September 1970
Memo On the Visual Perception of Tangible and Intangible Things
J. J. Gibson, Cornell University
The World Wide Web distribution of James Gibson’s “Purple Perils” is for scholarly use with the understanding that Gibson did not intend them for publication. References to these essays must cite them explicitly as unpublished manuscripts. Copies may be circulated if this statement is included on each copy.
Tangible things are almost always visible but visible things are not always tangible. The only invisible but tangible surfaces are large smooth plates of transparent material like glass, which do not exist in nature, whereas there are many visible but intangible things in the natural environment, such as the sun and stars, the sky itself, rainbows, sunsets, mirages, reflections, the beams of light in a dusty atmosphere, and the images in a pool of water.
The existence of things that can be seen but not touched has puzzled observers from the beginning of observation itself. The belief in spirits, ghosts, and Gods comes in part from this fact. Things that can be seen and touched do not puzzle us; we take their perception for granted, but the others are vary puzzling and we strive to explain them. The history of physics and physical optics is largely a history of our attempts to explain intangibles like rainbows, reflections, light-beams, and images (Minnaert, 1954; Bragg, 1933).
The optical explanation of how we see intangibles is highly successful and is nowadays widely understood. They are not ghosts or Gods but products of light. We know about light rays and their reflection, refraction, diffraction, and scattering. But the explanation of how we see tangibles is not at all clear and is not understood. Every student of high school physics knows why a rainbow looks as it does but no one knows why the ground under one’s feet looks as it does. Why should this be so?
It is probably because we have a persuasive physiological theory of how we see radiant light and spectral color, the theory of sensations of brightness and color caused by the stimulation of photoreceptors, but no adequate theory of how we see illuminated surfaces and the layout of such surfaces. We have assumed that the perception of surfaces is constructed out of sensations of lightñthat the elements of perception are sensations. We have assumed that an illuminated surface is composed of a dense set of luminous star-points, or bits of the rainbow. And, indeed, something like this is what most physicists believe. We take the phenomena of light and color in our environment as a model for the phenomenal surface s of our environment. Because we understand the curious phenomena of meteorology and astronomy we have tried to understand the ordinary and familiar phenomena of terrestrial objects in the same way.
The intangible and often ephemeral things like rainbows, sunsets, light-beams, the play of sunlight, and fires are unlike the surfaces and objects of the world in that the perceptions are not invariant despite varying sensations of light (Gibson, 1963). The perception changes as the light changes whereas the perception of an object, a layout of reflecting surfaces, is more or less constant as the illumination changes. The colors of the former are the spectral colors of light whereas the colors of the latter are the pigment colors of surfaces. The intangible phenomena generally do not have a definite surface texture whereas the tangible phenomena do. It is vastly more important for animals and man to see the composition of surfaces and their layout for what they are than to see the rainbow for what it is. To see the rainbow as a bridge in the heavens does no harm so long as one does not have to walk on it.
The unsolved problem of perception, then, is how we see the tangible things of the world. A persistent explanation, ever since Bishop Berkeley’s New Theory of Vision (1709) has been that the sensations aroused by these things are supplemented by memories of touching them whereas the sensations aroused by intangibles are not. The underlying assumption is that the perception of things by touch (mechanical stimulation) is self-evident while the perception of things by vision (light stimulation) is not. But this, of course, is not true.
The alternative is to assume that the tangibility of some things and the intangibility of other things is visible, and to search out the information in ambient light that makes it possible to distinguish the two kinds of things. This is what ecological optics tries to do. The invariant information in an optic array for perceiving surfaces and their layout is distinct from the changing variables in the ambient light that enable us to see the curious phenomena that result from regular reflection, refraction, diffraction, polarization, and scattering. When these phenomena are isolated and controlled in the laboratory they reveal the laws of radiant energy. But they are not the basic phenomena from which an understanding of surface perception can be derived.
I said at the beginning that the optical explanation of how we see these intangibles is very successful. But even so there is something strange about them which has been a source of confusion in psychology and philosophy: they do not look like what they really are, or, more exactly, the source of the colors and forms in the light is not revealed. A textured surface with its edges and corners looks like what it is; the layout is revealed. The refraction of light by raindrops, however, is not evident in the rainbow and the differential scattering of wavelengths is not revealed by the blue vault of the sky. The sun looks a bit like a chariot of fire and the stars look like chinks in a roof. The wondrous discoveries of physics and astronomy have led some observers to mistrust their vision of ordinary terrestrial things, to celebrate an unseen world, and even to believe the nonsense that things are never what they seem.
References
Bragg, W. The universe of light. Macmillan, 1933.
Gibson, J. J. The useful dimensions of sensitivity. Amer. Psychol., 1963, 18, 1-15.
Gibson, J. J. Chapter on ecological optics in The senses considered as perceptual systems. Boston: Houghton Mifflin, 1966.
Minnaert, M. The nature of light and color in the open air. Dover, 1954.