Humor & Comedy Craft

I don’t think of comedy as magic. I think of it as mechanics — a small set of load-bearing parts that, assembled correctly, produce a reliable result. That belief is the whole reason I can claim this as a domain: not because I’m the funniest person in the room, but because I’ve taken funny apart to see what makes it run, and I’ve built a working framework and a content engine (Library of Laughs) on top of what I found. My authority here is the systematizer’s, not the performer’s.

The engine is incongruity. Almost every laugh traces back to a violated expectation: a setup builds one thing, the payoff delivers another that’s surprising but, in hindsight, logical. And the rule underneath it is quantitative — the size of the gap between what you expected and what you got is the size of the laugh. Once you see comedy that way, you stop hoping for jokes and start engineering gaps.

Specificity is the fuel. This is the part most people get wrong, and it’s also the part that connects comedy to everything else I do. “A hat” is inert. “A 1952 Stetson that’s seen more funerals than a hearse” is funny. Concrete nouns, real numbers, named things — vague is the enemy of funny, exactly the way vague is the enemy of authority. The same instinct that makes a claim credible in the William Bernard Standard makes a line land in a joke: specificity earns the response. I didn’t discover two rules there. I discovered one rule wearing two hats.

The strongest humor isn’t invented — it’s recognized. Observational comedy works because the reader thinks “yes, exactly.” You’re not manufacturing the joke; you’re naming something they already felt but never put into words. That’s why truth beats cleverness: the laugh is really a flash of recognition.

Then there’s the delivery layer, which is pure craft. Economy — every syllable that isn’t serving the laugh is dampening it. Terminal placement — put the funniest word last, because the tiny pause before it does the work; comedy is as much percussion as content. The rule of three — two straight beats set a pattern, the third breaks it. And commitment — you cannot apologize for a joke and land it. No “lol,” no hedging, no explaining it afterward. Deadpan confidence is the delivery mechanism.

Because it’s mechanical, it’s repeatable — and that’s the real point. Humor fails when people treat it as inspiration you either have or don’t. I treat it as a process: draft the informative version straight first, with zero jokes (you can’t make a boring point funny, but you can make a clear one funny), then layer the humor on a second pass — and only at the positions readers actually see: headlines, subheads, the first line of a section, captions, and the parenthetical aside, which is my favorite tool because it reads as the narrator’s private commentary. Then read it aloud, because dead jokes and rhythm problems are audible instantly and invisible on the page. Then cut the weakest fifth, because mediocre jokes actively lower the perceived quality of the good ones.

And there’s a governance line I won’t cross: punch up, never down. At power, at myself, at shared absurdity — never at the vulnerable or at my own audience. Self-deprecation is the safest and most endearing default. That’s partly ethics and partly risk management: down-punching is the single biggest reputational hazard in branded humor, and I’m building durable assets, not disposable jokes. For the same reason I favor observational and structural humor over topical references, which decay fast and travel badly. I want a line that still lands in three years.

That’s the domain in one sentence: comedy is an engineerable craft whose fuel is specificity and whose ethics is punching up — and knowing that is what lets me write it on purpose instead of waiting to get lucky.