## **HORST LÖSSLEIN**

## Royal Power in the Late Carolingian Age

## **Charles III the Simple and His Predecessors**

Horst Lößlein · Royal Power in the Late Carolingian Age

## Published by Modern Academic Publishing (MAP) 2019

Modern Academic Publishing (MAP) is an initiative from the University of Cologne that contributes to the digital humanities, in the field of electronic publishing. MAP is led by Prof Dr Gudrun Gersmann, Chair of Early Modern History.

The MAP partners, the University of Cologne and the Ludwig-Maximilians-Universität Munich (LMU), are funding the open-access publication of selected dissertations and post-doctoral monographs by humanist scholars of both universities. The main target of MAP is combining the support of highly awarded researchers at an early stage of their career with new digital formats of publishing in the humanities.

www.humanities-map.net

Horst Lößlein

# **Royal Power in the Late Carolingian Age**

**Charles III the Simple and His Predecessors**

Published by Modern Academic Publishing University of Cologne Albertus-Magnus-Platz 50923 Cologne, Germany

Funded by the Ludwig-Maximilians-Universität Munich

Text © Horst Lößlein 2019

This work is licensed under the Creative Commons Attribution BY 4.0. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/. This license allows for copying any part of the work for personal and commercial use, providing author attribution is clearly stated. Figures may have different licenses. These will have been identified in the caption to relevant content.

#### First published 2019

Accepted as a dissertation at the Université de Limoges and Ludwig-Maximilians-Universität Munich 2017

Cover image: "Karolus divina propitiante clementia rex Francorum et vir illustris", detail of a royal diploma of Charles III the Simple for Saint-Maximin of Trier, issued 1st January 912 at Metz, Bibliothèque nationale de France, Latin 9264 N°12, [03.08.2015], Permalink: https://gallica.bnf.fr/ark:/12148/btv1b52505561w/f25.image (accessed 31.10.2018), CC0 1.0.

#### **Bibliographic information of the German National Library**

The German National Library lists the title in the German National Bibliography; to view detailed bibliographic information, visit http://dnb.dnb.de.

ISBN (Hardcover): 978-3-946198-48-2 ISBN (EPUB): 978-3-946198-49-9 ISBN (Kindle): 978-3-946198-50-5 ISBN (PDF): 978-3-946198-51-2 DOI: https://doi.org/10.16994/ban

Production & publishing platform provided as part of the Ubiquity Partner Network Ubiquity Press Ltd, Unit 2N, 6 Osborn Street, London, E1 6TD, United Kingdom

To read the free, open access version of this book online, visit https://doi.org/10.16994/ban or scan this QR code with your mobile device:

*Meiner Familie*

## **Contents**




## **Preface**

Die vorliegende Untersuchung wurde im Wintersemester 2016/2017 an der Faculté des Lettres et des Sciences Humaines der Université de Limoges und der Fakultät für Geschichts- und Kunstwissenschaften der Ludwig-Maximilians-Universität München als Dissertation unter dem Titel "Possibilities of Royal Power in the Late Carolingian Age: Charles III the Simple" angenommen. Für die Publikation wurde sie geringfügig überarbeitet und neu erschienene Literatur nach Möglichkeit berücksichtigt.

Die Université de Limoges und die Région Limousin haben die Arbeit an dieser Studie mit einem Stipendium unterstützt. Dankbar bin ich zudem den Monumenta Germaniae Historica in München, bei denen ich zuletzt nicht nur im Rahmen eines Editionsprojektes wertvolle Erfahrungen sammeln durfte, sondern vor allem auch Zugang zur großartigen Institutsbibliothek erhalten habe.

Zutiefst verpflichtet bin ich meinen Doktoreltern, Professor Dr. Irmgard Fees und Professor Dr. Philippe Depreux. Sie haben mir diese Studie ermöglicht, mir den Weg gewiesen, mich gefördert und mir mit Rat und Tat zur Seite gestanden. Dank gebührt in diesem Zusammenhang darüber hinaus Professor Dr. Régine Le Jan, Professor Dr. Geneviève Bührer-Thierry, Professor Dr. Knut Görich, Professor Dr. Simon MacLean, Professor Dr. Jacques Péricard und Dr. Anne Massoni für mancherlei Hinweise und Unterstützung sowie ihre Mitwirkung an der Disputatio. Hilfe habe ich auf die eine oder andere Art und Weise auch von einer ganzen Reihe weiterer Personen erhalten, von denen ich gerne die folgenden nenne: Professor Dr. Bertrand Lançon, Dr. Tobie Walther, Dr. Johannes Bernwieser, Dr. Hélène Caillaud, Sarah Ewerling, Lucile Jaeck, und Anna Nierhoff. Hervorzuheben ist jedoch vor allem Dr. Fraser McNair, dessen Korrekturen, kenntnisreichen Hinweisen und scharfsinniger Kritik ich außerordentlich viel zu verdanken habe. Auch drei weitere Personen seien besonders hervorgehoben: Natascha Wanninger, für ihre Freundschaft und für ihre sprachlichen Korrekturen, Fabien Cerbelaud von der Université de Limoges für die fantastischen von ihm erstellten Karten sowie Dr. Claudie Paye für ihre Anleitung durch den Publikationsprozess.

Herzlicher Dank geht darüber hinaus an Professor Dr. Martina Hartmann, nicht nur für die Möglichkeit in den Hallen der Monumenta Germaniae Historica zu arbeiten, sondern auch für die mir dabei eingeräumten Freiheiten und die Unterstützung.

Schließlich danke ich meinen Freunden**—**unter ihnen insbesondere Dr. Hannes Hain—die mich in den vergangenen Jahren unterstützt, mit mir gelacht und gelitten haben und mit mir ungezählte Kilometer auf verschiedensten Gewässern gerudert sind. Die letzten Worte seien jedoch meiner Familie gewidmet: Ohne sie, ihre Liebe und ihre Zuversicht hätte ich diese Arbeit weder beginnen noch beenden können.

