인과의 조종가능성 이론과 간섭

Abstract

Manipulability theories of causation have considerable intuitive appeal and are popular among scientists. The central idea of manipulability theories of causation is that causal relationships are relationships that are potentially exploitable for purposes of manipulation and control. That is, if C is a cause of E, then if someone can manipulate C in the right way, this should be a way of manipulating or changing E. In order to clarify the central concept of this theory, manipulation, recent developments in manipulability theories of causation concentrate on the concept 'intervention'. Pearl suggests atomic intervention, which brings about a surgical change, using equations and directed acyclic graph. It controls the variable in question leaving aside the rest of the causal relations. Woodward claims that Pearl's characterization of intervention is merely calculational different from his own which aims to clarify the meaning of causation. In this paper, I argue that Woodward's characterization of intervention is not different from Pearl's. I show that Woodward's characterization of intervention materializes Pearl's concept of intervention.