I have been reading and writing several different factual and fictional narratives over the past few weeks, particularly as I am starting to get to grips with my current project on the divorce of Lothar II and Theutberga. And I'm finding myself more and more uncertain about whether some of the factual narratives I'm trying to create have much more claim to reality than many fictional ones.

Normally, I am used to writing analytical history, which essentially works by aggregating and averaging data (even if for early medieval history, not that much data). If I can show a pattern in a dozen uses of the term 'viriliter' (manfully) for example, it doesn't matter if I may be wrong about one individual source. And humans are also relatively predictable on average. If I say that early medieval peasants were often exploited by their lords, that's a generalisation from a few data points combined with a knowledge of human psychology to make a sensible argument.

But individuals aren't average (at least on average). Statistically, men make up around 80% of people with maths and computer science degrees. But if you deduce that because I'm a woman I didn't study mathematics, you're wrong. This confusion between the average and the specific is something that was first really brought home to me by a TV programme by Simon Schama, Murder at Harvard. Schama is trying to solve the historical problem of the murder of a prominent doctor in Boston in 1849. At one point in the transcript Schama starts discussing the behaviour of a janitor, Ephraim Littlefield, who he believes was responsible for the murder:

Unlike his rich employers, Littlefield left behind scant evidence of how he thought or spoke. But the voice of men like Littlefield does survive in letters and diaries of the time, and I read enough of them to make me think I could hear him speak.
...
I heard a voice full of resentment toward the Harvard professors -- those lofty men of science...As he cleared their rubbish. . . and got them their cadavers.
...
I knew I was crossing a line historians don't usually cross, the line that separates history from fiction. But instinctively I felt I was coming to understand Ephraim Littlefield's motivation. He was not a murderer. It was bitterness towards Webster and towards his own lot in life that drove him to do that dirty job.

It's reasonable to think that on average someone like Littlefield would resent his superiors. It's an entirely different matter to say that Littlefield did resent a particular doctor enough to kill him. Perhaps it's just because it's accusing someone of murder who can't defend their own reputation that I feel a bit queasy about this.

Until recently, I'd have said that the problem here is just that Schama is trying to take his imagination too far, that it's a technical error of his as a historian that is the problem. But for the last few weeks I've been turning over a clause from a Carolingian church council (Council of Metz 893 c. 10):

Quaedam foemina, nomine Ava, cum sui fratris consilio & auxilio, qui vocatur Folcrius, & cum aliis consanguineis suis, suum maritum dimisit, & ad eum redire noluit. Unde illorum sacerdos servus Dei, vocabulo Folcardus, & ad suam dominam & ad ejus fratrem veniens, ut eos a tanto scelere traheret, confestim ab eodem & suis complicibus castratus est. Pro his omnibus ad synodum vocati venire noluerunt: & idcirco ad satisfactionem excommunicati sunt.

A certain woman called Ava, with the counsel and aid of her brother, who is called Folcrius and of other relatives of hers, sent her husband away and did not want to return to him. Therefore, their priest, the servant of God called Folcard, coming to his lady and her brother so that he might drag them away from such a great crime, was immediately castrated by him and his accomplices. They were called to the synod for all these things and did not want to come and therefore were excommunicated until they gave satisfaction.

This is a story, but possibly not a plot, in E.M. Forster's sense (“'The king died and then the queen died'" is a story. 'The king died, and then the queen died of grief' is a plot."). But it also looks bizarrely like an RPF (real person fiction) prompt, asking someone to write a larger story around it. (This is by no means the most bizarre RPF prompt going – those of a nervous disposition or who admire Bernard of Clairvaux should not click here). There are different fictional stories that could be written around this clause: some in which Ava's behaviour is somehow justified, some where Folcard is recklessly irritating (the Thomas Becket or Peter Abelard of ninth century Germany), some in which Folcrius is a sadistic joker who argues that it doesn't matter to a priest if he's castrated.

But what worries me more is how to decide between some of the possible factual stories this clause could support, because though it can't have only one meaning, there are some meanings it could have that are incompatible with others. Does this story tell us something about the Carolingian period as a whole, or only one late corner of Carolingian Europe? Are men are getting attacked and castrated quite frequently, or is this tale is as freakish to people then as it to us now? Does this tell us something about lay attitudes to clerics, or noble attitudes to subordinate clerics ('their priest'), or does it tell us only about Folcrius?

I've used this story before in my work to talk about how noblewomen could exploit their family connections to escape from marriages they disliked, and I still think that's a reasonable aggregate of this tale and other texts. But I'm not sure now that I can get any sense of Ava or Folcrius or Folcard themselves, or at least any sense that is based on my own narrative imagination. These are minor characters, it may not matter that I can get no hold on them. But I worry that some important figures in De Divortio may share this same indeterminacy. I don't worry that Hincmar or Lothar II are purely figures of my invention, I feel there is something there that I know, not just imagine. But I'm still not sure whether I can discover Theutberga, the woman at the heart of this case, or whether all I am doing is creating her as the wronged heroine, the woman who must suffer the accumulated assaults of Carolingian patriarchy. How do I recreate her without inventing her as I want her, as she ought to be to make the story dramatic? I'm not yet sure how or if I can do that.