Hamburg, Juni 2018 Horst Lößlein

## **Summary**

Charles III the Simple (893/898−923) only became king when nobles rebelling against the Robertian Odo were in need of a candidate for the West Frankish throne. Posthumously born to Louis II the Stammerer, he was of Carolingian blood and thus able to provide the rebellion with an appearance of legitimacy. The rebels on the other hand offered Charles the opportunity of a lifetime: after 14 years of being ignored by the leading nobles of the West Frankish realm, he was finally able to succeed his father. Yet, while his reign lasted for 25 years, it ended how it had started and Charles was deposed by a rebellion led by Robert of Neustria. The circumstances of Charles' elevation and deposition are among the reasons for his image as a weak king, unable to control the nobles, and why he counts as a prime example of the "decline and fall" of the Carolingian empire towards the end of the 9th century. Yet, what does "weak king" mean? Modern scholarship has long discarded the view of kingship as a question of royal orders and noble obedience. Instead, it is understood as the result of a process involving both the ruler and those around him. Successful kingship depended on the ruler's ability to integrate the nobles into this process, to mediate between their and his own interests and to create consensus.

This understanding serves as basis for this new approach to fathoming out the possibilities and limits of late Carolingian royal power. First, the focus is set on the relations between the king and the nobles around him, interpreting royal actions as the result of their interactions. Second, the customary hierarchy of the source material is inverted. Royal diplomas, ideally suited to reveal the networks of royal power, are placed at the centre of the analysis and subjected to rigorous contextualisation, treating narrative sources as secondary. Third, the timeframe of this study is extended back to the late 870s, covering the decades during which the political landscape of the West Frankish realm underwent drastic changes that determined the framework for Charles the Simple's rule. Thus, not only these developments are revealed, but also comparisons can be made.

Charles' first task after he became the sole king of the West Frankish realm was to integrate his old opponents into his rule. This meant that he had to bridge the old rivalry between these individuals and his allies from the struggle with Odo, allies who now occupied key positions in the circle around him. A dominating group of nobles agitating against their political rivals at the royal court was nothing new and can also be observed during the reigns of Louis the Stammerer and his sons Louis III and Carloman II. Under Charles the Fat this situation changed. The emperor was able to promote men of his own choice since his power base was located in the East Frankish realm and, equally important, key members of this dominating group such as Hugh the Abbot and Gauzlin died. Death also opened the door to new political solutions for Charles the Simple. In his case it was the murder of his key supporter, Fulk of Reims, which allowed him to integrate his most important opponent into his rule: Robert of Neustria.

Ensuring Robert's cooperation was certainly a crucial factor to Charles' rule. Yet, his dependency on the *marchio* (or others like him, notably William the Pious or Richard the Justiciar) should not be overestimated. Early on, Charles was able to create a network of alliances that served as a counterweight. However, to stabilise the realm in the long run, such opposition needed to be overcome. Charles' remarkable gift in integrating Robert and other powerful nobles into his rule is demonstrated by the great successes of his rule: the change in strategy towards the Northmen represented by the treaty of Saint-Clair-sur-Epte as well as the integration of the leading Lotharingian nobles into the circle around him after the acquisition of the *regnum* was accomplished without estranging those of the Western realm. This acquisition also reveals how his political room for manoeuvre had increased compared to his predecessors: given to Louis the Younger as a lease after the death of Louis the Stammerer, Charles' brothers efforts to regain the *regnum* had been thwarted by their dependency on the alliance with the East Frankish rulers to defend the realm against the Northmen and the rebellious Boso. Charles, unhindered by such alliances, was able to pursue his interests much more aggressively against his neighbours. However, the lack of such alliances meant that he missed out on their stabilising effects in regard to the relations between him and his nobles, a circumstance he tried to correct when his relations with the said nobles deteriorated.

The key to understanding this deterioration lies in the importance of trust in the relations between the ruler and the nobles around him. The rebellion against Odo was the result of a crisis of trust that developed when Odo repeatedly acted against the interests and expectations of the West Frankish nobles. Similarly, Charles also appears to have developed a strong tendency to emphasise his majesty and royal prerogative towards the end of his rule, his famous favouring of the illliked Hagano being but one example. Yet, where Odo also achieved suppression of the rebellion by taking actions that restored trust in him, Charles continued on his path up to the point where even his closest allies turned away from him. Thus, Charles' neglect of ensuring the cooperation of the nobles and the creation of consensus deprived him of the foundations of his rule and marked the limits of his royal power.

## **Abbreviations**


## **Introduction**

*Karolus stultus* ("the Stupid" or "the Foolish") and *Karolus follus* ("the Crazy"), but also *Karolus pius* ("the Pious") and *Karolus sanctus* ("the Saint")1 −these are only some of the *cognomina* that 11th to 13th century authors attributed to Charles III, who in modern times is called "der Einfältige" in German, "le Simple" in French and "the Simple" in English. The meaning of the most common byname, *simplex*, has long been analysed by Bernd Schneidmüller,2 who notes that it carried a positive meaning up until the 11th century, depicting a virtue rather than carrying the negative connotations that became dominant later and which are still reflected in its modern translations. Recently, however, Geoffrey Koziol has criticised Schneidmüller's results as revisionist.3 He argues that *simplex* is a word typically associated with monks, equated with "innocence" and thus conveys the image of a disadvantaged ruler. According to him, it was Charles' naivety which was perceived as having caused him to fall into Count Heribert II's hands−the event that branded him as *simplex* in the eyes of the medieval chroniclers.4 However, Koziol's own judgement differs fundamentally: "Charles went down in West Frankish histories as 'simple' because once defeated at Soissons, he trusted the word of the count of Vermandois and foolishly walked into a trap. But Charles was anything but simple. He was one of the most intelligent and complex of the Carolingians."5 Charles the Simple as a kind of failed genius? This assessment is far removed from the judgement of Auguste Eckel, author of the first and so far the only study of the Carolingian's reign, who depicted him as "naturellement bon, un peu faible de caractère et crédule, mais ne manquant, au besoin, ni d'énergie ni de volonté"6 and even more removed from that of Ernst Dümmler, who described him as incompetent, unwarlike and far less intelligent than his grandfather Charles the Bald, yet ambitious all the same.7 Of course, those latter characterisations are mainly based on Charles the Simple's final failure, his deposition at the hands of the nobles and his inglorious capture by Count Heribert II of Vermandois in 923. Also, perhaps, episodes described by the late 10th century authors Richer of Saint-Remi8 and Dudo of Saint-Quentin9 may have been influential.


