The state they were in

My next catch-up session on IHR seminars from earlier this year discusses a couple of papers about early medieval state formation. First up was Marie Legendre in February on “Not Byzantine nor Islamic? The Duke of the Thebaid and the formation of the Umayyad State”. This has recently won the Pollard Prize for best presentation at an IHR seminar by a postgraduate/recent PhD, and so will be published in Historical Research. I’m therefore not going to say much about the detail, especially since there was a lot of it, and it’s a long way from my field of expertise, as this map will show:

Map of the administrative divisions of Egypt in 400 AD. The Thebaid is the one at the bottom.

Marie was focusing on the southern part of Egypt in the second half of the seventh century, after its conquest by ‘Amr ibn al-‘As in 639-642, and trying to answer the question of whether there was a state there before the reign of Abd-al-Malik in the late seventh century. She was arguing that the framework of continuity versus change wasn’t really useful, because it tended to separate Byzantine from Islamic aspects and some things changed, while others didn’t. Fred Donner’s influential concept of the state, as expressed in “The Formation of the Islamic State,” Journal of the American Oriental Society 106 (1986), 283-296, also doesn’t really fit this world.

Instead, a complex system of overlapping authorities is visible. This included successors to the Byzantine administration of the pagarchs and their superior, the Duke of the Thebaid (the duke governed the whole province, with both civil and military authority). But these acted alongside the Amir of Hermopolis, and moagaritai are also present, who look like soldiers.

There were around eight dukes of the Thebaid, between the Arab conquest and 710, after which they are no longer visible. Some of the later ones have Arab names and some have the title of both duke and amir, but since some locals were also adopting Arab names, we can’t always be sure whether the dukes were locals or not. The administrative situation was very complicated: although the province of the Thebaid was merged with Arcadia under Duke Jordanes in the 660s or 670s, there seem still to have been two provincial systems operating in parallel, and two parallel hierarchies, one under the duke and one under a governor.

Changes are gradually visible however, such as fiscal documents being written in Coptic; from about 706, numerous epikeimenoi (overseers) can be seen in administrative roles, all with Arab names, dealing with matters previously deal with by dukes. By 714 the pagarchies were also being replaced, by districts called kura, whose administrators had Arabic names and who issued documents in Coptic, Arabic or bilingual Greek/Arabic. Marie thought the dukes were becoming unnecessary as links between the governor and the pagarchs, and they cease to be visible.

Marie concluded with reference to Pierre Bourdieu’s theory that the successful state renders its formative phase forgettable, since the elite comes from a social order they have to destroy. The duke of Thebaid had an important role in the change from Byzantine to Islamic order, but despite the continuous presence of such a figure, he can’t be seen as a Byzantine official, but instead one of the main actors in forming an Islamic hierarchy.

It’s these insights that link to the paper a few weeks later by George Molyneaux on “The Formation of the English Kingdom in the Tenth Century”. I see that Jon Jarrett has already given a fairly detailed write-up of what sounds like a similar paper George gave at Oxford a year previously, so those who want details of George’s arguments should probably start there.

George was essentially asking the question of how you get from point A in England (900, when there’s no political unit corresponding to the “kingdom of the English”) to point B (the mid-eleventh century when the entry for 1017 in version C and E of the Anglo-Saxon Chronicle say that Cnut (Canute) “obtained all the kingdom of the English [eallon Angelcynnes ryce]”. The period between the reign of Alfred (871-899) and 1066 tends to be taken as one block, but George is trying to work out when particular “state” developments actually take place, and seeing more of them occurring in the third-quarter of the tenth century rather than earlier. Jon’s talked about the idea of intensification of the power of kings (and also whether it’s a Good Thing), so I want to be a bit more basic and go back to the evidence that George was using.

Speaking as a non-specialist, he did seem to be making two very important points. One was that the control of coinage is much tighter from the 970s onwards: there’s a uniform design and relatively stable weights. The second is that there is legislation from the first half of the tenth century to compare with that from later in the tenth century and it very rarely talks about hundreds and wapentakes in the places it should do (e.g. when witnessing).

In other words, with both the coins and the hundreds as administrative units, we don’t just have a lack of evidence from the previous period; we have some evidence of a different social situation. Therefore, arguing that there were hundreds (or areas which corresponded to later hundreds) before the mid-tenth century doesn’t in itself wreck George’s argument. You’d have at least to add in a kind of “documentary mutation” section to the argument in which later tenth-century kings suddenly feel a new need to discuss administrative structures which were too obvious to be worth mentioning before.

I was less convinced by George’s argument that the (temporary) cessation of hostilities with the Vikings and attempts to conquer Britain were the reason that these changes happened in the later tenth century rather than the first half of it. As Charlemagne’s empire shows, you can have continual warfare and considerable legal/economic/governmental developments happening at the same time. But although there were a lot of people disagreeing with George in the discussions afterwards, I don’t think anyone came up with convincing proofs about why most of the English administrative structures had to exist much earlier. (The nearest was Stephen Baxter saying Patrick Wormald had done unpublished work showing sureties being taken from local communities in the late ninth century. I’m afraid I can’t give more details than that, being pretty ignorant about Anglo-Saxon law generally).

George is now writing the book of this argument, so anyone who wants the gory details should probably wait and find them there. But it was only when I came to write up his paper that it abruptly connected in my mind with Marie’s. What have we got here again, but the forgettable (if not actively forgotten) formative stage of the successful state? In an early medieval culture that venerates custom and fears novelty, of course the state is going to be made to appear as traditional as possible by reusing existing structures. We also know that the later tenth century was a hotbed of ideological propaganda, creating the idea of an English state. So we’re probably not going to be able to find its origins entirely accurately, but we need to be quite wary of later claims that it had existed for centuries.

Secondly, Marie’s paper suggests that a third alternative to (abrupt) change or continuity is continual small change – the Thebaid in the 710s was very different from the 640s, but it’s hard to see any obvious single breakpoints. The idea of the “precocious” English state (which Susan Reynolds was objecting to) tends to imply an Alfredian Big Bang and then just a smooth motoring on for another 150 years. But if we look at the Carolingians, we see another state that’s developing new institutions over at least fifty years, if not a century. Charlemagne’s first really big reform capitulary (the Admonitio generalis) comes twenty years into his reign, and Louis still changing things seventy years after his grandfather first became king. If you have some kind of partial ratchet effect, in which developments that make kings more powerful don’t voluntarily get reversed by them, then you don’t need very many small steps to get a considerably more powerful state. The seventh-century Islamic world wouldn’t be my first point of comparison for what happens in late Anglo-Saxon England, but it can still get us thinking more creatively about how and when administrative structures develop.