This from Colm Toibin's NYTBR review of a new biography of the writer Donald Barthelme:
By early 1963 he had enough stories for a book. He also had an aesthetic vision. In the second issue of [the arts magazine] Location he took part in a debate about the future of fiction in which Saul Bellow argued that the modern novel was "predominantly realistic" because "realism is based upon our common life." Barthelme countered that a "mysterious shift...takes place as soon as one says that art is not about something but is something," when the literary text "becomes an object in the world rather than a commentary upon the world."One could chew on this paragraph for a long time. What exactly is the relationship of a work of art to the world? A Mark Rothko or Franz Kline, say, might seem to satisfy Barthelme's notion of "an object in the world rather than a commentary upon the world" -- pure abstraction, nothing "realistic" to get one's teeth into. But then what is the source of the power of these works to move us deeply? Surely they draw that power from the world, if nothing else from human psychology which is part of the world. It is hard to imagine a work of visual art or literature -- or music, even -- that does not in some direct or indirect way make commentary upon the world. What gives a work of art its power is the energy that flows back and forth between the representation of a thing and the thing itself. When we look at a Mark Rothko we are looking at an object that is as realistic as a tree; we are also looking in a mirror at our own deepest selves.
Where does science fit in this discussion? Bellow's realism might seem to be the obvious answer. Science is a matter of consensus knowledge "based upon our common life." But a scientific theory also "is something," rather than being just "about something." Einstein said that we discover the deep truths of nature only by "free inventions of the mind." The equations of quantum electrodynamics, for example, sprang as much from a search for pure aesthetic symmetry as from any empirical imperative. They predict that the electron should have a magnetic strength of 1.00115965214; the measured value is 1.00115965219. The "mysterious shift" of which Barthelme spoke flows both ways, from theory to world and back again, like a spark leaping between two poles.
This is a subject that would require a book to properly explore, or at least a long summer night on the porch.