Evaluating Charles' reign is a difficult task indeed. Louis II the Stammerer's third son, posthumously born by his second wife Adelaide, Charles disappears from the sources for ten years after his birth, although undoubtedly there would have been plenty to report about him. Passed over for the throne after the deaths of his half-brothers Louis III and Carloman II, as well as that of his relative Charles III the Fat, it is only due to the ascent of the Robertian Odo to the throne in 888 that the sources permit a second glimpse at Charles in the context of the new king's first visit to Aquitaine. Then again the sources remain silent about him for another five years until he suddenly reappears in Reims being crowned king himself by a group of nobles around the archbishop of Reims, Fulk, who were in rebellion against Odo. Over the following four years, a war ensued in which the Robertian's military superiority became increasingly evident. Nevertheless, in the end, the two parties reached an "astonishing"10 agreement: Charles not only remained king over a part of the realm, but would also succeed Odo on the throne−an event that took place only some months later, when the latter died in late 897.

The narrative sources do not tell us much about Charles' reign. The *Annales Vedastini* alongside Regino of Prüm's chronicle, our main contemporary sources for the late 9th century, break off in the year 900 while the annals of Flodoard, the sole record of the early 10th century, only resume the narration in 919. While the former are mostly neutral towards Charles, some remarks should be made on the latter. In his annals, Flodoard betrays a hidden bias against Charles. An example of this is his portrayal of Charles as a king who repeatedly violates God's laws by attacking his enemies on the most important Christian holidays and ruthlessly devastating the realm that had been entrusted to him.11 One can hardly doubt that Charles did pillage his opponents' possessions and did not halt his war to celebrate Pentecost; yet Flodoard's silence on what his enemies were doing speaks volumes. It is hard to imagine that their actions differed in any way from those of the king they tried to depose.12 While his annals are not propaganda, they are victims of the circumstances of their creation, namely the dominance of Charles' enemies.13

*Francorum igitur precibus compulsus, jussit cuidam militi pedem regis osculari. Qui statim pedem regis arripiens, deportavit ad os suum, standoque defixit osculum, regemque fecit resupinum. Itaque magnus excitatur risus magnusque in plebe tumultus.*


#### Introduction 3

Thus, Flodoard's account does not lose its credibility, but we should be aware that as often as he reported some events, he also turned a blind eye to others.14

As mentioned, for most of his reign there exist hardly any sources at all. Therefore, we know about two of the main events of Charles' rule almost exclusively from the meagre accounts composed east of the Rhine, later narratives, like those of the aforementioned Richer of Saint-Remi and Dudo of Saint-Quentin and Charles' royal diplomas. In an event traditionally dated to 911, after the battle of Chartres, Charles granted the Northmen under Rollo a territory that was to become the heart of the future duchy of Normandy by the treaty of Saint-Clairsur-Epte. Later during the same year, Charles became king of Lotharingia when the nobles of this *regnum* chose the Carolingian over the new king of Eastern Francia, Conrad I. Flodoard alone provides more detailed information about the last years of Charles' reign, which were overshadowed by different conflicts. In Lotharingia, the king fought Count Gislebert while in the West the nobles around Odo's brother Robert rebelled against him. This latter conflict proved to be fatal for Charles' reign: the nobles made Robert their new king and even after his death shortly afterwards in the battle of Soissons, they chose his son-in-law, Raoul, over Charles. Seeking new allies, Charles put himself in the hands of Count Heribert II of Vermandois, was imprisoned and was then used by the count to exert pressure on Raoul until his death in 929.

It may be because of the general lack of sources for Charles' reign that scholars have devoted little attention to him and his reign. Since Auguste Eckel produced a biographical analysis in 1899,15 only Geoffrey Koziol has published a number of articles16 and dedicated parts of his monography on the politics of memory and identity to him.17 That is not to say that Charles' time, its problems and its general developments have been completely ignored by scholarship. A number of studies on specific aspects of this period have been brought forward, covering key political figures like Archbishop Heriveus of Reims,18 Count Heribert II of Vermandois19 and Count Hagano20 or key events such as the conflict around the episcopal siege of Liège in 920/92121 or the conversion of the Northmen under Rollo. 22 Most commonly, however, Charles has been treated as one amongst many rulers in general surveys23 or handbooks.24

14 For the biases of Flodoard's other great work, the History of the Church of Reims, see chapter III.3.


<sup>15</sup> Eckel, Charles.

Scholarly discussion about the reign of Charles the Simple revolves around three main axes. The first concerns the general state of the realm, that is to say, the balance of power between the king and the nobles. The old view, emphasising the infidelity of the nobles towards the king as the main problem of Charles' reign,25 reading royal-noble relations as a constant struggle for power,26 has long been challenged and modified. For example, Bernd Schneidmüller, although still close to the old readings, has brought forward the creation of a new ministerial class created from the Lotharingian lower nobility by Charles, which was supposed to act as a counterweight to the powerful higher nobility.27 More influential was Karl Ferdinand Werner's proposal of cooperation between the king and the highest nobles in relation to the emergence of the principalities in the West Frankish realm. According to him, in return for giving those powerful men a vice-regal position, Charles would have gotten their support and thus stabilised his reign.28 While the overall image of powerful nobles becoming *marchiones* and controlling large parts of the *regnum* has been generally accepted,29 Werner's suggestion that this system was consciously and purposely created by Charles has been criticised. Instead, it has been emphasised that the foundations of this development had already been laid earlier, during the reigns of his immediate predecessors30 or even during the reign of Charles the Bald31 and that his grandson Charles the Simple only slowly succumbed to circumstance.32 Thus, his rule is not seen as having been built upon the investiture of powerful individuals as his surrogates, but founded on changing alliances with different groups of nobles.33 Further criticism of Werner's hypothesis highlights the differences in Charles' control over the various regions of the realm. While intense in the area north of the Loire, south of the river his rule would have been no more than nominal.34

A second line of thought revolves around the problem of Charles the Simple's legitimacy and its influence on his position as king. His father's two marriages and their validity have always been subject to scholarly study,35 yet whether this actually played a role in late 9th century politics remains controversial. Those emphasising its importance claim it was the reason why Charles had been passed over for the throne, at least in 883/884,36 or why he underlined his Carolingian decent


<sup>25</sup> Eckel, Charles.

once he had become king in order to strengthen his rule.37 Charles certainly seems to have possessed a strong consciousness of his family background and of questions of legitimate and illegitimate birth—according to Geoffrey Koziol, this latter also resulted in a distinct dislike of his half-brothers Louis III and Carloman II. 38 However, while this emphasis on his Carolingian heritage seems to be the general consensus of scholarship, the political significance of Charles' problematic legitimacy is more often doubted than stressed. In particular, other reasons have been brought forward concerning its role in the royal succession,39 thus questioning the political significance of Carolingian legitimacy for the late Carolingian age.

