Assuming an essential difference between scientific data and phenomena, this paper argues for the view that we have to understand how empirical findings get transformed into scientific phenomena. The work of scientists is seen as largely consisting in constructing these phenomena which are then utilized in more abstract theories. It is claimed that these matters are of importance for discussions of theory choice and progress in science. A case study is presented as a starting point: paleomagnetism and the use (...) of paleomagnetic data in early discussions of continental drift. Some general features of this study are presented in formalized language. It is suggested that the presentation given is particularly suited for a semantic conception of theories. Even though the construction of scientific phenomena is the main topic of this paper, the view presented here is more adapted to realism than social constructivism. (shrink)

The general thesis that science is essentially a problem-solving activity is extended to the development of new fields. Their development represents a research strategy for generating and solving new unsolved problems and solving existing ones in related fields. The pattern of growth of new fields is guided by the central problems within the field and applicable problems in other fields. Proponents of existing research traditions welcome work in new fields, if they believe it will increase the problem-solving effectiveness of their (...) tradition. Correspondingly, researchers in new fields will graft their work onto established traditions, if they believe it will augment the problem-solving effectiveness of their work. The above claims are defended through using the development of paleomagnetism as a case study. (shrink)