Some brief comments on historiography

April 2017

There is a duality in the meaning of historiography, perhaps reflecting a duality in the meaning of history itself. History can refer to past events as they actually occurred (we're going to call it actual history), or it can refer to a narrative that claims to report past events. We thus use the word to refer both to whatever actually happened and to what some people, preferably historians, tell us about what happened. Those two histories, obviously, can differ. This is too often forgotten, not always inadvertently, giving rise to all manner of equivocations.

Historiography likewise has two primary senses. It may refer to what has been written about past events, or it may refer to the study of how such writing is done, has been done, or ought to be done. These excerpts are from the first four pages returned by Google on one occasion in response to the query historiography definition:

Unless otherwise required by context, I shall use the word to mean the study of how history should be written. Assuming that we intend to write a factual history of whatever events we're concerned with, this entails an examination of how we know what actually happened—the who, what, when, where, why, and how of actual history. This in turn raises epistemological issues, since there is continuing controversy over whether we even can know such things or, if it is stipulated that we can know, over what are the proper methods for obtaining such knowledge. Historiography, in this sense, is just the effort to find out what actually happened at some time in the past in order to produce a truthful narrative, or a narrative as close to truthful as we can make it. I have seen it claimed that the writing of history is more art than science, and that efforts to do it scientifically are misguided. The methods of science, we are told, cannot be applied to any study of the past. I hope to show that such claims reflect both a mistaken kind of skepticism and a misunderstanding of the nature of scientific inquiry.

We begin with some preliminary observations. Human cultures that existed before roughly 6,000 years ago are called prehistoric because history, as popularly understood, began only with the invention of writing sometime around 4000 BCE. This usage has come into disfavor in many intellectual circles as a disparagement of preliterate cultures, but no such disparagement need be implied. A distinction between the historic and the prehistoric, in this sense, is simply an acknowledgement that what we can learn from documentary evidence is more in quantity, different in kind, and in some important respects more interesting than what we can learn without any documentation. Much can be learned about humanity’s past from archaeological evidence, but material artifacts can tell us little if anything about how people, especially individual people, used to think. Archeology might tell us that a certain group of people engaged in some battle at some time and place, but it is unlikely to give us any idea of the reason for that battle.

These essays are about, among other things, the history of certain ideas in Western civilization. To reliably associate particular individuals with particular ideas, we need documents with those individuals’ names on them. The only documents to have survived from the beginning of Western history are of stone or clay, and they were not written by philosophers. For scholars investigating the origins of Western philosophy, the paper trail all but ends barely a millennium ago. Except on a few statues and other monuments, there exists today no document relevant to Western history actually written by any famous person who lived more than a few centuries ago. What we have are purported copies of some documents written by famous and not-so-famous people who lived decades or centuries before the extant texts were produced. For most of our knowledge we are at the mercy of countless generations of anonymous scribes who made copies of copies of copies of documents whose originals were said to have been produced by certain people whose identities we get only from the scribes themselves.

Writers who ignore this may assert bits of historical trivia with greater certainty than the evidence warrants. For just one example, one often finds Thales credited, without qualification, with predicting a solar eclipse in the 6th century BCE. How do we know he did that? Well, Aristotle said he made the prediction, but Aristotle lived some two centuries after Thales, so how did he know? Actually, he didn’t. He was just repeating a story he had heard. Of course the story could have true, but “could have been” is not the same as “certainly was.” And for that matter, how do we know that Aristotle told that story? The oldest copies of anything he allegedly wrote were made in medieval times, upwards of 14 centuries after his lifetime. It would take us too far afield to discuss how historians settle questions about the authenticity of their source documents, but a couple of points will become relevant later. One is that none of their answers are regarded as infallible. What we think we know about who wrote what, and about how closely the existing manuscripts match their originals, is almost universally considered a matter of probabilities, not certainties. In many cases the probability of both authenticity and accuracy is deemed high enough to leave little if any room for reasonable doubt. Even so, it remains a matter more of collective judgment than of historical fact. It is rhetorically convenient to treat a settled judgment as if it were a fact, but there are occasions when convenience must yield to precision.