A third discussion centres on the reasons for the final failure of Charles' reign, which culminated in his deposition. Some scholars attribute the rebellion of the nobles solely to the promotion of Charles' intimate Hagano. 40 Others interpret the king's favourite as an example of a general policy pursued by Charles, emphasising the royal prerogative to choose his councillors, as opposed to the seemingly more successful strategies of Berengar I or Henry the Fowler, who acted as *primus inter pares.*41 Still others argue for a change in the structure of Charles' rule, caused by more aggressive royal policies against the leading nobles.42 This view of a change in royal politics has been criticised by Geoffrey Koziol as an argument *a silencio*  based on the lack of sources for most of Charles' reign. In turn, he argues that Charles' rule was characterised by a constant rivalry between the king and Robert of Neustria.43 In this context, special roles have been attributed to the effects of the treaty of Saint-Clair-sur-Epte as well as to the acquisition of Lotharingia. Thus, Rollo's investiture with the *pagi* along the Seine is sometimes interpreted as an effort to establish a counter-balance against Robert. 44 Similarly, the extension of Charles' rule over Lotharingia, while sometimes seen as having had a stabilising effect on his reign,45 is more commonly interpreted as having had a negative influence on his relations with nobles from the Western realm. Royal presence in Lotharingia is supposed to have led to a shift of Charles' main power base away from Western Francia, which resulted in him being estranged from the Western nobles,46 or, at least, in an increased competition for influence at the royal court, thus destabilising the newly found balance.47


This reading of Charles' reign as a struggle for power between the king and the leading nobles points to the wider discourse among scholars over the structures of Carolingian rule and the diminution of royal power over the course of the 9th century. Step by step, the boundaries for the beginning of this decline have been moved from the last years of Charlemagne's reign48 to the end of that of Louis the Pious, 49 then to the death of Charles the Bald50 and finally to the deposition of Charles the Fat. 51 However, there are some important gaps left in the study of late Carolingian kingship. Leaving aside Charles the Simple, for Louis the Stammerer and his sons Louis III and Carloman II no published studies exist at all52 and the sole monograph on the reign of Eudes dates back to 1893.53 This lack of modern analysis may be a reason why the period after 888 is often still assessed as a time when royal rule is supposed to have suffered from a crisis of legitimacy and authority.54 Especially concerning the reign of Charles the Simple, these assessments are often particularly pessimistic, with Gerhard Schmitz's (already rather dated) conclusion about the restraints placed on his rule representing the low point. According to him, Charles' "Handlungsfähigkeit nach innen und nach außen [war] nicht nur begrenzt, sondern fast aufgehoben. Letztlich hatte sein Königtum nur so lange Bestand, wie die großen Vasallen es zu tolerieren bereit waren; Unterstützung konnte Karl von ihnen nicht erhoffen, einen Konflikt mit ihnen aber auch nicht riskieren, denn dessen Ausgang war bereits zu Ungunsten des Königs vorprogrammiert."55

#### **Possibilities and limits of late Carolingian rule**

Schmitz aims to capture what in more general terms can be described as the king's room for manoeuvre or, to apply another term, as his *Handlungsspielraum*. This latter concept provides a great basis for the exploration of the limitations a ruler was subject to and the possibilities available to him in political situations. However, while this term is frequently used especially by German historians, attempts to actually define its meaning and to use the concept behind it as an analytical tool are, given the number and variety of publications on the matter, very rare.56


<sup>48</sup> Ganshof, Fin.

"Political *Handlungsspielraum* is the sum of possible options of action which can be implemented successfully by the political decision-maker"57 is a simple definition used by political scientists. The existence of *Handlungsspielraum* is therefore bound to two conditions: a) a political situation requiring the political actor to become active and b) this actor having at least two different options that may lead to success and that he can choose from.

However, the range of options available to the actor is subject to certain constraints.58 One of these is the human limitations of the actor himself. The knowledge available to him is limited, as are his cognitive abilities to process this knowledge in regard to the possible consequences of his options. Thus, he only acts under "bounded rationality"59 Hence, human nature itself limits the actor's options by preventing him from knowing about all of them and everything about them. Yet, there are also other constraints to his pool of options. In sociological approaches *Handlungsspielraum* defines the degree of freedom of an individual embedded into a society. "The more choices this individual has in decision situations, the more limited the restrictive conditions, the bigger is the *Handlungsspielraum*."60 Thus, the decision-making individual is conceived as part of a bigger group, a society, which in turn limits his options. The practical philosopher Werner Stegmaier has included these restrictive conditions in his own definition: "A [*Handlungs*] *spielraum* is the 'space' of a movement limited by rules, in which another, 'playful' movement, not subject to these rules, in this sense a 'play' free of rules, becomes possible; in short: the *regulated limits of unregulated behaviour*. 'Within' these limits the behaviour may very well be subject to its own rules."61 Thus, if an individual decision-maker is part of a greater community, he is subject to the rules existing in this society. Each of his actions underlies these predefined rules, yet, within these rules a certain space for own choices exists, which may possibly be dependent on other rules. The rules existing in a society are defined by whether other members of said society will consider the actual action decided upon as normatively and cognitively appropriate or inappropriate.62 Therefore, as a set, the rules existing

Resch, Freiheit, gives some thoughts on the restrictions of individual *Handlungsspielräume*. Notable exceptions are Auge, Handlungsspielräume and Poguntke, Handlungsspielräume.

57 "[Politischer Handlungsspielraum ist] die Menge möglicher Handlungsoptionen, die politische Handlungsträger umzusetzen in der Lage sind." Sattler and Walter, Handlungsspielraum, 465.


in a society constitute the framework defining appropriate behaviour within this society. Thus, each decision taken is the result of a process of negotiations to which rules apply and regarding which of those rules are to be considered stronger.63 In some cases, this framework limits the possibilities of how to act in certain situations, while in others a large pool of options may exist. The range of these options within a given framework is expressed by the term *Handlungsspielraum*.

