This week I’ve been discussing the early medieval ‘state’ with my students, and chewing over the various complexities of the topic. Was early medieval rulership too personalised to count as a proper state? Was it too undifferentiated from religious authority? Was it too broken up by alternative and rival jurisdictions?
Here’s a charter (pdf) issued by King Lothar II in 856 which illustrates the issues at stake. In it, King Lothar II grants a kind of personal fiscal and legal immunity to a man named Winebert and his family. Winebert had given over all his property to the monastery of St Arnulf of Metz, but he remained a secure tenant on his lands, in exchange for paying the monks a modest three pennies worth of wax each year. In return, he now benefited from the ‘immunity’ of the monastery of St Arnulf.
So, in this charter Lothar confirmed that neither Winebert nor his family could now be summoned to military service, or pay the stofa (a mysterious, and presumably irregular, tax). No public agent or royal missus could make public demands of Winebert, his family, or their heirs, at all. Winebert had privatised himself and his family, so to speak.
The charter gives us a precious indication of the sorts of claims that kings such as Lothar II could make of their free subjects. But for previous generations of historians, this kind of action weakened the ‘state’; through this document, King Lothar was surrendering his authority and diminishing his revenues.
But Barbara Rosenwein has taught us to be more subtle about immunity charters such as this. These grants had their costs, but they also brought kings benefits. After all, the monastery of St Arnulf was a family mausoleum for the young king, where his grandfather was buried. Helping this monastery boosted royal prestige and credibility at a time when Lothar needed both. Foregoing whatever Winebert might have given the king’s agents was a price worth paying for that.
As the UK government announces a new generation of spatial zones subject to special taxation and regulation arrangements, known as freeports, this point might become easier to grasp. Freeports are supposed to generate economic capital, not the spiritual capital created by Carolingian immunities, but both are and were intended to feed directly into politics. Now that we see how the modern UK government is deliberately surrendering powers – in part in order to raise tangible revenue, but perhaps in part to demonstrate sovereignty through the very act of renunciation – might the power games of Carolingian kings around immunities no longer seem so strange?
Chris Wickham has argued that polities whose political economy is based on grants of land (as opposed, say, to tax and salaries) are intrinsically unstable. Here’s a text in support of his argument, written by Archbishop Hincmar of Reims in 876 – the De villa noviliaco, a text surviving in a single ninth-century manuscript (Paris lat. 10758).
In it, Hincmar recounts the complex history of the villa (estate) of Neuilly. King Carloman granted this estate to the church of Reims in the late eighth century; the grant was confirmed by Charlemagne who nevertheless gave it in benefice to a Saxon named Anscher. Later, Charlemagne’s son Louis gave it to a count named Donatus. Legally, these grants in benefice did not overturn Reims’s ultimate ownership. However, Donatus craftily sliced off some holdings from the benefice which his family would later claim were entirely his.
When Emperor Louis’s sons rebelled against him, Donatus had to make a political choice; unfortunately for him, he made the wrong decision, joining Lothar just before Lothar surrendered. Neuilly was duly taken from Donatus, and given to a certain Hatto. But when Hatto died, Donatus and his family got it back again. However, when Louis the German invaded the western Frankish kingdom in 858, Donatus’s widow Landrada read the politics wrong once again, deserting Charles to join Louis. When Charles re-established control, he therefore took the estate from Landrada and gave it to the monastery of Orbais. Only then did Hincmar finally manage to get Neuilly back for Reims, shortly followed by the associated holdings stolen by Donatus. Hincmar did not keep the estate in house, however, preferring to grant it out in benefice to clients, Rothaus and Bernaus.
Donatus’s family was down but not out, however, for when Louis the German invaded again in 874 (while King Charles was in Italy), his and Landrada’s sons managed to get the estate back from Queen Richildis, presumably in return for their political and maybe military support. Not until Charles’s return from Italy was this grant overturned thanks to Hincmar’s lobbying.
The estate of Neuilly thus changed hands at least eleven times in around a century. Hincmar thought Reims had the better claim – and used the written word to prove it, referring to several charters none of which now survive. But the family of Donatus thought otherwise, and had their own established hereditary claim which had twice been honoured. We cannot know what they would have made of Hincmar’s arguments, but mostly likely they would have argued that although Reims owned Neuilly, they had a family claim to it as a benefice. Perhaps the families of Hatto, Anscher and Bernaus (each of who had held at it some point) would have seen things differently again.
