In the theory presented in this paper, humor occurs when a perceiver views a situation simultaneously as being normal and as constituting a violation of the ``subjective moral order.'' The ``subjective moral order'' is defined as the set of principles which an individual both has an affective commitment to and believes ought to hold. Individuals' varying degrees of attachment to the principles violated in different situations are at least in some cases independently verifiable, and are found to be consistent with the theory in all cases examined here, thereby making the correlations stated by the theory (between personal moral attachments and perceived humor) substantive and non-circular. The three conditions of the theory were shown to be both necessary and jointly sufficient for humor perception.
The paper then explored the ambiguity of something being ``not funny'', and developed a three-level scale of violations of the moral order was developed, which ranges from ``no violation'', to ``funny violation'', to ``threatening violation'', depending on the presence or absence or relative intensity of N, normal interpretations and V, perceived violations. The formal logic of these affective evaluations was explored. The intensity of perceived humor was explained by a ``more is better'' principle. Humor's role in transforming the affective evaluation of situations was derived from the theory. Also the irrelevance of the order of presentation of N and V in joke-telling was pointed out.
A number of properties which form the basis for various theories of humor were explored and put in their place, including absurdity and incongruity, timing and surprise, aggression, superiority, and comprehension difficulty. Two contemporary partial theories of humor, due to Gleitman and Raskin, were shown to be derivable as special cases of the present, more general, theory. Then a number of humor-related phenomena were explored, with a focus on those where interpretations vary: offensive or sick jokes and the humor of children (elephant jokes) and babies (peekaboo). Then, it was shown how various forms of laughing fit into the theory, including tickling, exaggeration, puns, satire, and ridicule. Biological evidence consistent with the theory was adduced, including first, the isomorphy between the organization of the physiology of tickling and the elements of the present theory of humor, and second, a comparison with the physiologically similar responses of crying and gasping in pain. The former can be seen as isomorphic to humor in its internal organization: Tickling involves simultaneous normal and painful (tactile) perceptions. And the latter, crying and gasping in pain, share an affective element of pain -- along with humor, according to this theory, which thereby provides an explanation for the similarity of the physiological responses. Finally, because the theory determines certain properties of shared humor, a rule of inference was derived for exploring the thoughts and feelings of people who do -- and do not -- laugh at particular situations. Thus the present theory shows both how humor communicates and also, when people don't share one another's perceptions of humor but want to understand each other, it shows how they can figure out what is in each others' minds. The theory can be applied as a fine-grained diagnostic tool for learning about different individuals and communities of individuals. Thus it may be useful in ethnographical or anthropological investigations of people and of systems of moral commitments, both for academic investigators interested in such topics, and for lay people in a complex world, who simply want to understand what makes those crazy people tick, and why they laugh at those stupid jokes.