The Perception of Pictures - Part 2 - A Wished-For World?


[Bosch - The Haywain]


To begin this discussion of Gibson's theory of perception and pictures, I want to look at a 1960 essay published by Gibson in Daedalus’ special volume on ‘The Visual Arts Today,’ titled “Pictures, Perspective, and Perception.” [PPP] In PPP, Gibson describes his understanding of pictures in the theory of ecological optics. I have already mentioned much of what he describes, including the new notions of the ambient light array, the optic array at a station point, and the generalized geometry of perspective. In his initial discussion, Gibson also repeats his point about the irrelevance of the retinal image (or any other image) to visual perception. At the end of this summary he states:

The human retinal image is only one stage in the human process of seeing. What the instrument makers, the photographers, the visual educators, and the artists in their own way are trying to do is to aid or enhance the process of seeing. By ‘seeing’ I mean understanding, not the special process of considering one’s sensations or the special act of seeing in perspective. [219]

It is worth noting that this assumption of the importance of the retinal image is almost ubiquitous in art theory and history. John Ruskin presents a nice version of the unexamined assumption in The Elements of Drawing:

The whole technical power of painting depends on our recovery of what may be called the innocence of the eye, that is to say, of a sort of childish perception.

Everything you can see, in the world around you, presents itself to your eyes as only as an arrangement of patches of different colours variously shaded.


The Impressionists often described their intention to capture just that image -- the fleeting or momentary experience of color and light. The Cubists and Abstract Expressionists often cited their intention to turn away from such a 'sensory image' and focus instead on the 'form' or structure of the object or scene depicted. Gibson would say that all of this talk is misguided. As we will see -- all artists only ever depict invariant properties of the ambient light array -- to do otherwise would be to make something that could not be looked at.


Gibson also discusses his theory of ‘how perception works,’ describing the direct pickup of invariant properties and the dependence of the pickup upon structured stimulation and “the interests of the individual observer” [220]. I haven't discussed these theoretical notions in detail, but it should suffice to say that Gibson conceives of the information in the world as unlimited and available for selection by an individual. That selection will follow the interests of the individual -- interests as low-level as 'good-for-eating' and 'good-for-using-to-cut' to interests as high-level as 'those-are-my-shoes' and so on.

He initially characterizes the artist as a sort of specialized perceiver, one “who pays special attention to the points of view from which the world can be seen, and one who catches and records for the rest of us the most revealing perspectives on things” [220]. He also characterizes the perception of objects depicted as necessarily second-hand. For Gibson, perception can be mediated in various dimensions, and one of them is the immediacy of the perception. Perception of objects depicted will always be second-hand perception and can never approximate or transform into first-hand perception, no matter how skillful the depiction. Gibson returns to this later when he refutes the traditional view that there is a sliding scale between artistic representation (illusion) and reality.

So, what is a picture?

A picture is a human artifact which enables another person to perceive some aspect of the visual world in the same way that the artist, the maker of the artifact, has perceived it. This definition is intended to apply to any picture – any drawing, painting, photograph, motion picture, or television image, whether representational or not – so long as it is intended to be looked at.

Concretely, a picture is always a physical surface, whether of canvas, paper, glass, or some other substance, which either reflects light or transmits it. It is an object, in short, commonly a flat rectangular one, but what is unique is the light coming from it. The surface has been treated or processed or acted upon in such a way that the light causes a perception of something other than the surface itself. It delivers a sheaf of light rays to a station point in front of the surface, rays that contain information about quite another part of the world, perhaps a distant world, a past world, a future world, or a wished-for world; a delicious or a horrifying world; but at any rate some part or aspect of a world which is not literally present at the station point. [221]

My first potential problem with Gibson begins to take shape here – this description of what a picture is seems to presuppose that the object represented was perceived or is perceivable. Combined with his description of what the artist does, it seems like art is, at its best, a faithful and natural depiction of an actual visual experience. Gibson later amends these statements, but the question arises – what if no artist could have ever seen what they depicted? How do they make such a picture, and what sort of information does it contain? These questions can be lumped under the more general question of how Gibson can account for artistic invention, if at all.

Next -- on pictures that are faithful to the things depicted!