Cartesian Optics and the Geometrization of Nature

Abstract

In His Dioptrics, Descartes puts forth a mechanistic theory of perception?mechanistic, in the sense that matter (for Descartes, mere extension) and motion (a mode of extension) are its funda mental concepts. The most striking component of this theory is Descartes' very careful treatment of the visual perception of distance as utterly different in kind from the visual perception of color. For in the explanation of distance perception, Descartes invoked the use of a natural geometry, innate in the percipient. This natural geometry, he claimed, processes the data of sensation so as to produce a judgment about natural spatial relationships. Significantly, Berkeley, in his Essay Towards a New Theory of Vision, leveled a sustained attack on just this geometrical theory of distance perception.1 At first glance it may seem, as it did to Berkeley, that Descartes' geometrical theory is produced by a simple error: namely, by the idea that a physiological optics provides an adequate description of the psychological processes of judging distances.2 In truth, this is the weakest of Berkeley's objections to Descartes' theory. Obviously we do not see the angles and lines of convergence when we focus on a distant object, nor are we aware of