problem you can actually solve.
Here's the through-line that holds the whole course together, applied to its very first gear. A joke is a controlled accident in someone's expectations. You lead a person to believe one thing is true, and then you reveal that something else was true all along. Philosophers have a name for this, and it's the dominant modern account of what makes things funny. It's called incongruity theory. The Stanford Encyclopedia of Philosophy traces a version of it all the way back through Kant and Schopenhauer, and the basic claim is simple: humor lives in the gap between what you expected and what you got.
Now, that idea sounds almost too tidy, so let's slow Wright's joke down to bullet time and watch the gears turn. The setup does a specific job. It's not there to be funny — it's there to plant an assumption. "I installed a skylight" tells your brain, without you noticing, that this man owns the ceiling above him. You never decide to believe that. You just do, automatically, because that's what a skylight implies. The comedian Andy Bumatai, who won an Emmy and wrote about this for the Toastmasters magazine in 2018, put the structure plainly: a joke that works means the audience believes one thing to be true, only to have it revealed that something else is actually true. The involuntary reaction, he wrote, is usually laughter.
So the setup builds the wrong belief, and the punchline breaks it. But there's a subtler move hiding in the middle, and it's the one most people miss. It's called misdirection — actively steering the listener away from the real meaning so the reveal hits harder. Bumatai makes this point with a great before-and-after. Strip Wright's joke down to just "I've been getting into astronomy, so I installed a skylight," and there's no joke at all. It's just a guy telling you a fact. The line "the people who live above me are furious" is what does the work, because it leads you somewhere — into picturing his cozy upstairs apartment — right up until it yanks the floor out. Misdirection isn't decoration. It's the thing that builds the height you're about to fall from.
This is where most people get stuck, so let's name it directly. The natural assumption is that the funny part is the surprise itself — that a joke is just an unexpected ending, and the more shocking the better. That's wrong, and it's wrong in a way worth understanding. Your electric bill doubling is surprising. Your kid wrecking the car is extremely surprising. Neither one makes you laugh. Bumatai's word for the difference is that a comedic surprise is a twist without serious repercussions, and we'll get deep into exactly why that "safe" part matters in the next section. But there's a second, more mechanical reason raw surprise isn't enough, and it's the part that really makes the engine click.
The brain doesn't just need to be surprised. It needs to be able to solve the surprise. Go back to the skylight. The reason it lands isn't only that you're caught off guard — it's that in the same instant, a second meaning snaps into place that explains everything. Oh. He's not on the top floor. He cut into the neighbor's apartment. That's why they're furious. The whole picture reorganizes itself around a new logic, and that reorganization, that little click of "ohhh, that's what was going on," is the satisfying part. A surprise you can't resolve isn't funny. It's just confusing, or unsettling. Incongruity has to be resolved to pay off — your brain has to get to throw open a door it didn't know was there and find the room already furnished.
Bear with this for one more step, because it leads somewhere genuinely surprising. If a joke is really an ambiguity you resolve, then in principle you ought to be able to measure funniness — to put a number on how much hidden meaning is packed into a sentence. And it turns out somebody did. A 2016 paper in the journal Cognitive Science, by Kim Binsted and colleagues, built a computational model of how the brain processes sentences and pulled out two measurements. They called them ambiguity and distinctiveness. Ambiguity is roughly how many different meanings a sentence can support at once. Distinctiveness is how sharply those meanings differ from each other.
They tested this on a pile of puns and ordinary sentences — and here's the share-worthy part — both measures lined up significantly with how funny human beings rated those sentences. Even better, within the set of puns alone, the distinctiveness score could separate the genuinely great puns from the mediocre ones. Let that sink in. The thing your gut feels as "that's clever" turns out to be partly a quantity: how many meanings are hiding in there, and how far apart they are. The researchers used the example, "The magician got so mad he pulled his hare out." Two meanings — pulled his hair out in frustration, and pulled a rabbit out of a hat — riding on one string of words. The funnier a pun feels, the more cleanly those two meanings split. As the philosopher Henri Bergson described a pun back in 1914, it's a sentence where two different sets of ideas are expressed, and you're confronted with only one series of words.