However, these frameworks are not static.64 Rules may change over time and actions that have been considered acceptable before can, in similar situations at later points, become unacceptable and *vice versa*. Especially in times of crisis and disorientation, the existing framework is questioned and shifts within it may take place.65 It would also be wrong to assume that at any given time only one framework exists which determines the appropriateness of actions for any member of the society. In fact, different frameworks may coexist at the same time, each with its own set of rules. An example for these would be rules of behaviour stemming from one's perception of oneself, that is to say, from one's identity. These often collide with frameworks provided by society, thus leading to behaviour outside the limits of what is generally considered appropriate.66 This is especially the case in less stable and well-defined political systems,67 such as those existing in the late Carolingian era.

Hence, sounding out *Handlungsspielraum* is a process of negotiation between different systems of rules. However, this implies that there is a certain degree of flexibility inherent to these rules. Following Philippe Buc, it seems appropriate not to speak of "rules," but rather of "norms" to emphasise this aspect.68 When referring to the realisation of these processes of negotiation, we can go even further and introduce another nuance. In the end, these processes are not only determined by the norms of the participating actors, but by their expectation that their opposite knows, acknowledges and acts not only according to his own norms, but takes also those of the other participants into consideration. This expectation can be described as trust, with the assumed probability that this expectation will be met by the other according to the degree of trust in the other.69 Hence, the existence of a certain degree of trust can be described as the prerequisite for negotiating since

macro-environment described by Lagroye (the Church or another big organisation partaking in the political competition), this influence would be rather limited when it comes to individuals.


it ensures the willingness to cooperate70 and thus becomes an important part for determining one's *Handlungsspielraum* within a society.

When determining the possibilities and limitations of late Carolingian rule this leads to another question. The result of these negotiations is not only a question of the norms of each individual actor and of the trust existing between the participants. The strength of the individuals' identities also plays a role, as does the assessment of the situation as well as the available resources to realise any option.71 The importance of the last point is revealed when taking into consideration that this process of negotiation was not always one of equal interaction in the late Carolingian age. Late Carolingian politics were a constant flux of negotiations between the interests of the king and the nobles and, therefore, as Simon MacLean put it, "potentially unstable."72 Each of these negotiations can be read as sounding out the possible ways of action, for the ruler as well as for the other nobles participating in the decision-making process or affected by their results. If the latter disliked the royal actions, they would offer resistance, either forcing an adjustment to or even change of the royal politics (possibly even by changing the king himself) or else being forced by the king to submit to his view. Thus, the royal *Handlungsspielraum* was not only determined by norms, but also by the capacity the king had to enforce his own norms over those of others. This even goes so far as to limit the importance of trust: overwhelming power on one side could simply leave no other option than to submit. The ratio of the power between the ruler and the nobles opposing his politics thus defined the possibilities and limits of royal power. When asking for these possibilities and limits during the late Carolingian era, it is therefore best to start by analysing the actual functioning of royal power during this time.

#### **The functioning of royal power**

The traditional starting point in analysing the potential power of any Carolingian king is the royal fisc.73 The lands under direct royal control had originally been created by taking over the Roman *civitates* and imperial estates during the 5th century and later supplemented by the Carolingian family lands when they superseded the Merovingians as the ruling dynasty. From the fisc, the kings drew the resources to equip their soldiers, to supply the court on its travels through the realm and to maintain their palaces and estates, the latter standing as reminders of

<sup>70</sup> For the importance of trust in interpersonal relations, see, for example, Marzano, Confiance.

<sup>71</sup> March and Olsen, Institutions, 22; March and Olsen, Logic, 492−493 and Gulati and Srivastava, Agency.

<sup>72</sup> MacLean, Kingship, 96.

<sup>73</sup> On the problems associated with the fisc, see Barbier, Fisc. Fundamental also Brühl, Fodrum and Barbier, Palatium.

their authority even in their absence.74 Trying to establish the extent of the fisc at any given moment, however, poses problems that are impossible to solve. Almost all of our knowledge concerning royal property derives from the royal diplomas donating parts of it to individuals and institutions,75 telling us only which parts the king had decided to part with and not how much he actually possessed and controlled. Yet, it seems clear that it was concentrated around the palaces and in the core regions of royal power,76 while those parts given away were often situated outside these heartlands, where they would have been difficult to control.77 But the fisc did not simply undergo a process of continual diminution through grants; it was also replenished by acquisition, inheritance, gifts or confiscation of property in cases of treason, all of which were hardly ever documented by royal charters. Thus, the fisc was in constant flux and, given the gaps in our sources, any attempt to reconstruct it at a certain moment or shorter period must prove futile.78 In any case, the landed possessions only provided part of the royal income. Other wealth was also drawn from different fiscal rights on trade, such as tolls and markets, or from minting, which is equally impossible to quantify.79 In times of war, plunder and tribute could be obtained from the defeated, which provided an important means of rewarding armies.80 Finally, becoming ever more important during the second half of the 9th century, the kings could also access the resources of ecclesiastical institutions, churches and abbeys, both being under royal protection and therefore open to royal influence.81

Counts were charged with the care of the royal estates, as well as keeping the peace in their counties, doing justice, putting royal decrees into practice and raising levies for royal campaigns, thus transmitting royal power to every corner of the realm.82 West of the Rhine, these counties appear to have been often centred on the old Roman *civitates*; east of the river, they were newly created and oriented towards landscape and settlement structures. However, within these counties enclaves existed, chiefly landed possessions of ecclesiastical institutions that had been granted royal immunity. Still others were created by the control of the unfree by the free. Since counts only had jurisdiction over the free, those areas which


were inhabited only by the unfree remained outside their direct control.83 While the importance of these offices is clear, more interesting for us is the question of royal control over the office holders. One possibility for the kings to exercise control was to install counts of their choosing. However, the ability to do so appears to have been limited. Most of these cases documented by the sources occurred when the county system was introduced in newly conquered territories and even then, the choice often fell on local nobles, since they had to rely on their own resources and connections to fulfil their tasks.84 Another possibility was to take control of a county after its count had died. For a long time, scholars assumed that, from Charles the Bald's capitulary of Quierzy at the latest,85 the office of count had become hereditary.86 While this reading misses the true character of the clause, which in fact confirms the royal prerogative to control succession,87 nevertheless, the passing of a county from father to son already appears to have been common practice for a long time.88 Actual royal control over the countships appears to have ranged between these two extremes. The redistribution of counties was a generally accepted right of the king, as long as he could claim that the office holder had failed to properly fulfil his duties.89 Kings certainly claimed to exercise influence in such matters, yet whether they were actually able to implement their view differed from case to case.90

