DOI: 10.4324/9780415249126-W047-2
Version: v2,  Published online: 2010
Retrieved July 07, 2020, from

1. Historical background

Theorists of vision have proposed various accounts of the nature of the processing responsible for our perception of size, shape and distance. Geometric models, popular among optic theorists in the seventeenth century, and suggested in some of René Descartes’ work on vision (particularly, his Sixth Set of Replies1641:–9), construe visual processing as a species of mathematical calculation (see Molyneux problem). Geometric models can therefore be seen as precursors of modern-day computational models of vision (see §§4–8 below). According to one geometric model, the visual system computes the distance of an object in the visual field from the angles at which the light from the object strikes each eye, and the distance between the two eyes. Some of the knowledge required for the calculation, including knowledge of the relevant mathematical theorems, was thought to be provided by innate mechanisms, rather than acquired from experience (see Innate knowledge). A significant defect of geometric models is that they failed to provide an account of how the requisite knowledge is made available to the visual system (including, for this calculation, the distance between the eyes, which is not itself perceived, and which changes as the subject grows) and how it is deployed in calculations that were presumed to be unconscious.

The philosopher George Berkeley, in his influential ‘Essay towards a New Theory of Vision’ (1709), questioned the psychological reality of the geometric models, arguing in effect that the information upon which the postulated calculations are based is not available to the visual system. Berkeley agreed with the geometric theorists that retinal information alone is insufficient to account for our perception of distance and size, but, consistent with his more general empiricism, he claimed that the process by which we acquire such knowledge is not a species of calculation based on innately specified information, but rather associative and learned.

According to Berkley, our ideas of distance and size, unlike our ideas of colour, are not really visual ideas at all. Whereas light reflected at different wavelengths affects the retina differentially, and so (the special case of metamers aside) gives rise to different colour sensations, light reflected from different distances does not. There is no characteristic retinal pattern associated with something’s being 10ft away. As Berkeley put it in his famous ‘one point’ argument, ‘distance being a line directed end-wise to the eye it projects only one point in the fund of the eye, which point remains invariably the same whether the distance is larger or shorter’ (1709: §2). Similarly for size: there is no characteristic retinal pattern produced by our looking at an object that is 6 ft3 (cubic feet) in volume. A larger object placed at a greater distance along the line of sight will have the same geometric effect on the retina. Our ideas of distance and size, Berkeley concluded, derive not from visual experience, but from touch and movement, from the time and effort it takes to make contact with objects, and from the way they feel in our hands. We can tell the distance and size of objects by sight only because we learn to associate visual cues, including sensations caused by the convergence of the eyes and the accommodation of the lens, with ideas originally derived from our tactual sense.

Central to Berkeley’s account of distance and size perception is the empiricist doctrine that there are no meaningful abstract ideas, that is, ideas not reducible to sensation (see Empiricism; Sense-data). He rejected the possibility that we might possess abstract spatial ideas that are shared by visual and tactual experience (see Molyneux problem). Later theorists of vision who do not share Berkeley’s epistemological and metaphysical assumptions have found his claim that our ideas of distance and size are derived from our sense of touch uncompelling, and recent work on object perception in infants has independently undermined this claim. Nonetheless, Berkeley’s discussion of the phenomena to be explained by a theory of vision shaped the field well into the twentieth century.

Citing this article:
Egan, Frances and Nico Orlandi. Historical background. Vision, 2010, doi:10.4324/9780415249126-W047-2. Routledge Encyclopedia of Philosophy, Taylor and Francis,
Copyright © 1998-2020 Routledge.

Related Articles