Models in scientific practice

The main topic of this paper is the roles that models play in scientific practice. Although mechanistic models have been used since the emergence of Newtonian classical mechanics in the 17th century, philosophers of science have disagreed about their importance for scientific theories. Logical empiricists between the two wars strove for a rational reconstruction of science and the construction of a normative methodology of science in which models had no place. During the 50s of the last century, the Copernican turn in the philosophy of science took place, shifting the focus from rational reconstruction to actual scientific practice in which models are widely employed. The paper shows the ways of their construction, basic characteristics (mediation between abstract scientific theories and concrete physical phenomena, interpretative and representative function) and cognitive significance (descriptive adequacy, explanatory and predictive power, contribution to understanding and research). A number of examples have shown that scientific models are so diverse that it is almost impossible to form a unifying theory about them - they do not share any essential common properties, but are connected by a network of Wittgensteinian ”family resemblances”.