Next to the counties, a parallel and at least as important administrative system existed based on ecclesiastical organisation. Episcopal power was typical of the way Carolingian society worked, as it was situated at the frontiers between the profane and the sacral worlds and thus reached into both.91 Apart from their spiritual responsibilities, within their dioceses bishops were also responsible for keeping public order and speaking justice. Like the counts, they participated in public assemblies and political decision making and provided, probably more reliably than the counts,92 substantial resources for royal politics.93 Like counts they acted as royal legates, like counts they acted as conduits of royal power. However, unlike


that of the counts, episcopal investiture was officially regulated by canonical law, that is to say, their election resided with the clergy and people of their respective dioceses.94 Reality, of course, was different, as Carolingian kings often aimed to install their own candidates.95 For example, Charles the Bald exercised close control over the sees of Reims, Sens and Rouen.96 Therefore, while the royal role in these elections led to conflicts with important ecclesiastical dignitaries, most often both parties appear to have sought a consensual agreement, resulting in the West Frankish realm in the integration of the royal prerogative into the canonical procedure of the appointment of a new bishop.97 Besides the king, the local nobility also had its say in the elections, trying to raise their own candidates, often relatives, to the episcopal throne. This did not necessarily counteract royal aspirations since the member of a local family might well be the choice of the king too if said family belonged to the circle of his own supporters.98 Thus, episcopal succession presents itself as the competition of various, sometimes opposing, interests, yet it was one with seemingly more possibilities for the rulers to install candidates of their choice.

Limiting our view to resources and offices, however, fails to portray the functioning of Carolingian kingship. As has already come through in the previous comments, we can make out its true basis: relations between the king and the nobles.99 Counts and bishops provided resources and transmitted royal power even into the distant regions of the realm, but royal control over both was limited. The kings tried to influence elections and install their own candidates, yet to succeed, they depended on the cooperation of the local nobility. In the words of Martin Gravel: "L'Empire carolingien doit être conçu comme un réseau de relations, comme l'enchevêtrement des liens du baptême, du sang, de l'alliance, de l'amitié, de la servitude et d'une multitude de fidélités."100 Only by creating bonds between themselves and the nobles, by creating consensus about their kingship in general and each individual decision, could their rule succeed.101 Kingship was not simply the giving of orders, but necessarily needed to show consideration for the interests of those involved. In consequence, ruling consisted of constant


negotiations between different interests and each balance found between these interests was fragile and in need of constant care.102

A key factor for the relations between the king and his nobles was the distribution of *honores*, that is to say, land and offices.103 The former could be part of the fisc, belong to abbeys or churches or be confiscated from disloyal nobles; the latter could be counties or, more often, abbeys, commonly in the form of lay abbacies. Receiving a donation from the king reinforced the bond between himself and the receiving noble,104 who then acted as a conduit of royal power, transmitting the ruler's will even to the frontiers of the realm. This does not mean that donations created loyalty. Far more often, they are instead the sign of already existing ties. For example, in border regions, small and unimportant abbeys were granted to influential nobles to install trusted men in potentially endangered zones, thus ensuring royal control over these areas.105 Hence, giving away *honores* served to create and expand networks of power consisting of a larger number of individuals who would lend their support to the king. These networks were the means to exercise control over the realm and to transmit royal power even into regions that hardly ever saw the king in person.106

But to reduce the relations between the king and the aristocracy to a mere model of giving (making donations) and taking (withdrawing *honores* in the case of infidelity) would mean to limit our view. Besides *honores*, influence, partaking in the decision-making process, could draw the nobles to the king. In the ideal court as described by Hincmar, the king was surrounded by close advisors with whom he discussed the affairs of the realm before deciding on which measures to take.107 This circle only consisted of those closest to the king, those in the possession of the most *Königsnähe*. 108 As Hincmar notes, these men close to the king participated in the decision-making process and thus were able to influence its outcome.109 Therefore, being close to the king directly translated into political power. As we have already noted, the nobles close to the king served as conduits, transmitting the royal will into the regions of the realm. This also worked the other way round, with locals coming to the respective noble asking him to bring their


106 Wickham and Reuter, Introduction, 1−16. On these networks, see also Althoff, Verwandte 134−181 and Werner, Adelsfamilien.


<sup>109</sup> On the role of advisors and their influence see also Althoff, Kontrolle.

wishes to the king.110 Thus, having access to the king meant to possess the ability to grant this access also to others or, in other words, to exercise control over the access to the king.111 Like participating in political decisions, this was a further guarantee of political power since, on the one hand, it allowed the noble in question to influence the topics being discussed in the circle around the king and, on the other hand, strengthened his position at home. Since only he could provide access to the ruler, the petitioners needed to court his favour. Being close to the king and thus managing the affairs of the realm, being able to influence the decisions made at court and controlling the access to him was certainly an incentive for cooperating with the king that was at least as strong as getting donations from him.112

These thoughts can be developed even further: Stuart Airlie has argued that royal service not only served noble self-interest in *honores* and offices but also became a constitutive factor of noble identity.113 Belonging to those around the king, counselling him and serving him, meant being part of the most privileged elite, a status that elevated them not only over their peers but also over their relatives those with whom they were in competition for social advancement and proximity to the king. Their own and their family's past actions served as sources of legitimacy and so did the prospect of future deeds done in service of the king, creating a line of continuity stretching from the times past to the times to come. Social status depended on how one was perceived by other nobles around oneself, and for contemporaries the significance of one's family depended first of all on its connections with the royal household and its access to the court.114

However, this leads to another problem: who belonged to these other nobles? Trying to find a convincing, all-encompassing definition for "aristocracy" proves to be a fruitless undertaking. To be noble meant to belong to the leading members of the society. In some cases this was undisputable but in others it was hard to draw the line. "'Nobility' was a moral distinction, a moral distinction which (as moral distinctions usually do) tended to reflect social status. To be 'noble' in this sense was to exercise social power in the proper manner."115 Thus, the aristocracy consisted of individuals for whom belonging to this group depended on their own behaviour and the acknowledgement of their status by other members of the same group.116 Access to this group was open while within it a hierarchy existed in which advancement was possible.117 When talking about the aristocracy, we there-


fore should avoid imagining one single block in opposition to the ruler, but rather think of individuals and groups consisting of relatives and friends118 competing with each other for social advancement.119

