Proximity and territory versus space in regional science

Empirical evidence tends to show that regional development occurs only in certain places. Therefore, one can ask the central question of the role of physical proximity in economic development processes. Economists try to answer this question through the concept of externalities, but this does not explain why certain externalities are linked with physical proximity. In this paper the author tries first to show why mainstream economists are not able to catch the problem of proximity and territory in their model. It is argued that they make a systematic confusion between physical space and abstract mathematical spaces. In the second part of the paper a tentative definition of a terrority is given and a first mathematical formulation of this concept is proposed. Starting from this definition, the author shows in the third part of this paper how a consideration of territory and proximity should change our understanding of economic development processes in general.