Abstract

The arguments presented in this thesis aim to show that a Kantian model in which a theoretical framework is constitutive of empirical reality denies the possibility of a full grown experimental science. In contrast, a logical empiricist model according to which empirical reality is described by means of a theoretical language such as Euclidian geometry or Newtonian mechanics allows for a possible disagreement between experiment and theory. One of the main strengths of this model consists in its ability to handle cases in which theory is shown to be false in respect to empirical reality. However, in order to achieve this compatibility with experimental science, I argue that we must abandon the assumption that there is a single, unified language of science and replace it with the more general assumption of a plurality of languages. Under this new formulation, a perfect unity of science cannot be granted solely by a coherent formulation of all scientific knowledge in a unique language of science, but must be complemented by the common reference of all scientific theories to the same empirical reality