The Encryption Theory: Studies in Brazil and the United States Testing an Evolutionary Explanation of Humor as an Honest Signal

There are few behaviors in the human repertoire as consistently and reliably universal as humor. The ubiquity of this phenomenon suggests some kind of species-typical architecture, and recent years have seen a number of evolutionary hypotheses for its origins. This dissertation presents a novel theory of humor which, though it shares features with some existing theories of humor, proposes a new mechanism through which humor operates, as well as an ultimate function. The Encryption Theory of humor proposes that a necessary component of the structure of humorous production is the presence of multiple, divergent understandings of speaker meaning, some of which are dependent on shared access to implicit information on the parts of both speaker and audience. Only with access to this background knowledge can an audience member "decrypt" the implicit understandings, which also entails the inference that the speaker shares that access. This provides a channel for the honest signaling of personal features, which is proposed to have evolved to aid within-group assortment for long-term interaction partners such as friends or mates. The information encrypted by humor may be propositional knowledge, personality traits, attitudes and values, or personal experiences. Experimental studies conducted in the U.S. provide support for the hypothesis that humor depends on this encryption-decryption hypothesis. Ethnographic and experimental studies conducted at a collective farm in rural Brazil provide further support not only for the encrypted structure of humorous productions, but also for the relevance of humor in the actual social assortment found there, as it reflects dyadic similarity in personal features like knowledge, preferences, or experience. Further evidence collected in Brazil is consistent with Encryption Theory's prediction that evaluations of the senses of humor of others is a subjective consideration, dependent on the preferences of the evaluator, and not an objective assessment; these predictions are also consistent with a critical review of the existing literature. Finally, an acoustic analysis of spontaneously produced conversational humor shows a predicted lack of prosodic marking, reflecting the strategic interests of humor production. Implications for the future study of humor as an honest signal of within-group individual variation.