Very roughly, if I understand him rightly, the difference between a representation and an object is that the former is intended. The difference between a representation and an artistic representation is that the latter is artistically intended, meaning, the artist, with his knowledge of the artworld and art history, intends to make a work of art. So Andy Warhol's Brillo Box is a different thing from the Brillo box in the supermarket even though they look indiscernibly the same. In order for the viewer to grasp the artistic intention behind the art work, the viewer must know or come to understand the meanings that the artist infused into the work. The structure of the artwork is thus very close to the structure of a metaphor.
So the artwork is constituted as a transfigurative representation rather than a representation tout court, and I think this is true of artworks, when representations, in general, whether this is achieved self-consciously, as in the arch work I have been discussing, or naively, when the artist simply happens to vest his subject with surprising yet penetrating attributes. To understand the artwork is to grasp the metaphor that is, I think, always there. (172)
Danto draws out many implications from the last statement, one of which has to do with the limits of art criticism:
The first is that if the structure of artworks is, or is very close to the structure of metaphors, then no paraphrase or summary of the artwork can engage the participatory mind in at all the ways that it can; and no critical account of the internal metaphor of the work can substitute for the work inasmuch as the description of a metaphor simply does not have the power of the metaphor it describes, just as a description of a cry of anguish does not activate the same response as the cry of anguish itself. (173)
The task of art criticism is not only to interpret the metaphor but to provide the viewer with the necessary information to respond to the artwork, information lost to time or unknown due to place.
Criticism then, which consists in interpreting metaphor in this extended sense, cannot be intended as a substitute for the work. Its function rather is to equip the reader or viewer with the information needed to respond to the work's power which, after all, can be lost as concepts change or be inaccessible because of the outward difficulties of the work, which the received cultural equipment is insufficient to accommodate. It is not just, as is so often said, that metaphors go stale; they go dead in a way that sometimes require scholarly resurrection. And it is the great value of such disciplines as the history of art and literature to make such works approachable again. (174)
The argument about art leads Danto to expound a view about man.
My view, in brief, is an expansion of the Peircian thesis that "the man is the sum total of his language, because man is a sign. . . . it is not merely what a man represents, but is the way in which he represents it, which has to be invoked to explain the structures of his mind. This way of representing whatever he does represent is what I have in mind by style. If a man is a system of representations, his style is the style of these. The style of a man is, to use the beautiful thought of Schopenhauer, "the physiognomy of the soul."