Who had the best claim to Neuilly was therefore a political question, which is why Hincmar wrote and preserved his (doubtless partisan) account. But what Hincmar’s history does show quite unequivocally is how tensions over landholding made Frankish politics in the ninth century very unstable. For there was always someone waiting for the right opportunity to press long-harboured claims over some estate – and no shortage of rival kings willing to provide that opportunity.
All historians, I think, are attracted to the gaps in the archive – the silences, the absences, the things that aren’t there. For historians of early medieval Europe, it can sometimes feel like there’s more gap than record, though really this isn’t such a poorly documented time and place, especially Carolingian Francia. But it’s the inconspicuous absence, not the glaring one, that’s often the most telling: and here’s a case in point.
On 17th January 866, King Lothar II granted his wife Queen Theutberga twenty estates in Francia, in a charter issued at the royal palace of Aachen. For kings to transfer lands to their queens was not itself unusual. But such transfers normally took place at the beginning of a marriage, as a dowry, not as in this case eleven years later, which makes this charter rather strange.
That’s not however the only remarkable thing about this charter. What’s also odd is how it describes Theutberga – or rather, how it doesn’t. She is described as dilectissima nostra, ‘our most beloved’. But ‘our most beloved’ … what? The adjective dilectissimus is quite common in royal charters from the ninth century, but it’s usually applied to a noun: ‘our most beloved’ sister, son, wife, daughter, mother, etc. In this charter, there is no noun, Theutberga is just ‘dilectissima’. Why?
Perhaps the context of the charter might help explain it. As all readers of De Divortio will know, in 866 Lothar was still grimly struggling to separate from Theutberga, and this charter has been interpreted as a pay-off or compensation in exchange for separation. In that context, he could hardly call her his ‘wife’ (coniunx, uxor). That would be an awkward acknowledgement of a status that he was insistently denying. To call her his ex-wife would however have been politically risky – he wasn’t quite there yet. Better perhaps just not to say anything at all.
In a later charter, it seems that Lothar applied similar discretion to Waldrada, his mistress, who is described as ‘our beloved’ with no further qualification – though in this case we cannot be entirely sure, because the key passage was later tampered with, when someone changed Waldrada’s name to ‘Rotrude’.
But Lothar’s grant to Theutberga has not been tampered with, and survives in its pristine original. And that allows us to see something rather peculiar – something that it’s tempting to ascribe to the hesitation of the scribe, identified by modern historian as a man named Rodmund (who had also worked for Lothar I, and who seems to have in practice run Lothar II’s chancery), in the face of this unusual phrasing.
Though it’s not marked up in the standard edition, Dupraz noted that there was a significant gap after Theutberga’s name. And on inspecting a facsimile, he’s right. There’s in fact a clear gap in both occurrences of the phrase “Theotbergae dilectissimae […] nostrae” in the charter. The first, at line 3, is this blog’s cover image. The second, at line 7, is visible here:
The intact ascender of the L reaching up from the line underneath suggests this gap isn’t an erasure (though one might need to check the original to be sure). But if the gap isn’t an erasure, what is it?
Dupraz suggested it was to enable a suitable noun to be added later. Perhaps, but I’m not quite sure the space is quite long enough for that. Rather, the little gap seems to express the uncertainties of Lothar and Theutberga’s circumstances, years into their relationship’s tragic breakdown. A moment when the scribe stopped, paused and moved on?
Image: Lothar II, D. 27. Original charter in Parma. It seems possible that Lothar II took this charter with him to Italy, along with a reissued version of 868 (a topic for another blog), and it was abandoned in Piacenza when he died there in 869. More research required…
*Updated 10 August 2018 in light of comments from Clemens Radl and Levi Roach (thanks to both).*
*Updated 4 Jan 2021 to add draft translation*
 Louis Dupraz, ‘Deux préceptes de Lothaire II (867 et 868) : ou les vestiges diplomatiques d’un divorce manqué’, Zeitschrift für schweizerische Kirchengeschichte 59 (1965), pp. 193-256.
Further reading. The charters are briefly discussed in Heidecker, Divorce of Lothar, p. 171, with n. 93. You can see a full facsimile in Chartae Latinae Antiquiores, vol. 93, no. 7, pp. 40-43, with useful technical commentary on the scribe. For helpful context, see Roberta Cimino, ‘Royal women and gendered communication – Female voices in Carolingian diplomas’, L’homme 26 (2015).
