Arthur Danto, "The Artworld"
Danto suggests a picture of change in artistic theory which resembles change of scientific theory. 'Discovery' of a whole new class of artworks is akin to the discovery of a whole new class of facts. In science, such a discovery is often met with a conservative extension of existing theory, one which attempts as best as possible to explain existing phenomena while also accounting for the new ones. This is paralleled in art by the development of a new theory of art which accounts for both old artworks and the newly 'discovered' ones.
A change of this sort occurred with the acceptance of postimpressionist painting. Under the standards by which older artworks were judged to be artworks ('imitation theory', or 'IT', also mimesis: the idea that art is imitation of reality), postimpressionist painting failed: it was not sufficiently mimetic. The acceptance of postimpressionist painting required theoretical changes. Danto subsumes these changes (while admitting the historical complications) under one theory, 'RT' (probably 'reality theory', though he only makes this implicit).
"According to it, the artists in question were to be understood not as unsuccessfully imitating real forms but as successfully creating new ones, quite as real as the forms which the older art had been thought, in its best examples, to be creditably imitating." This gets at what Danto sees as the substance of RT, but consider also the following, cribbed from H.H. Arnason's History of Modern Art:
Impressionism also marks the beginning of a movement in art, in which a group of artists began to assert (half unconsciously) the identity of a painting as a thing, a created object in its own right, with its own structure and laws, apart from any it might have as an imitation of the world of man and nature.
This indicates in part the relevance of the 'reality' portion of the name. The acceptance of postimpressionist painting involved a turn away from looking at paintings as solely imitative, which in a sense ignores their reality as objects, because of the greater interest in the real objects they represent. It is not as if paintings were never recognized as real objects (canvases covered with paint) before postimpressionist painting; indeed, consideration of the formal elements of painting was certainly involved in IT. But under RT, it became acceptable to look at paintings more on the basis of their formal properties (that is, their properties as real objects), and less on the basis of the quality of their imitation or representation.
Danto claims that it is in terms of RT that we must understand contemporary art. At the very least, it provides an example of how theory makes art possible: if it weren't for RT, postimpressionist painting would not be considered art. It is important to make a distinction here between 'it would not be possible to paint these paintings', or 'it would not be possible for these paintings to exist', and 'paintings with certain sorts of properties would not be counted as artworks'. Though it is true of some paintings that they would not have been made if it were not for the development of certain ideas about painting, that development possibly motivated by earlier paintings which may or may not have had an explicit theoretical basis prior to their painting, Danto is not talking about the existence of these paintings, as a contingent matter. He is talking about how paintings of that sort are considered, and whether or not they are taken up into the artworld.
Now Danto takes up the question of how it is that one can mistake an artwork for a real object. His paradigmatic examples are genuine beds made by Robert Rauschenberg and Claes Oldenburg. Asking how one can mistake them for real beds, instead of artworks, is equivalent to asking what makes them artworks. To answer his question, he introduces the "'is' of artistic identification". This 'is' is used in sentences like "That a is b" where a is some specific physical property or part of an object. It is a necessary condition for something to be an artwork that is can be the subject of such a sentence.
Danto's first attempts at explaining the 'is' of artistic identification all seem to be examples where the paintings are representational - but it seems that his example, "That white dab is Icarus", is meant to indicate that the 'is' is not solely the 'is' of representation, in the sense that the subject of the sentence visually resembles (in a strong way) b. We can imagine a painting of a seascape, with a bright sun overhead, and a tiny white dab of paint beneath the sun: Icarus falling. This may not be apparent from the painting itself, but what if it's called The Death of Icarus? Then (though there may be argument over how much this is warranted, since it's so hard to tell from just the dab that it represents Icarus) it would seem quite reasonable to most people discussing the painting to say "That white dab is Icarus": the dab is identified with Icarus.
Danto presents an example in which the 'is' of artistic identification is relied heavily upon in the viewing of two identical paintings (each a white rectangle with a black line through the center). His comments indicate how use of the 'is' of artistic identification is guided by (backed by, supported by, constrained by) theory, understood in the loosest sense possible. At its extreme, this theoretical support allows a hardcore abstractionist to refuse to identify his black-lined rectangle with anything - "there is nothing there but white paint and black". The difference between his statement, and that of a philistine who utters the same thing but does not see the painting as an artwork, is that the abstractionist is employing the 'is' of artistic identification, and the philistine is not.
The different rows of the style matrix are not as legitimate as one another, in light of the fact that which ones the artworld will accept as expansions of the style matrix is a contingent matter, dependent on the existing style matrix, and also on how much of it different segments of the artworld are willing to accept.
Danto's black-lined rectangles point to a problem with his claim that the 'is' of identification is at work behind all art's acceptance into the art world. I think his account of how Brillo boxes or the rectangles could be artworks is plausible. But their acceptances seem to me to be in a very significant way more theory-bound than the acceptance of postimpressionist painting or the identification of a white dab with Icarus, because the acceptances of the former rest on more abstract ideas - ones more divorced from the formal properties of paint on the canvas, or of meaning as it works on a more casual level.