The aim of the paper is to show that in the framework of philosophy of science it is possible to choose such an approach to the reconstruction of scientific laws and scientific explanation and also to the testing of scientific theories which enables us to avoid the well-known raven paradox as stated by C. G. Hempel. The paper starts with an analysis of Hempel’s approach to the so-called “purely qualitative predicates.”Then it shows how this approach, which has its origins in the perceptualist phi- losophy, led Hempel to a completely distorted view on the structure of scientific laws and scientific explanation, that in turn led him to a distorted view on the testing of scientific theories. As an alternative the paper proposes an another view on the structure of scientific laws, scientific explanation and introduction of theoretical magnitudes using the method of definition by abstraction. Consequently, it also proposes an alernative view on the testing of scientific theories which eliminates the raven paradox.
Definition by abstraction, Explanation by gradual concretization, Hempel, Paradox of the raven, Structure of scientific law