Crucial for social advancement was the cooperation with the king. As Régine Le Jan expresses it: "Le roi carolingien fut le pivot de la circulation des honneurs et des richesses, contrôlant la hiérarchie sociale par le système du don, imposant la mobilité et la fidélidé en confisquant les honneurs pour cause d'infidélité."120 The king was the pivot in the competition for influence over the affairs of the realm as well as the source of noble identity. He could take advantage of his position within this system by choosing his councillors and promoting them at his pleasure. Yet, there were limits to his choices. Having reached a certain level within the hierarchy, the nobles not merely passively hoped for the king's goodwill. From their own self-conception they derived a right to be treated according to their position, to take part in the affairs of the realm and to receive *honores*. 121 What is revealed here is a complex system of sometimes diverging claims. On the one hand was the king, exerting his right to choose whom to favour and further, on the other hand, was the noble, deriving from his already elevated position the belief that he needed to be chosen by the king. This system of relations between the king and the nobles was an unstable one, one that needed constant care and one in which there was often no ideal choice which could satisfy all parties involved, only one that would limit the damage done within a bigger part of the network.

Our comments up to this point have been aimed at casting some light on the functioning of royal power and the determining factors in the relations between the king and the nobles. However, the political landscape of the late 9th century underwent some important changes that need to be taken into consideration when studying this period. Since the reign of Charles the Bald, within Western Francia some nobles had created large conglomerates of *honores* that were to become the future principalities.122 The original perception of these developments by Jan Dhondt,123 who promoted the idea of the "rise of the aristocracy" as a struggle of two opposing powers—the Carolingian kings and "the aristocracy"—has since been questioned and modified. For example, Karl Ferdinand Werner emphasised royal consent in this development, describing the principalities as sub-kingdoms led by nobles in vice royal positions who exercised powers delegated by the kings.124 Furthermore, as Simon MacLean in his study on the reign of the Emperor


<sup>118</sup> On such groups, see Althoff, Verwandte; Althoff, Amicitiae; Le Jan, Famille, 77−85 and Epp, Amicitia.

Charles III "the Fat" has explained: "Carolingian kings and emperors had always been accustomed to dealing with powerful aristocratic individuals, and not with the 'aristocracy' (or individual aristocratic families) as a monolithic entity."125 The rise of powerful individual nobles was not something that was only typical of the late Carolingian period. Kings already had had to cope with such individuals and managed to integrate them into their rule. Therefore, we should not perceive this period as a constant struggle between the "aristocracy" and the kings, but rather as one in which their cooperation, their joint work towards a well-ordered world, should be emphasised.126

This is not to claim that the framework of late Carolingian politics was static. Duke Boso, who in 879 convinced the local nobles at Mantaille to elect him king, struck a first blow against the functioning of this system. Although married to a Carolingian princess, he destroyed the illusion of a Carolingian monopoly on kingship and made the office targetable for other ambitious nobles.127 In addition, noble families became more and more autonomous from the royal centre as the noble self-conception became increasingly detached from royal service and instead depended on family traditions and property rights.128 They now developed a dynastic identity and began to ascribe their power not to royal generosity but to God's will.129 This effect became pronounced when in 888 kings came to power in the former Carolingian *regna* whose kinship with the Carolingian family was at best problematic. To name only two examples, in Eastern Francia, Arnulf of Carinthia, an illegitimate son of Carloman of Bavaria, now assumed power while in Western Francia, Odo, a Robertian without a single trace of Carolingian blood, made his move for the throne. Carolingian legitimacy remained a strong argument in the political theatre130, but from now on one could also become king by possessing sufficient power to impose oneself on the other nobles.

These two developments, the rise of individual nobles amassing *honores* with royal consent on the one hand and the loss of the Carolingian monopoly on the throne on the other, may have commenced independently of each other, yet by the end of the 9th century, they were inseparably linked. Boso had only been able to claim a crown because he had already risen to a position from which he was able to do so, yet he had chosen a moment in which the Carolingian monopoly still stood unwavering. However, his rebellion was a first crack in the dam which broke nine years later when dynastic coincidences had done away with all legitimate male Carolingians north and south of the Alps, apart from a nine-year-old boy, Charles the Simple. These developments, the weakening of Carolingian legitimacy and the increasing power of some nobles in combination with their growing emancipation

<sup>125</sup> MacLean, Kingship, 77−78.

<sup>126</sup> West, Reframing, 100−102.

<sup>127</sup> Airlie, Semper fideles. On Boso's rebellion, see also chapter VI.4.

<sup>128</sup> Airlie, Semper fideles, 139−140.

<sup>129</sup> Le Jan, Royaume, 90.

<sup>130</sup> Le Jan, Royaume, 88.

from the crown, led to a shift in the relations between the king and the nobles which started to unhinge the old system.131

#### **Analytical approach**

While it is undeniable that such shifts took place, the basis of royal power did not fundamentally change. Its basic principles remained intact and ensured the cooperation of the nobles. The better they were integrated into the king's rule, the better the royal will was transmitted even into the most distant corners of the realm. When it comes to determining royal influence, this understanding also has a major impact on the value of the royal itinerary. It has been rightfully noted that in the late 9th/early 10thcentury, West Frankish kings were limited in their travels to the north-east of their realm,132 yet the conclusion that their sphere of action was as limited is misleading. As long as the king was in contact with nobles from the regions outside the radius of his movement, he could influence the local politics. Thus, analysing royal power means that an approach focused merely on direct royal actions is too limited. Defending the realm, dealing with conflicts and ensuring the royal prerogative certainly can be used as benchmarks for the possibilities of royal power,133 yet these actions are only the surface of the underlying network of nobles centred around the ruler. An analysis of the possibilities and limits of royal power has to take these two levels into consideration.

Networks can be understood as ensembles consisting of individuals (or groups) who are linked with each other by social relations,134 or in other words, are expressed by the sum of their interactions and connections.135 While they do constitute rather stable collectives,136 the individuals belonging to a network can compete with each other.137 This means that networks are dynamic structures which develop over time and which may become fragile up to the point of breaking even under the occurrence of only minor pertubances.138 Therefore, when analysing networks the focus has to be set on the nature of the relations between the different actors,139


or in the case of this study, on those between a ruler and the nobles in contact with him. Only by analysing these in regard to continuity, intentionality and quality can the position of an individual within a network be determined.140 Based on this analysis, within a network different zones, or layers, can be identified, as defined by these relations:141 while frequent and intense contacts between a ruler and a noble indicate the importance of this noble for the ruler and their central position within the network, a single contact might not even be deemed sufficient to place a noble within the network at all.

