Philosophy of History XXVIII: Arthur Danto’s Narrative and Knowledge

Narrative structures penetrate our consciousness of events in ways parallel to those in which … theories penetrate observations in science. –Arthur Danto

Arthur Danto (1924–2013) was an American artist and analytic philosopher who is best known for his work in the philosophy of aesthetics. He also made an important contribution to the philosophy of history, however, in his ground-breaking text, The Analytical Philosophy of History (1968), which he later expanded to Narration and Knowledge (1985).

Danto defined his project in opposition to two powerful trends in mid-century American historical thought. The first was Carl Hempel’s application of logical positivism to historical methods (the subject of an earlier article); the second, skepticism about the possibility of historical knowledge offered by Bertrand Russell and Charles Beard. Carl Hempel had affirmed that historical explanation was possible, but required that it be given in terms of laws, boundary conditions, and logical deduction, else it was not an explanation at all, but only an “explanation sketch.” As it turned out, Hempel’s desire to extend the logical positivist project from science to history was premature, as the logical positivist program proved untenable even for scientific explanations, for which it had originally been formulated. Nevertheless, it stood then (and for some, stands now) as an indictment of historical knowledge that it is not given in terms of general laws or logical deduction, and Danto felt compelled to answer this challenge. Russell’s and Beard’s positions ran to the opposite extreme, challenging the possibility of historical knowledge as such. After all, Beard argued, we cannot witness the events of the past, and have only a fragmentary and self-interested record of them. How can we trust works written on such a basis? Indeed, Russell had argued, how do we know these records are about real events at all? Suppose the world were created five minutes ago—would we know it?

To Russell’s criticism, Danto replied that it was not so much a complaint against historical knowledge as a complaint against knowledge generally. It has to be admitted that it is possible that the world was created five minutes ago, is the dream of an oyster (or a computer simulation, as the more up-to-date version has it), or is in some other way completely different than we conceive of it, but so what? Nobody thinks this kind of skepticism invalidates mathematical or physical knowledge, so why does history come in for special attention? Beard’s challenge required a more careful response. Suppose, Danto argued, that we had a complete, and a completely accurate, record of events—just the sort of thing Beard asked for, an Ideal Chronicle. If we had such a record, we would still have to interpret it, for on its own it would simply be a mass of unstructured information. A historian of New York City would certainly benefit from having a camera on every corner and in every room, so he could consult a completely accurate record; but that would not be a history of New York City, for he would still have to select from, and abridge, that information in order to tell us the kinds of things we expect to find in a history book. It is certainly an added burden that the historian has to decide what actually happened before he can perform that selection and abridgement, but plainly it is not the whole of the historian’s task.

That abridgement, Danto argued, has to be given in the form of a “narrative sentence.” The structure of this sentence is necessarily temporal, for instance, “Julius Caesar conquered Gaul.” Such a seemingly simple sentence, of a type routinely found in historical descriptions, is actually quite complex. Caesar didn’t conquer Gaul all by himself, and he wasn’t conquering it every minute while he was there either. He had lots of help, and he also spent a great deal of time eating, sleeping, writing letters, and other mundane activities. The Ideal Chronicle would give us these activities as they happened, one by one, but the historian removes the extraneous material and tells us what we want to know. “What we want to know” will be different for different observers, to be sure; some may be more interested in everyday life in the Roman army, others in the heroic resistance of the Gauls, or in the political intrigue between Caesar and the Senate, etc. But however that interest is constituted, an abridgement will have to take place, and it will have to be done with reference not only to the event itself, but to subsequent events. In other words, it will have to be placed within a temporal structure.

This can be readily appreciated if we consider the changing significance of Caesar’s conquest of Gaul. At the time, its significance was that Caesar wanted military glory and plunder. He may have thought he was bringing Roman civilization to the barbarians, but he certainly didn’t think he was helping to found French civilization, for there was no such thing at the time. Nevertheless, it appears in retrospect that he was helping to found French civilization, for he introduced the Latin alphabet, Roman law, and urban life to areas in which they were relatively unknown. The current enthusiasm for environmental history, which is quite explicitly connected to anxieties over climate change, will no doubt see Caesar’s significance quite differently, for with urban life came deforestation, intensive agriculture, marsh drainage, and other alterations of the landscape. Eventually, environmental history will run its course and another interpretation will come about.

So the significance of events is derived from its place in a temporal structure, and is thus only knowable in a final sense when all temporal structures are complete—which is to say, never. It is not, then, an incidental, but a necessary, feature of historical explanation, that it is constantly changing, for it is a consequence of the temporal structure of historical reasoning. Historical projects that attempt to cut out or circumvent this temporality—such as the quest for predictive laws of history, which in effect attempts to write a history of events before they have happened, or the pragmatist account, which attempts to reduce statements about the past to statements about the present—are fundamentally misconceived. The significance of the events that historians describe must change as its relation to other events changes, and these are always being added to by the progress of time. Thus it is truly said that the work of historians is always a dialogue between the present and the past.

This post is the twenty-eighth is a series on the philosophy of history; the previous article in the series is here, the next is here.

Daniel Halverson is a graduate student studying the History of Science and Technology of nineteenth-century Germany. He is also a regular contributor to the PEL Facebook page.

by