A funny joke consists of:

  • A premise P that, for maximum humor, should be in some way familiar or relatable
  • A conclusion Q, that in some way involves injury, harm, embarrassment, or some other taboo
  • A "childlike" way to heuristically infer Q from P, which we may symbolize P ⊢ Q. A sensible adult would normally not conclude Q from P, but the joke does so perhaps by neglecting some common-sense information, or reasoning in a way different from usual, yet still partly justified by some aspect of P.
  • An element of surprise in the inference, where the crucial part of P ⊢ Q was not anticipated, even as a possibility, by the listener. In some real-life cases it is conceivable that P caused Q in actual physical fact, but P ⊢ Q should still not be what a sensible adult would expect to happen.

Is this "childlike inference" theory of humor associated with a philosopher who states it in approximately these terms?

  • 2
    This seems to be about jokes specifically, not humor in general, or am I mistaken here? Humor involves a lot more than just telling funny stories, doesn't it?
    – Philip Klöcking
    Apr 12, 2021 at 20:41
  • It doesn't sound mainstream, maybe could say is a mix of the superiority+incongruity theories of humour? plato.stanford.edu/entries/humor Humour makes more sense as play, indicated by queues, and what is otherwise uncomfortable can be explored eg real vs stated feelings. Satire and 'putting people in their place' in an appropriate creative entertaining way, are socially useful & require substantial intelligence & sensitivity, discussed here philosophy.stackexchange.com/questions/81070/…
    – CriglCragl
    Apr 12, 2021 at 22:07
  • 1
    I've only skimmed it but I think Inside Jokes by Daniel Dennett, Matthew Hurley, and Reginald Adams may have a similar idea. They add an evo-psych theory that it's an adaptation to help with correcting errors in our mental heuristics, and to motivate this debugging, nature "has to bribe the brain with pleasure. That is why we experience mirthful delight when we catch ourselves wrong-footed by a concealed inference error." So even if the evo-psych explanation is wrong it seems like they are focused on "inference error". Elsewhere they talk about why errors involving danger are funnier.
    – Hypnosifl
    Apr 12, 2021 at 22:20
  • @Hypnosifl that does sound related. Have they stated it all together, the familiarity + surprising childlike inference + taboo, as necessary and sufficient?
    – causative
    Apr 13, 2021 at 0:30
  • 1
    As I said I only skimmed it, but I didn't see them say those elements are all necessary. Do you think taboo is really needed though? There's a lot of absurdist humor that usually does seem to involve setting up some more familiar expectations and subverting them with something surprising that has its own kind of bizarro logic, but which doesn't seem to involve any particular taboo violations, like a lot of the jokes from Monty Python, or Steven Wright one-liners.
    – Hypnosifl
    Apr 13, 2021 at 3:55

1 Answer 1


Sounds like this kind of joking inference belongs to informal fallacy. Is Argument from analogy matching your expectation?

Electrons in an atom orbit a nucleus, and electrons jump instantly from orbit to orbit. Therefore, planets in a solar system jump instantly from orbit to orbit.

  • Well, that's kind of an example. I would say literally any joke is an example of childlike inference. In this case the "joke" about planets could be made funnier by making the rather austere premise ("planets in space") more familiar and relatable, by making the consequences more socially taboo, and by setting up the premise so that it's more understandable how someone could make that mistake with childlike thinking.
    – causative
    Apr 12, 2021 at 23:36
  • For instance, take: i.pinimg.com/originals/3c/da/c9/… The premise P is familiar: one guy has asked another whether his dog bites, and the dog owner said no, and the first guy presumably tried to pet the dog. The result Q, that the guy on the right was kicked by the dog, certainly involves harm. The inference P ⊢ Q is understandable under the idea that "another thing dangerous animals can do is kick," which is of course childish logic when applied to dogs.
    – causative
    Apr 12, 2021 at 23:39
  • Here's a way the planets one could be made a little bit funny: say that you are watching a science cartoon where an anthropomorphized electron ("Lenny the electron") is showing you the universe at different scales. Previously he demonstrated how he jumps among orbits in the atom. Then later he's scaled himself up to the size of a planet, and does the same orbit-jumping thing, bumping into an anthropomorphized planet who is very annoyed. "Hey buddy, watch it!" "Oops, sorry, things are different where I come from." Little bit funny.
    – causative
    Apr 12, 2021 at 23:47
  • The anthropomorphizing makes it more relatable, and there's social embarrassment in the conclusion, and it's more understandable how the electron could make that mistake with childlike logic (because that's how things are done where he's from).
    – causative
    Apr 12, 2021 at 23:49
  • @causative I wouldn't say literally any joke is an example of childlike inference, many jokes are caused by a pun (like informal equivocation fallacy). Such as "Light travels faster than sound. That's why some people appear bright until you hear them speak". Apr 13, 2021 at 2:00

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .