Psychologists usually work out their research programmes obeying to a principle called "Morgan's canon."  It says that a higher mental faculty should not be postulated to explain a behavior if the behavior can be explained by postulating only a lower mental faculty. However a new research programme in ethology has arisen that seeks to explain nonhuman animal behaviors in terms of anthropomorphism. The famous ethologist Francis de Waal argues that anthropomorphism is a modest programme because of evolutionary parsimony. So the question we should ask here is whether anthropomorphism is a better scientific programme than anti-anthropomorphism, or anthropodinial. In this paper, I criticize de Waal's argument for anthropomorphism. I show that his argument is based not on evolutionary parsimony but on overall similarity, and that whether his argument was based on overall similarity or on some kind of parsimony, both arguments would fail. Then I discuss which programme is better, anthropomorphism or anthropodenial from scientific point of view. In modern science, statistical null hypothesis testing is one of the most common methods. By analyzing it linguistically and historically, I argue that anthropomorphism has a greater sin than anthropodenial.

特集 : 坂上貴之教授 退職記念号#寄稿論文