The aim of this paper is an analysis of the different standpoints of Parsons and Schutz concerning Weber’s suggestion that sociological explanations have to include the subjective point of view of the actors, the Cartesian Dilemma that the actor’s consciousness is not accessible to the researcher, and the Kantian Problem that theories are necessary in order to interpret sensory data, but that there is no guarantee that these theories are true. The comparison of Schutz’s and Parsons’s positions shows that Parsons’s methodology is na?ve and unsuitable for a sociological analysis. But although Schutz’s methodological standpoint is much more reasonable, it is also problematic, because it excludes highly abstract social “facts” such as social systems from the research agenda. Parsons can deal with such highly abstract facts, despite the drawback that with his methodology the truth content of theories cannot be judged.