Towards a Typology of the Carolingian Count?

I wasn’t sure about writing this one. I started, and then went, ‘Insofar as this isn’t half-baked, I think I’ve just ended up at Matthew Innes’ arguments about royal power in the localities; and in any case this is an old battle to fight’. But, apparently there’s interest, so I might as well jot this down and see if people think there’s anything to it. (Plus, it’ll be a break from the Aquitanian stuff, although there’ll be more of that next week.)

The question of the Carolingian count is a big historiographical question, more or less revolving around the questions of ‘what did a count do?’ and ‘where did he do it?’ There’s an old model that a count is a government functionary, he exercises state power, and he does it in his comitatus (‘county’) which can be directly equated with the districts known as pagi (such as Flanders, Touraine, or Wormsgau). Modern historians have raised serious questions about this (many of these questions are in German, in a literature dealing with south Germany, which I have tried to read a little of in preparation for this but basically don’t really know, so consider yourself forewarned). Hence the idea that 1 pagus = 1 comitatus now seems very questionable indeed. I myself have observed that some pagi just don’t have counts at all. The Limousin is a case in point here: one will often find references to William the Pious being ‘count of Limoges’, but actually what that means is he shows up on a charter witness list in the 880s and again in a diploma dealing with land in the area a decade or two later, and there’s no evidence at all to assume that he had administrative jurisdiction or even real political clout there unless you assume that every pagus had to have a count.

I want to go further than that today, though. Like I said, maybe this will be obvious to people who know the literature better than me, but I would propose that the title comes (‘count’) did not have the same meaning all the time. So here is a list of all the different types of counts I think I’ve encountered. Some of them in fact overlap; but not all of them do.

First, you have counts who are counts of comitatus. These are much easier to see in Lotharingia and the East Frankish kingdom, where royal diplomas often refer to property as being in ‘the pagus of X, in the county of Y’. And we know that the pagus and the comitatus don’t necessarily overlap, so that you can have multiple pagi under the same comital jurisdiction; but you can also have multiple comitatus in the same pagus. I went looking for examples of the latter, actually, and found one in DD Arnulf no. 60, where a grant to Corvey is ‘in pago Huueitago in comitatibus Ecperti et Reithardi et Herimanni’; sometimes one finds one comitatus being held by multiple people but the plural here suggests that it is multiple counties. (Because this isn’t really what I’m supposed to be doing at all, I stopped there; but there may well be others.)

However, there’s a twist. Because much of this work comes from historians working east of the Meuse, they’re not necessarily as familiar with the West Frankish evidence, because I can tell you that we do in fact have counts who are explicitly counts of pagi. Thus Odo of Paris, the later King Odo, can describe himself in the 880s as ‘count of the pagus of Parisais’. So sometimes counts evidently were made counts of individual pagi.

‘Made’ is an important word there, because some of these counts are as well royal functionaries, or at least royal appointees, coming in from outside to run a locality. Odo is also a good example here: he seems to have spent most of his life in Lotharingia and became count of Paris mostly because Charles the Fat put him there.

Some of them, however, don’t appear to be. This post in fact comes out of the older one about Gerald of Aurillac, who I think is a good example of a figure who is locally entrenched already and whose acquisition of the comital title is a bottom-up process: he is first called count as a mark of respect in the locality, and then the king decides to roll with it as a way of co-opting them. My suspicion is that the evidence for Alemannia, Bavaria, or the middle Rhine would show more of these guys, if I had any familiarity with it at all…

800px-malles2c_chiesa_di_san_benedetto2c_ritratto_del_fondatore_della_chiesa2c_affresco_ix_secolo
Maybe a count, maybe not; but certainly a Carolingian aristocrat (source)

Another suspicion is that these people might be how to explain counts who it’s hard to place geographically. A shout-out here to my friend and colleague Jelle Lisson, who’s done some work on this regard about ancestors of the counts ‘of Vermandois’ (coming soon, he informs me, in the Journal of Family History as ‘Family Continuity and Territorial Power in Early Medieval West Francia: A Reconsideration of the “House” of Vermandois (9th-10th Centuries)’ [edit: and it is now available open access through this finely-crafted hyperlink.). What his work shows is that Heribert I of Vermandois is actually really hard to localise, and that past work calling him the count ‘of Soissons’ or ‘of Meaux’ or the like is reading too much into the contemporary evidence. I’m not sure he’d push it this far, but I think that this is because Heribert I doesn’t actually have a comitatus at all, just a cluster of local interests. Hence, he might be a count who controls an abbey in Soissons or territory in Vermandois, but he’s not count of Soissons or of Vermandois, or of anywhere: his title is unconnected to his territory.  

Opposite to him are people who definitely have comitatus (it’s an annoying word to write in English, because the Latin plural of comitatus is comitatus, not comitati, but it looks wrong…), but whose comital title seems to me unlikely to rest on that fact. The ancestor of the Capetians Robert the Strong looks to me like an example of this sort of figure: the Annals of Saint-Bertin show him being shuffled between the counties of Angers and Auxerre and Nevers, but I think his status is such that his comital title probably doesn’t depend on which specific counties he possesses. Some evidence for this comes from the Annal’s entry for 867, which describe how Charles the Bald took the county of Bourges away from Count Gerald and gave it to Egfrid; that Gerald is still described as ‘count’ suggests that his title was socially embedded rather than legally linked to possession of a specific official competence.

This leads to a final category, which is counts whose title was strictly palatine. There is a title ‘count of the palace’ (comes palatinus), which isn’t quite this, not least because by about 900 it’s become smooshed together into conspalatinus, a separate title you can hold together with comes. I mean counts whose title goes right back to the word’s origin – comes was originally the Latin word for ‘companion’, meaning men who were members of a ruler’s retinue and whose prestige came therefrom. It’s very hard to prove that a given count was a comes in this sense, but if I had to propose one figure, it would be Charles the Simple’s favourite Hagano. Everything we know about his career links him to the palace and only to the palace. I don’t think he ever was a comes in the sense of an administrative functionary in a locality; his title was always a product of his status at court.

Wow, that went on longer than I’d expected, and it’s still rather brief. In any case, it looks rather like the single word comes is hiding a number of different animals, different in degree and not just in scale… Is there anything to this, or have I stumbled the roundabout way onto something obvious?

3 thoughts on “Towards a Typology of the Carolingian Count?

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s