A theory of humor

Are you a developer?

Join my startup: Elody

Work together with other developers and build a Gestalt AI.

Elody can make your programs accessible to
non-technical users.

A theory of humor

Lots of people have tried to give a scientific explanation of humor, but I have found that all the explanations I have read so far are missing something and fail to account for one oddity or another.

Most theories of humor try to pare everything down to only a single overarching explanation for everything. While it would certainly be neat if that was possible, nature is messy and evolution does not usually develop a single solution to deal with a single problem.

This is my own theory, inspired by my research into the way the brain works. It is similar to the incongruity-resolution theory, but with an added evolutionary basis.

I call it FDIS for "flaw detection and ingroup signalling". This is because I believe that humor serves more than one purpose and those two are the most important ones.

Basic Idea

Humor is based on a concept almost but not quite making sense. A concept can be humorous if there is a tiny mismatch that causes everything to fall apart.

In the case of a joke, this mismatch is often what is revealed in the punchline. When a joke is told, the listener builds up a model of the world based on the joke's content. The punchline reveals this flaw. If the flaw is visible since the beginning the joke does not have a punchline. Instead, humor is derived from how far the speaker is able to strain the model despite the obvious flaw in it.

There is a lot of variety in what a 'tiny mismatch' in a concept needs to be like so that it is perceived as humorous. It depends on culture, intelligence, disposition (dark humor, fart jokes, violence-as-joke) and developmental stage (babies find very simple things funny, such as peek-a-boo). This is why 'humor' is so hard to define. It depends on the state of the brain perceiving it.

At its root, all humor has the same simple pattern "flaw in a concept", but whether or not something is actually perceived as funny depends on a plethora of factors so that it can't be easily defined scientifically, not with our current limited understanding of the brain.

Note that the flaw in the concept must be perceived and corrected on a subconscious level, not a conscious one: The brain must be capable of perceiving the flaw in the concept without promoting the concept to a higher level of awareness, analysis and introspection. This is why people are so often unable to tell why they find a joke funny. It is an important part of 'getting' a joke that it happens subconsciously, because finding the flaw in the system through analysis and introspection on a higher level of consciousness is not indicative of the ability to detect the same flaw on a subconscious level. This is also why "explaining the joke" is viewed negatively.

Evolutionary Purpose

Why would an animal evolve humor? The benefits of detecting flaws in a mental concept are self-evident, but there is a deeper reason for why we have evolved to react positively to humor and to laugh at jokes, even though laughter is loud and could attract predators.

I believe that humor was originally an out-of-control form of positive feedback: The brain isn't perfect, and sometimes the positive reinforcement of learning-by-resolving-an-inconsistency is much stronger than it should be.

This would mean that the origin of humor was originally a malfunction of the brain. However, evolution is ingenious at repurposing things: Even though humor may well have started out as a malfunction, it was quickly adopted as a social tool and became useful: Telling a joke means describing a concept and a flaw in that concept in such a manner that both you and the listeners have the same overreaction to it.

In order to understand jokes, the concepts in your brain must be similar to those of the speaker or else the joke won't be perceived as funny. Additionally, you need to be smart enough to see the flaw in the concept. Sharing similar mental concepts with other people (i.e. thinking in similar ways), and being smart enough to recognize flaws in those concepts, are both very useful traits in potential allies.

Somebody who laughs at your jokes, or somebody who tells a joke that you find funny, is both intelligent and similar to yourself. Therefore, telling and laughing at jokes is an excellent way to signal intelligence and ingroup membership.

Since humor is so heavily tied into social signalling, cultural norms and values can in turn influence what is considered funny. Examples: Not laughing at a mistake by a high-status person or institution, tragedies being 'not funny' when similar but older tragedies can be safely joked about, 'bad taste' in humor, 'crossing a line', aping the behavior of an outgroup member.

Because humor can become very complex, it is even useful for transmitting information while maintaining plausible deniability ("Of course we weren't really planning to do X. It was only a joke.").

