Stalking the beast: Art history as asymptotic exercise

Art historians produce discourse, ¹ working both syntactically and dialectically. In the syntactical mould they fit the object of their study, said by their predecessors or themselves to be art, into a context formerly supposed to be works perceived as similar. This context is now, increasingly, understood to be the culture or society that has a generative, if not an explanatory, role in the creation of the object. Art historians work dialectically or, as is sometimes said today, 'con testationally', first because objects (especially those perceived to be of high quality) do not go gently into cultural contexts nor readily lend themselves to social or anthropological explanation;² and, secondly, because their readings are reactions to what their predecessors have said about a work. Whichever mode — the syntactic or the dialectic — is operative, art history, like all human enterprise, is a contingent activity. On the one hand, it makes inferences about a work from the society that produced it, or about a society from the works that constitute it. On the other hand, its startingpoint is the inferences that others have made and its end the generation of responses that are dependent on, even while often being responses to, a published reading. Thus art history, like all scholarly enterprise, advances and only a positivist would doubt that the game is worth the candle.