So if someone stopped you right here and asked what the difference is between a pun that makes people groan and a pun that makes them laugh — what would you say? … It's not how clever the wordplay is. It's how far apart the two meanings are, and how completely the second one explains the setup.
Now, none of this means humor is settled science, and it's worth being honest about where the disagreement lives. Incongruity theory is the dominant account, but the researchers behind that computational model were careful, and the field's own skeptics are loud. A humor scholar named Graeme Ritchie pointed out years ago that there still isn't a rigorously precise definition of incongruity — one that would let you look at any given joke and objectively say "yes, incongruity is present here." That vagueness is exactly why some computational researchers, like the team behind an earlier study by Rada Mihalcea, found that incongruity-based features barely beat random guessing, and concluded you were better off using joke-specific templates instead. So the camps split. One side says incongruity is the deep, general engine under all humor. The other says, fine, but it's too fuzzy to build on, so just memorize what works. The Binsted paper is the strongest evidence for the first camp, because it took the fuzzy idea and made it numeric — and the numbers held up against human judgment. That's the side the evidence leans toward, and it's the side this course is built on, because a fuzzy-but-real engine you can train beats a tidy template you can only copy.
Here's where this stops being theory and starts being something you can do on a Tuesday. If a joke is just a broken assumption, then writing one is reverse engineering — you find an assumption nobody noticed they were making, and you break it. There's a teaching tradition built entirely around this, and you can find it laid out at a site called Creative Standup, which calls it the Broken Assumption joke. The method is almost insultingly simple. Start with something true. List the assumptions buried inside it. Then find a punchline that flips one of those assumptions into a different, surprising meaning.
Let's run it live. Bumatai's worked example starts with the truest, flattest sentence imaginable: "I had eggs for breakfast." No joke there yet — and that's the point. He warns against trying to be funny too early, because that's how you end up with something overdone and unbelievable. Start real, he says, and the audience won't see the surprise coming. Then he lists the silent assumptions packed into that boring sentence. The eggs were cooked. They were chicken eggs. It was a reasonable number of eggs. They were his eggs. Each one is a door, and the joke is just walking through the one nobody expected you to open. Break the "chicken eggs" assumption and you get: "I had eggs for breakfast. They were fresh, they were delicious — they were Cadbury." Break the "morning" and "food" assumptions another way and you get something stranger about the light turning green every time you take a bite. Same flat setup. Different doors. Different laughs.
You can hear the engine running underneath the pros, too. The comedian Bill Engvall has a bit about skiing with his family: "I hit two trees and fell down a ditch. And that was just walking from the lodge." Your brain assumes the disaster happened on the slopes, because that's where skiing disasters happen. The punchline reveals he never made it to the slopes — same machinery as the skylight, just a different door. John Mulaney, Jim Gaffigan, Dean Martin — comedians whose styles couldn't be more different — are all, underneath, doing the identical thing. Plant an assumption, then break it with something the listener can instantly resolve into a new and funnier truth.
So strip away the detail, and three things are doing the real work in every joke you've ever laughed at. The setup quietly builds an expectation you never agreed to. The punchline breaks it. And the laugh comes not from the break alone, but from the snap of a second meaning clicking into place to explain everything — the more meanings hiding in there, and the further apart they are, the harder it lands.
That's the engine. But an engine that fires the same way for everyone would mean the same joke works on everyone — and you already know that's not how it goes. The skylight bit kills in one room and gets a polite nod in another. Nothing in the broken-assumption machinery explains why. The missing piece isn't about how the surprise is built. It's about whether the listener feels safe enough to enjoy the break — which is the strange, table-splitting puzzle at the heart of what comes next.