It’s fair to say that the historian Orosius is no longer a household name. Yet his Histories in six books were, for centuries, extremely popular. There are circa 275 extant medieval manuscripts (this is a lot) of the original Latin, and medieval translations were also made into Old English, French, Italian, Spanish and Arabic.
Writing in the early 5th century, Orosius covered the history of the world, from Creation to his own time. He was a friend of St Augustine’s, and the stated purpose of his Histories was to bolster Augustine’s thesis in the City of God, that there was no correlation (as we would put it) between the Romans’ piety towards their gods and the rise or fall of their empire.
This framework naturally made Orosius’ work congenial in the Christian centuries. But its success is due above all to a simpler factor: it is a vivid and simple account, covering a broad period of history at a digestible length. In short, it is the sort of single-volume “all the history you need to know” book that, if well executed, will be popular in any era where a veneer of history is a requirement for the educated man or woman.
By the same token, such books’ popularity can never last forever: tastes change as to the contents and shade of the veneer. That Orosius was read for so long illustrates continuities of thought between Late Antiquity and the Middle Ages, just as the waning of his popularity points to a later rupture.
One element of this rupture takes us to the reason that Orosius is not only no longer a popular historian, but is also very little read even by specialists. Many classicists and ancient historians are at best dimly aware of Orosius, and very few would consider it their duty to read him through (contrast Livy: few classicists have read all of Livy, but we generally feel we probably should).
It is respectable to neglect Orosius—unless one is a historian of Orosius’ own period—because of a criterion that has become very meaningful to professional historians: for most of the period he covers, Orosius’ sources are all extant.
In other words, Orosius sewed together his ancient history from authors such as Herodotus, Livy, Suetonius, all of whose texts survive. It follows that, for those early centuries, historians cannot learn from Orosius “facts” that they could not get directly from where Orosius himself got them. Historians then concluded that they would do better to eliminate the middle-man, and so they set Orosius aside.
This choice may seem self-evident, but in fact its seeming self-evidence is itself a historical product: when we speak of the birth of “scientific history” in the early modern or modern period, one of the developments we have in mind is the rule that a historian must seek out and use the earliest extant sources of a story, rather than later sources that derive from these.
“Any idiot”, as Moses Finley wrote when discussing the emergence of this norm, can understand the difference between an original and derived source, and why that difference might matter. Both ancient and medieval historians discuss the problem. Yet it is seemingly not till the 16th century that we can see any attempt to systematically sift sources and work out their chain of derivation, and not until the 19th century that such work becomes, at least in theory, a sine qua non of serious historiography.
There is much to say about these developments, and I hope to return to them. A preliminary observation for today is that we cannot separate the development of the sort of source criticism that sidelined Orosius from the development of printing. Printing introduced not just a new way of making books, but, gradually, a new way of thinking about them.
Compared to hand-copying, even the earliest forms of printing enabled the production of vastly more copies of a text within a given space of time. Moreover, for any specific print run, these copies were all identical, which handwritten copies never are. The result was that the printed edition of a text gave it a permanence and solidity which texts previously lacked. It then became possible to imagine imposing a new level of order onto the world of old books, creating definitive lists of what was extant, and making whatever existed available to anyone who wanted it.
At that stage, demanding that historians always work out and then use the earliest extant source for a story likewise became imaginable, and in due course reasonable. I can find, at my university library, or indeed without leaving my computer, every extant source that Orosius used.
But for Orosius himself, back in that world where the only books were handwritten ones, it would have been quite impossible to locate the earliest extant source for each story he wished to tell. And it would have been equally impossible for Orosius’ medieval readers or recyclers to locate all the sources Orosius used, even though there were copies of them somewhere in European libraries. Accordingly, if we think of these medieval historians as lacking “critical sense” or a “scientific” mindset, we are in a certain way putting the cart before the horse.