Some Examples

  • Tickling, which works on all higher primates, overloads the brain with lots of unpredictable signals in vulnerable parts of the body, but paradoxically no pain is received (concept: I am being repeatedly hit in a vulnerable spot ; flaw: I am not feeling any pain). This explains why only vulnerable parts of the body are ticklish and why tickling does not work if you use too much force so that it hurts.

  • Inside Jokes are primarily funny due to their social aspect. Only someone who has shared past experience with you is able to understand an inside joke, which makes it more useful for signalling ingroup membership.

  • There are gags that continually repeat the same scenario multiple times (for example: The parrot sketch of Monty Python).

    Here, the humor consists in repeatedly attempting to force the mismatching concept to work even though it has already been revealed as flawed.

    Importantly, the same general scenario is explored in variations: Telling the exact same joke twice is not funny because both the concept and the flaw in it are already known. In contrast, if either the concept (returning a dead pet when the shopkeeper pretends ignorance) or the flaw shown in it (the absurd way in which the shopkeeper pretends ignorance and the turns of phrase the pet owner is forced to use) are always slightly different, it stays humorous.

    Instead of only noticing a flaw in the concept once, your brain is repeatedly shown ever more absurd ways to attempt to make the concept work, and each such attempt just reveals another flaw.

  • Immature humor vs refined humor: The humor preferred by younger people is very simplistic (e.g. fart jokes). As a human grows up and becomes more experienced, their sense of humor gradually shifts away from low-brow humor to high-brow humor. This is a type of social signalling. Younger and less intelligent people are simply not able to 'get' the jokes of their more intelligent and/or more mature contemporaries, and the latter draw pride in that.

Comparisons to other theories of humor

Let's compare FDIS ("flaw detection and ingroup signalling") to some other theories of humor listed on Wikipedia

  • Incongruity theory

    The root of my FDIS matches incongruity theory, but FDIS adds an evolutionary basis and adds social aspects.

  • Relief theory

    Relief theory states that humor exists to relief tensions. Suppose that a person is very stressed out. Several parts of his brain are raising alarms at each other in a constant feedback loop that just stresses him out even more. It is known that a joke can break someone out of despair, at least temporarily. Relief theory explains this. Can FDIS explain this as well?

    Yes, it can: Humor in this case is about pointing out a flaw in the feedback loop of alarms in the brain. Whether or not this flaw is actually factually correct, it will at least temporarily interrupt the feedback loop, which causes relief. If this explanation based on FDIS is correct, it would imply that humor can only provide relief if the stress is caused by a complex system of factors, since simpler systems are too simple to easily find a flaw in them, and so humor can not be found in them.

    This would appear to be correct: I can not imagine that someone who is afraid of a fire, or a spider, or some other perceived immediate threat, could be broken out of it with humor.

  • Superiority theory

    This matches the social aspects of FDIS very well.

  • Benign violation theory

    This theory sounds too broad to me to be useful, since it does not adequately explain what a 'benign violation' is.

    Ironically, FDIS is even broader. Only a small subset of 'flaws in a concept' count as humorous, and we simply lack the ability to accurately model which ones. FDIS is too broad a theory to be useful, but only because we currently lack the ability to build models that are accurate enough.

    However, the general tendency that smarter and older people have more refined/complicated humor agrees with FDIS.

  • Computational-Neural Theory of Humor

    Computational-Neural Theory of Humor states that humor is about timing, i.e. about recognizing things more quickly. In contrast, FDIS says that humor is about recognizing things at all.

    Both "he doesn't get it" and "he took a while" are frequently used phrases, but making complex jokes and understanding complex jokes at all seems to be more important than the amount of time it takes to recognize the joke.

    More importantly, the Computational-Neural Theory of Humor ignores social aspects of humor, which are evolutionarily relevant.

  • Detection of mistaken reasoning

    This is the most similar to FDIS. It is however missing the social aspects, which I believe are crucial to understanding humor.