It is always more attractive to examine scientific debates in terms of differences in types of discourse rather than examining theories. Independent of anything else, the following is the most significant feature of a type of discourse: It provides a framework where it becomes legitimate to pose certain kinds of questions and to discuss a particular class of phenomena. This legitimizing framework provides the possibility to discuss a whole new set of issues and, at the same time, it creates all those problems that make it difficult for the scientific community to have a consensus both about the formulation of the new questions as well as about the proposed answers. The success of a new theoretical approach has always been the result of a curious mix of persuasion and proof. Though absolutely essential, proofs were by no means sufficient. Persuasion, however, became indispensable whenever the novelties introduced by a new type of discourse were at stake. "To persuade" meant two things at the same time: consensual activities and legitimizing procedures.