How do we quantify an aspect of complexity in humor studies? Consider Gell-Mann's simple and elegant "effective complexity".
"Effective complexity is a simple and elegant measure of complexity. Every physical system has associated with it a quantity of information - the amount required to describe the physical state of the system to the accuracy allowed by quantum mechanics. The basic way to measure something's effective complexity is to divide that amount into two parts: information that describes the regular aspects of the thing and information that describes its random aspects. The amount of information required to describe a system's regularities is its effective complexity.
In a humor system, such as telling jokes, the effective complexity is essential to the equal length of the system's blueprint (set-up and punchline): it is the amount of information required to put the system together.
In a joke, for example, the blueprint specifies the type of set-up (word, gesture, a question-answer, or a whole short story) and the content and procedure (irony, sarcasm, word play and other devices) to deliver the punchline.
Words and gestures (in relation to a particular context) are regularities of the joke design; the bits that specify these features have to take on specific values if the joke is to be funny. These bits figure in the joke's effective complexity. But the blueprint does not specify every nuance of the human experience which makes the joke funny. The bits that specify just what it is that makes one joke funny and not another is an accident; these things do not contribute to the funniness of the joke, nor are they an indicator of its complexity.
Complexity is a key issue in engineering. So, how do we engineer a humor system that is robustly funny in its delivery? Give it a big, fat KISS!
Keep It Simple, Silly! You want a joke that is just complex enough to make it funny, no more. You want the joke to maintain its ability to carry out its functional requirements (clever one-liner delivered precisely at the moment you wish you had something witty to say) without dragging it along with TMI (too much information).
Determining the effective complexity of a humor system obviously involves a good sense of humor as well as good judgment about what constitutes a good joke and what does not. This is, you must establish criteria that indicate when a bit is an "important" bit in a joke, a bit of regularity, and when it is "unimportant," a bit of randomness you bit off and then spat out.
In an engineered system, the important bits are those that have to take on particular values or else the system will not do what it's supposed to do. In an evolved system, such as those associated with the nuances of the human experience (i.e., joking around), it is less obvious which bits are important and/or funny and which ones are unimportant and/or not funny.
A simple criterion for figuring out whether a bit is important and so contributes to the effective complexity is to flip it and see what happens.
If flipping the joke around (set-up and punchline) has a significant effect, it is important to the joke and probably funny; if flipping the joke around has no significant effect, it is unimportant and most likely, no one laughed.
If a bit within a joke affects the joke's ability to survive and be retold, then that bit contributes to the effective complexity of the joke. A joke's important bits are those that affect the joke's repeatability in a significant way (Q. Hey, did you hear the one about the statistician? A. Probably.); the effective complexity of any humor system that exhibits purposeful behavior (laughter) can be similarly measured. Any bit that affects the ability of a humor system to attain its purpose (make people laugh or otherwise feel good) contributes to the humor system's effective complexity.
Of course, the definition of humor is to some degree subjective. But suppose we focus on joke that allows a humor system (a) get energy and (b) use that energy to construct copies of itself. Living systems devote most of their effort to eating and reproducing. However one defines life, any system that can accomplish those two actions has gone a long way on the road to being alive.
Some jokes, as they say, never die for this very reason. Once we identify as purpose those behaviors that enhance a joke's ability to generate laughs (get energy) and be easily retold (use it to reproduce), then measuring the effective complexity of a joke will someday teach us how to finally tell a joke that gets told time and time again for generations to come.
Humorous Adaptation of Seth Lloyd's
Effective Complexity (pps. 193 - 195)
No comments:
Post a Comment