One of the problems of studying the Frankish kingdom of Lotharingia – the ‘lost’ kingdom between France and Germany – is that the main narrative sources for the time were written outside its borders: they are external perspectives, looking in. It is this that makes the charters issued by Emperor Lothar I and by his son King Lothar II (855-869) so important, as ‘internal’ evidence. A full book-length study of the former by Elina Screen is in hand; this short blogpost focuses on the latter, all available in a high-quality and open-access edition.
Only thirty-six charters from Lothar II’s nearly fifteen-year reign survive (including a handful of originals with images available online).. This isn’t an enormous number, either in absolute terms or compared to the 139 from his father’s admittedly larger kingdom. But the charters nevertheless provide a great deal of useful information, in this as in other early medieval contexts.
For instance, we know that issuing a charter was a ceremonial act, by which a king demonstrated his authority. Where it took place mattered, then, as the scene for the expression of royal power. Lothar II issued charters at sixteen different locations, including palaces, monasteries and cities. Plotting these locations on a map creates a view of Lothar’s kingdom defined not by its borders but by its centres of power – a king’s eye view of his kingdom, as it were (see the map above, showing how his kingdom spanned five modern countries).
His great-grandfather’s palace of Aachen towers above all the others, as the site for almost a third of Lothar II’s charters. But the charters also show change in his rule: for instance, the grants he issued in and around Lyon mark Lothar’s takeover of most of the kingdom of his younger brother, Charles of Provence, in 863.
Many charters also mention important political figures at the time, allowing us to reconstruct something of their careers. It’s striking that the first two of Lothar’s charters, both issued in 855, accord a prominent position to Hubert, the brother of Lothar’s first wife Theutberga. Hubert is described in the first of the two as an influential courtier (‘our beloved adviser Hubert’). Hubert then however disappears – only to crop up again in 868, but this time as a dead rebel whose property had been confiscated.
Another important figure at Lothar’s court was Archbishop Gunthar of Cologne. Although deposed by the pope in 863, Gunthar tried to cling on, and a document from 866 suggests that in that year Lothar II supported him. It states that Gunthar, though longer an archbishop, was now the gubernator et rector of the church – we might say ‘manager’ – and was making efforts to keep the Cologne clergy onside, with Lothar’s help.
But as so often with Lothar II, it’s his marriage that attracts the eye. Here’s an English translation (pdf)of one of the most evocative of his charters, granting property to a convent in Lyon and issued in the early summer of 863. This was just a few weeks before a council at Metz confirmed, with the approval of papal legates, that he could be married to his mistress Waldrada. And sure enough, in this charter Waldrada is given a prominent place, as Lothar’s ‘dearest wife’; and their son Hugh gets a mention too. Issued on the occasion of Lothar’s successful acquisition of part of his deceased brother Charles’s kingdom, and when all his marital problems seemed set fair for resolution, the charter may have marked the apogee of his reign.
Unfortunately, a few months later Pope Nicholas I stepped in to annul the Metz council, and threatened Lothar with excommunication unless he returned to his wife Theutberga. Waldrada vanishes from the charters. She reappears only in early 869, as ‘our beloved’ (without any reference though to her marital status). At that point, Lothar probably hoped that the new pope Hadrian, who had succeeded to the intransigeant Nicholas, might be more open to negotiation. But fate decided otherwise, since Lothar himself died, still a young man, just a few months later. The 863 charter survives, then, as a poignant reminder of what might have been, for Lothar, Waldrada, Hugh – and indeed for European history more broadly.
 The amazing Abbildungsverzeichnis der europäischen Kaiser- und Königsurkunden‘s page for Lothar II lists facsimiles and published photos of all the king’s charters; half a dozen of these can be viewed online, including the charter for Lyon discussed above. [29.08.20: another charter, for the monastery of St-Denis, is now available in colour from the Archives Nationales in France].
How did the Carolingian Empire, established by Charlemagne through his coronation in Rome in the year 800, differ from the Byzantine Empire centred around Constantinople (modern Istanbul)?
The ideological and cultural aspects of this enormous question have been abundantly studied for these two heirs of the ancient Roman Empire, but the issue is perhaps less clearly understood when it comes to the practical exercise of power. Historians have plenty of sources at their disposal for both these political units, but these sources are culturally embedded in such a way as to make direct comparison difficult, especially when it comes to thinking about how power worked at an everyday level, and how rulers controlled their territories.
There is however one source that not just allows but encourages us to think comparatively about how these two empires actually worked, and that is the Rižana Dispute (sometimes also known as the Plea of Rižana). You can read a full English translation, but here is a brief summary:
Around 804, Charlemagne sent three legates to Istria, a peninsula to the north-east of Italy, to resolve a dispute that had broken out there.
