Abstraction is a method of inquiry into how seeing works

The argument

A second argument for this account treats abstract painting as inquiry rather than expression. Its materials are the components of visual experience themselves: hue, value, saturation, edge, scale, interval and surface. When a painting contains nothing to recognise, the viewer's attention has nowhere else to go, and what the work does to the eye becomes the whole of its content. Proponents argue that this makes the abstract work an instrument, a controlled situation in which the behaviour of seeing can be observed. The findings, on this account, are ones anyone can check. A colour looks different according to what surrounds it. A large field of one hue alters the appearance of whatever the viewer turns to next. Edges meeting at slight angles generate a sense of movement that no element in the painting actually has. Scale changes the sense of distance even where nothing in the work is a thing at a distance. None of this can be learned as readily from a picture of something, because there these effects are at work but are attributed to the depicted world rather than to the seeing of it. Proponents conclude that abstraction has a purpose in the strict sense: it is how an artist goes looking for what lies past the tangible surface of things, by removing everything that would let a viewer look through the painting to a subject instead of at it. The work is not a statement about the invisible but an occasion for noticing what vision does, which is itself not visible in any object.

Premises

Counter-arguments

Critics answer that this turns painting into an experiment it was never equipped to be. A controlled situation requires a question, a variable and some way of being wrong, and a painting supplies none of these: whatever the viewer notices counts as a finding. They add that the effects described are already well understood by those who study perception, so if abstraction's purpose were inquiry it would be a redundant and badly instrumented version of work done more rigorously elsewhere. What the account captures, on this view, is one of the ways abstract painting can be interesting to look at, which is an account of its appeal rather than of its purpose.

Rejecting the premises

[Rejecting P1] Hue, edge, scale and interval are the materials of all painting, abstract or not, so listing them does not single out abstraction. [Rejecting P2] A viewer given nothing to recognise generally starts recognising anyway, reading figures, landscapes and depth into the work, which undoes the controlled situation the premise needs. [Rejecting P3] Attributing an effect to the depicted world does not prevent it being observed, and painters studied these effects for centuries inside representational work. [Rejecting C] At most it follows that abstract work isolates certain perceptual effects, not that perceiving beyond the tangible is the purpose of the art.