The uncertainties multiply when we talk about people for whom firsthand evidence is not even alleged. Nothing is reliably known about any of the first Greek philosophers—specifically, all the ones who predated Socrates and are thus referred to as the Presocratics. Nothing they might have written has survived, and the earliest references to them are by people who lived, in most instances, long after their purported lifetimes. Those references sometimes include a few quotations said to be from the now-lost writings. Those quotations are typically referred to as “fragments,” as if they can be considered pieces of the original documents, but that is not what they are.

Suppose I have a friend who wrote me a long letter about something you happen to be interested in. I could write you a letter in which I quote a passage from my friend's letter to me. Alternatively, I could cut out that passage and mail the clipping to you along with a letter in which I offer my own thoughts on the matter. In the latter case, you have a fragment of my friend's letter. In the former case, all you have is a quotation from his letter. In some sense an accurate quotation is just as good as a fragment. And maybe you trust me to have quoted accurately, but maybe you don't. Whichever your judgment, you should have a good reason for it, and if you trust me to have quoted accurately, you should at least have good reason to think I actually got the letter I was quoting from. When ancient writers quoted earlier writers, we don't always know that they had a copy of whatever they were quoting from.

It is conceivable, i.e. actually possible, that some or all of the people identified as Presocratic philosophers are fictitious, although few if any professional historians take that hypothesis seriously. Possibility is not equivalent to probability. Scholars in general are becoming increasingly skeptical about the reliability of ancient documents, but what the skeptics usually question are certain kinds of uncorroborated details. A consensus among multiple independent sources that, for example, a certain person lived and said certain things is usually taken as likely (but never certain) to be true. At the same time, a claim by only one of those sources that that person did something improbable will be regarded as not so credible.

About even the later philosophers, for whom at least some documentation seems reliable, some skepticism is justified, however rarely it be expressed in books or articles for general readerships. Socrates’s historicity is a good example. We have no solid evidence for his existence that was produced during his lifetime. He does appear as a character in a play written while he was alive, but the play is obviously a work of fiction, not history. Our best sources are documents produced many centuries after his purported lifetime in which his existence is alleged. Some of them are supposed to be copies of documents whose originals were produced by men who knew Socrates. And, they probably are just that. However, while alternative hypotheses are currently hard to justify, they cannot be declared impossible. Occam’s razor speaks loudly for a real Socrates, but a fictional Socrates is not entirely indefensible.

In any case, the essays in this series are more about the history of certain ideas than about anybody’s biography. Some of the ideas I will be discussing were recorded in a set of books thought to have been written by a man named Plato living in Athens during the fourth century BCE. Other ideas are set forth in other books attributed to another Athenian (originally Macedonian) named Aristotle. That writer, whoever he was, claimed among other things to have been a student of Plato. For various reasons, historians are convinced that those two men were real people and really wrote most (not all) of the documents attributed to them. For our purposes, it will not matter whether the historians’ consensus is correct. The ideas in those books were widespread in the Western civilized world within a few years after their purported authors' lifetimes, and the documentary evidence for that is overwhelming. To keep the conversation simple, we might as well assume that the purported authors were the actual authors except when we have good reason to think otherwise.

We should remember, too, that the validity of no philosophical or scientific idea is contingent on the truth of anybody’s biography. Galileo almost certainly never dropped anything from the top of the Leaning Tower of Pisa, and he very certainly did not append “Eppur si muove” to his recantation. But neither gravity nor heliocentrism depends on whether he did either of those things. For a counterfactual example, we can imagine (with great difficulty) somebody proving that Charles Darwin never existed, that every document with his name on it was a forgery produced by some anti-creationist conspirators. The theory of evolution by natural selection would be unaffected by such a discovery. Scientific theories do not depend for their acceptance on the truth of stories about how or by whom they were discovered. These issues surrounding the historicity of some famous thinkers will become relevant in subsequent essays. For the time being, our narrative is based on the apparent consensus of historians, and to keep the narrative readable it assumes, except where otherwise noted, that the consensus is substantially correct. When I get to a point where I think the consensus should be challenged, I will explain why I think so.

Site home.

This page last updated on April 24, 2017.