Istria was a region that had previously been under Byzantine control, but Charlemagne had conquered it in 788, and annexed it to the kingdom of Italy (which he had also conquered a few years earlier, in 774). This conquest had brought with it some changes in how Istria was governed: and it was to these changes that people in Istria were objecting in 804.
The source is a record of the meeting that Charlemagne’s legates arranged to sort things out at Rižana, an area now just outside the modern Slovenian town of Koper, a couple of kilometres from the Italian border. The legates summoned 172 leaders from cities and fortified settlements in the Istrian region to hear what the issue was: and these people did not hold their punches.
First of all, the Istrian witnesses complained that since the Frankish conquest, the church had started to throw its weight around much more. They began by accusing the Patriarch of Grado, the most senior churchman of the area, of evading his financial responsibilities. Then they moved onto the other bishops, who they said had been dodging their obligations, evicting people from church lands wrongfully, and generally bullying the local free population.
But their real anger was reserved for Duke John, the main secular agent appointed by the Franks. John, they said, had appropriated the revenues meant for the imperial court for his own purposes, and had also taken over a great deal of property. And that was just the start of it! He’d abolished the traditional hierarchy of offices held by regional leaders, along with the cherished privileges that went with them; he’d demanded extra humiliating services, such as feeding his dogs and onerous long-distance transport duties; and he’d raised tax demands considerably. Getting direct access to the centre of power was much harder than it had been, the Istrian leaders claimed, under the Greeks.
What does this account tell us about the Carolingian and Byzantine empires? Both were obviously agrarian empires, getting their revenues from an overwhelmingly agricultural economy. Rights over grazing livestock were particularly important in this region. So in focusing their demands for revenue on these resources, to some extent the new rulers of Istria simply stepped into the shoes of the old.
But the differences are striking, too. Power in Carolingian Istria was organised and exercised in new ways. The Carolingian regime had apparently empowered bishops at the expense of what seems like the relics of a Roman-style city-based administration maintained by the Byzantines. The Carolingian administrative apparatus was also far less elaborate: instead of a series of grand titles and offices, each with their own administrative privileges, rituals and customs, there was just the duke and his all-purposes “sheriffs” (centenarii), together with his own extended family. And, as already mentioned, the new emperor was harder to reach than the old one. To get to the court, and thus to reach Charlemagne’s attention, the Istrians had to go through mediators whom they didn’t altogether trust, whether it was a duke or a church patriarch.
The new Carolingian regime may seem therefore less sophisticated than the Byzantine administration had been. Yet despite our modern preconceptions, and as Matthew Innes has pointed out, that doesn’t mean that it was less effective at flexing its muscles. In fact, judging from the complaints, it was actually more successful at extracting revenues from the region than the Byzantine rulers had been, even if not all of the proceeds ended up at the imperial centre any more.
And, as Stefan Esders has noted, the new regime was also just as literate. The Rižana record itself demonstrates that, since it was probably the first time that the customs of the region had been written down, in this case under the auspices of the Patriarch of Grado (in some ways as much a representative of Frankish authority in Istria as Duke John).
So, a relatively unelaborated and informal apparatus of government posed no barrier to efficient and effective rule. Carolingian Istria seems to have been just as controlled, and just as exploited, as Byzantine Istria, even though (as Rachel Stone has also argued) the Carolingian Empire was organised very differently.
Can we securely draw sweeping conclusions about how power worked in two empires just on the basis of a single document? Of course not. Matters in Istria were complicated by its recent conquest. This was a frontier zone between two empires, so things may well have been very different elsewhere. But the Rižana Dispute at least gives us a glimpse into how the Carolingian and Byzantine empires were ‘experienced’ differently by the same regional community – and that’s a precious insight of which we should make the most.
For more discussion (in English) of the Rizana (sometimes also spelled Risano or Riziano) dispute, see
J. Davis, Charlemagne’s Practice of Empire (Cambridge, 2015), esp. pp. 102-4 and 274-277.
M. Innes, ‘Framing the Carolingian Economy’, Journal of Agrarian Change 9 (2009), 42-58.
F. Borri, ‘Neighbors and Relatives”: The Plea of Rižana as a Source for Northern Adriatic Elites’, Mediterranean Studies
17 (2008), pp. 1-26
 M. Innes, ‘Framing the Carolingian Economy’, Journal of Agrarian Change 9 (2009), 42-58.
 Stefan Esders, ‘Regionale Selbstbehauptung zwischen Byzanz und dem Frankenreich: Die inquisitio der Rechtsgewohnheiten Istriens durch die Sendboten Karls des Großen und Pippins von Italien’, in Eid und Wahrheitssuche. Studien zu rechtlichen Befragungspraktiken in Mittelalter und früher Neuzeit, ed. S. Esders (Frankfurt, 1999), 49-112.
