The Shape of the Signifier

Hence the difference between losing a game and losing an argument: you don't lose at chess when you are convinced that you cannot move your king out of check; you lose when, whatever your views, you cannot, within the rules of the game, move him. The point can be put more generally by saying that in any game the players' moves have a force that is utterly undetermined by their beliefs about them. Beating someone at chess has nothing to do with changing his or her mind. And it can be put more generally still by saying that just as two players in a game cannot be described as disagreeing, two players playing two different games can't be described as disagreeing either, not because they have the same beliefs but because, once again, their beliefs are irrelevant. Chess isn't a set of beliefs; it's a set of rules. That's why the redescription-in philosophers like Richard Rorty and Jean-Francois Lyotard-of people who have different beliefs as people who are playing different "language games" amounts to a repudiation of the idea that people actually have any beliefs. The point here is not a skeptical or even a relativist one about the possibility of having true beliefs; postmodernism, properly understood, is required to be just as skeptical about the possibility of having false beliefs as it is about the possibility of having true ones. Book jacket.

Walter Benn Michaels is Professor of English at the University of Illinois, Chicago