George Lakoff and Mark Johnson's theory of metaphor (1980, 1999) provides a basis for describing everyday cognitive structures using linguistic models and thus, making it possible to uncover both individual and collective patterns of thought and action. Lakoff and Johnson have not, however, developed a workable system for carrying out qualitative research. This paper outlines the fundamentals of this approach and proposes a procedure for the reconstruction of metaphorical concepts. As is normally the case in qualitative research, such guidelines can only ever represent the interplay between the ability of the researcher to understand the sense of things and the rules of the methodology. An overview of the typical interpretations that a metaphor analysis allows is also given. (Rudolf Schmitt)