Most ancient theorists, from Plato and Aristotle on, saw jokes as an expression of superiority, humor as "mockery and derision," and laughter, therefore, as "a slightly spiritualized snarl." This is fine, argues Holt, for a range of unpleasant jokes at the expense of the other races and religions, of the poor or otherwise unfortunate ("How did Helen Keller burn her fingers? She tried to read a waffle iron"). It also works for the kind of temporary triumph we might feel if we saw "Bill Gates get hit in the face with a custard pie." It does not seem up to the task of explaining why we might laugh at puns, for example. You might conceivably argue that we are there enjoying a sense of superiority over language itself. "But now," as Holt observes, "the superiority theory has become elastic to the point of meaninglessness."
More popular among modern theorists is the "incongruity theory" of joking, which sees humor and laughter stemming from the inappropriate mixing of categories or registers of meaning ("Work is the curse of the drinking classes," as Oscar Wilde quipped). Or, as Kant put it more opaquely in the Critique of Judgment, a joke arises "from the sudden transformation of a strained expectation into nothing." Kant's own example of this was a story of an Indian who looked astonished when an Englishman opened a bottle of beer and the contents frothed out. When asked why he was so surprised, the Indian replied, "I'm not surprised at its getting out, but at how you ever managed to get it all in." The problem here is not that jokes trade on incongruity, for almost all of them in some way do. It is why incongruity should give us pleasure, and why some sorts of incongruity prompt laughter and others (such as Oedipus' parenthood) do not. Besides, as Holt goes on to ask, why should the reaction either to incongruity or to a feeling of superiority be "a bout of cackling and chest heaving"?