A bunch of early medievalists have been discussing textbooks over at A Corner of Tenth Century Europe and several laments have come up over the lack of narrative histories for the period. This got me thinking more generally about the purposes of narrative history and why it’s often so hard to find (or write) good ones.

In a sense the purpose of a narrative history is there in the term: it’s to tell a story. And speaking as someone who was a (failed) novelist before becoming a historian, it’s not easy to write a story. In particular, what stories really need are a coherent theme (the narrative arc) and also memorable characters.

Narrative history, therefore, means selecting and describing historical events so that they no longer look just like one damn thing after another. There are some aspects of history that are relatively easy to treat in this way (though it still takes skill on the part of the author). There are lots of good narrative histories of wars (as three random examples I’d take Stephen Runciman on the Crusades, Veronica Wedgewood on the Thirty Years War and Shelby Foote on the American Civil War). Historical biographies can be very good reads and dynastic histories also often work well, because there’s a ready-made narrative of rise and fall.

Narrative histories of nations have become harder to do now that historians are suspicious about teleological narratives: you can’t easily do a Whig history of Britain now. And, in fact, very long-term narratives (more than about a century) are hard anyway, because there is so much turnover of characters. I can never really get into the dynastic novels where you have to get to grips with a completely new set of characters every 100 pages or so: it becomes increasingly hard to remember which one is which and to care about any of them. The rare narrative histories that can sustain interest over such a length of time are usually those where a country/city can be seen as a ‘character’ in its own right (I’m thinking here particularly of John Julius Norwich’s histories of Venice and Byzantium, but the same is true of many histories of ‘Rome’).

One of the two big problems with trying to write a narrative history of early medieval Europe is therefore: what is the story? To the extent that there is an overarching one, it’s initially the fate of the post-Roman world. You can just about maintain a vaguely coherent narrative that contrasts the fate of Britons, Vandals, Ostrogoths, Visigoths, Merovingian Franks and Byzantines. (I know this ignores the Burgundians, but frankly, who cares about them?) The problem is, you can maybe keep this narrative up till about the seventh century. After that, what do you do? There is no coherent narrative that can tie the different kingdoms together (unlike with the early modern period and even, to some extent, the later medieval period, where there are some pan-European events to tie things together (the Crusades, Battle of Bouvines, Black Death). You can’t make a story out of ‘well this was going on here, and that was going on there’; the most you can do is make a series of short stories. And it’s hard to make such short stories of interest: the Lombards rise and fall, but so what?

So if you are trying to write a narrative history of the early Middle Ages, you either have to produce a series of chapters that don’t connect together or you have to cheat by restarting the story (this is normally done by ending the contrasting fates of sub-Roman kingdoms with Merovingian Francia and Clovis and then jumping to the Carolingians, at which point you have a coherent vaguely pan-European story again). Similarly I don’t think anyone has yet tried to do a narrative history version of the theory that everything before the year 1000 is really late antiquity (as Richard Sullivan and others have claimed) because you’d then have to have the fall and rise and fall of the Roman Empire and it really doesn’t work coherently as a narrative.

The second big problem for writing early medieval history is memorability: how do you get people to remember your story afterwards? For a novelist, this isn’t necessarily a problem: there are some novels where the pace of events is enough to carry you through the book and if you forget all about it afterwards you can just read the novelist’s near identical next-book. But if you can’t remember anything about the period after reading a history book, it hasn’t been of much historical use, even if it was entertaining. I have this problem myself with a lot of books on unfamiliar historical periods; even with a good narrative history I tend to forget a lot of what I’ve read soon afterwards. So every year or so I feel I ought to know more about Chinese history and read something on it and it all goes in one ear and out the other. The combination of unfamiliar names, geography and events is particularly difficult for the memory.

Early medieval Europe is appalling for this, with large numbers of unfamiliar and similar names (how many King Theudeberts are there and are any significant?) and severe problems for those of us who can still never remember with certainty where the Meuse is (and that it’s also the Maas). The most effective way of getting round a lack of memorability is obvious (for novelists as well as historians): a well-used anecdote (or a nickname, which is a kind of condensed anecdote). I can still remember that Antigonus One-Eyed was a successor to Alexander the Great long after forgetting almost all the rest of Peter Green’s wonderful book ”Alexander to Actium”. And almost the only bit of Chris Wickham’s discussion of post-Carolingian Italian politics in Early Medieval Italy that I can remember is his statement that Berengar of Friuli ‘in forty years of campaigning is not recorded as ever having won a battle’.

With that single statement, Berengar suddenly becomes memorable, steps out of the pack of other would be rulers of Italy. But for the rest, Chris’ overview of Italian events in the book, while perfectly readable, is deeply unmemorable. Chris’ writing style is perfectly clear, but that’s not enough for good historical narrative. For example:

In 900, the magnates of the north-west, now led by the Anscarids, marquises of Ivrea, a Burgundian family installed by Guy [of Spoleto], revolted and elected Louis of Provence as king. Louis went to Rome and was crowned emperor, but his support fell away and he left in 902. When he returned in 905, Berengar captured and blinded him. Berengar then ruled unopposed until 922.

How much of that could you remember even by the time you turned the next page? This is basically a slightly expanded version of a chronology; useful for reference, but not a lot else. This isn’t to get at Chris: you get a similar effect reading Roger Collins or many other narrative histories. If you want to write memorable early medieval history you have to have room for more expansive writing. You are also likely to have to use the kind of dubiously sourced anecdote that Victorian writers of narrative loved and modern historians spit at. Alfred and the cakes and purple prose may not appeal to modern academic historians, but if you want a narrative history that’s more than a reference work, I don’t see any alternative to that.