I write this blog on my way back from an inspiring workshop held in Vienna on early medieval local identities (the programme is online here). A published volume is in preparation, but to whet your appetite, I’ve taken advantage of free airport wifi to provide a rough English translation of a text that was presented there by Steffen Patzold – an account of a trial at the French village of Courtisols. (You can read a recent discussion of it by Josiane Barbier in the book on Hincmar that Rachel Stone and I edited).
The text records how some residents of this early medieval village near Chalons-sur-Marne claimed to be free, but lost their case when a considerable number of their neighbours testified against them in court. It’s a great example of how an early medieval village community could be split down the middle by the intervention of a lord (in this case Archbishop Hincmar) – or, from a different perspective, how factions within a village could harness the power of the lord for their own purposes (who, after all, had started the rumour about the upstarts’ original unfreedom?).
The judgment of Courtisols, 13 May 847
“On the command of Archbishop HINCMAR, his legates – that is Sigloard the priest and head of the school of the holy church of Rheims, and the noble Dodilo vassalus of the bishop – came to Courtisols. Sitting at the public court, and investigating the justice of Saint Remi and of the already mentioned lord [Hincmar], they heard a rumour [sonus] about the mancipia whose names are given below, and about their genealogy: that they rightly ought to be servi and ancillae, because their grandmothers Berta and Avila had been bought by the lord’s price. The above-mentioned legates, when they heard this, diligently looked into the matter.
These are the names of those who were present and questioned: Grimold, Warmher, Leuthad, Ostrold, Adelard, Ivoia, and the daughter Hildiardis. They said in response “That is not so, for we ought to be free by birth”.
The already mentioned legates asked if there was anyone there who knew the truth of this matter or who wanted to prove it. Then very old witnesses came forward, whose names are these: Hardier, Tedic, Odelmar, Sorulf, Gisinbrand, Gifard, Teuderic. And they testified that their origin had been bought by the lord’s price, and that they ought by justice and law more to be servi and ancillae than free men and free women.
Then the legates asked if the witnesses against them were telling the truth. They [the mancipia] saw and accepted the truth and proof of the matter, and at once re-entrusted themselves, and re-pledged the service that had been unjustly held back and neglected for so many days, through the judgement of the scabini, whose names are these: Geimfrid, Ursold, Frederic, Urslaud, Hroderaus, Herleher, Ratbert, Gislehard.
ENACTED in Courtisols on the 4th Ides of May in the public court, in the sixth year of the reign of the glorious King Charles; and in the third year of the rule of Archbishop Hincmar of the holy see of Reims.
Sign: I Sigloard the priest was present and subscribed with my own hand to all these truthful matters. I Heronod the chancellor signed. I Dodilo signed with my own hand. Sign of Leidrad the monk. Sign of Adroin the mayor. Sign of Gozfred the advocate. Sign of Flotgis. Sign of Guntio. Sign of Betto. Sign of Rigfred. Sign of Urinus. Sign of Alacramn, Altiaud, Balsmus, Balthard, Fredemar, Tuehtar, Atuhar, Geroard, Wido, Righard, Amalhad, Rafold, Alter, Amalbert. I Hairoald the chancellor authorised and signed.
The above mentioned witnesses also proved that Teutbert and Blithelm were by origin servi, and they repledged their service in that court meeting, by the judgement of the scabini whose names are written above.”
Mancipia is a term that generally means ‘unfree people’, and that would traditionally be translated as ‘slaves’. In property transfer records, mancipia are listed as part of an estate’s assets, along with livestock and agricultural infrastructure.
 These people are listed in the estate survey for Courtisols that was made around the same time (in the polyptych of St-Remi). It is to be noted that many of them were joint tenants of holdings along with people of free status, which may well be why they claimed that they were free too.
 All these witnesses were legally-free inhabitants of Courtisols.
Scabini were residents who enjoyed a special status: something like jurors or local councillors.
 Most of these names were other residents of Courtisols.
A research project blog by Charles West (Department of History, Sheffield)