This leads to a key problem of network analysis for historians and especially for those studying early medieval history. Even in best-case scenarios, the sources available are so scarce that it becomes questionable whether the identification of general structures and developments is possible at all.142 While the scarcity of sources is especially true for the late Carolingian era, there is one type of source that offers considerable advantages when it comes to analysing networks and determining the relations between individuals: charters and specifically royal diplomas.143 They not only allow us to establish relations between individuals (in the case of the royal diplomas the ruler and the nobles petitioning and receiving these), but also offer insight into the quality, frequency and intensity of these relations.144 Therefore, these royal diplomas form the basis of those parts of this study dealing with the networks of royal power while narrative sources are used to complement and contextualise the information gained from the diplomas. This approach also allows us to mitigate the problems caused by the gap in said narrative sources between 900 and 919.

While this study is dedicated to exploring the possibilities of Charles the Simple's royal power, his rule cannot be fully understood without going further back


to the 870s, to the times of his father, Louis II the Stammerer, his brothers Louis III and Carloman II, to the Emperor Charles III the Fat and finally to the Robertian Odo. It was during their reigns that the frameworks determining at least the early years of Charles the Simple's own developed. Taking the study back to the years following the death of Charles II the Bald allows us to make comparisons between Charles the Simple's reign and those of his predecessors and to reveal the continuities and changes without having to revert to the questions of "structures" often used in research on the 9th and 10th centuries. It allows us to analyse the reasons for Charles' very late succession to his father and the basis of his kingship (chapter I), to reveal the networks of royal power and the ruptures they underwent before and during Charles' reign (chapters II and III) and to use the results gained from these to assess the rulers' actions towards their neighbours, the Northmen, and their own nobles (chapters IV, V and VI).145 Each chapter of this study adds another layer to the analysis of the relations between rulers and nobles and thus to the possibilities and limits of royal power in the late Carolingian age.146

<sup>145</sup> On the importance of relations between different actors, for the context of their actions and the analysis of their room for manoeuvre, see Preiser-Kapeller, Calculating, 105.

<sup>146</sup> This approach means that we will return to the same events at numerous times over the course of this study, leading to certain redundancies. Given the advantages of being able to assess each event and each action from different perspectives and in different contexts, this is a price we consider to be worth paying.

## **I. Becoming king: The questions of legitimacy and support**

In hindsight, Charles III the Simple's accession to the throne appears as a surprise. Born posthumously as Louis II the Stammerer's third son, he was passed over in the royal succession no less than four times until finally, a rebellious noble faction chose him as king. Although the ensuing war was lost, he was nevertheless confirmed as king and succeeded the victorious Odo shortly thereafter. This brief account raises a number of questions which serve as a starting point to understand Charles' rule and the possibilities and limits of his royal power. Why was Charles passed over in the royal succession so many times? Or, in other words, why did the nobles rebelling against Odo chose Charles as their king when seemingly his claim to the throne was weak enough to be ignored until that moment? This, in turn, leads to the issue of the composition and structure of the network these nobles formed and to the question of their various motives. Essentially, this means examining the basic conditions under which Charles became king: to evaluate the nature of his legitimacy as well as its importance as an argument in the politics of the West Frankish realm and to analyse the composition and cohesion of the network of nobles elevating him to the throne.

### **I.1 Charles' claim to the throne**

When Archbishop Fulk wrote his letter to King Arnulf of the East Frankish realm in 893, announcing Charles' elevation to the throne and asking for the king's support, he made some remarks concerning certain rumours that had apparently reached Arnulf 's ears, according to which Charles was not Louis II the Stammerer's son. Fulk, however, assured the king that these rumours were unfounded, and that in fact the young Carolingian resembled his father in a way that made his descent above suspicion.1 From what we know from the sources, these rumours, probably spread by Odo's supporters to contest Charles' claim to the throne, were never spread again. It is nevertheless interesting to note the way they were constructed. Undoubtedly referring to his posthumous birth, they denied Louis' parentage and consequently Charles' Carolingian blood, the basis of his claim. This strategy certainly made sense, yet it is surprising that they did not attack the youth's legitimacy in any other way.

<sup>1</sup> Flodoard, HRE IV, c. 5, 382: *Denique, quod audierat ipsi Arnulfo dictum fuisse, quod hic Karolus filius Ludowici non fuerit, asseverat neminem se posse credere fore, qui, eum si viderit et parentum ipsius effigiem cognoverit, non recognoscat illum de regia processisse progenie; quedam quoque patris sui Ludowici signa gestare, quibus agnoscatur filius ipsius fuisse.*

#### **I.1.1 Louis the Stammerer's two marriages**

The problem of the legitimacy of Charles' birth, or more precisely that of his father's two marriages, has long been the subject of scholarly discussion.2 Louis the Stammerer and his own father, Charles II the Bald, had had a difficult relationship. Several times, the son had revolted against the father, been reconciled and then revolted again.3 During one of these cases in 862, Louis, following the advice of some of his supporters, had married Ansgarde, daughter of Count Harduin and sister of Odo, count of Châteaudun, one of his close supporters.4 Louis' revolt was subsequently put down, but the marriage with Ansgarde remained, soon to bear fruit in the form of two sons, Louis III and Carloman II, as well as a daughter, Hildegard. Some time later, Charles forced his son to abandon his wife and marry Adelaide, daughter of one of Charles' own close supporters, the Burgundian magnate and count of the palace, Adalard. 5 This count's importance is also underlined by his appearance in the capitulary of Quierzy from 877, in which Charles laid down preparations for his journey to Italy: Adalard was among those appointed to stay close to Louis during the emperor's absence and was also in charge of the royal court of justice.6

According to Karl Ferdinand Werner, Adalard, and consequently Adelaide, came from an old family, going back to Bego, count of Paris, who had restored and reformed the abbey Saint-Maur of Fossés and died on the 28th October 8167 , and his wife Alpais, daughter of Louis the Pious. In addition, Bego's nephews, Count


Gerard of Vienne8 and Adalard the Seneschal, uncle of Charles the Bald's first wife Ermentrude, 9 were influential under Charles the Bald. Bego's son Leuthard held the county of Paris and his second son Eberhard became well established in northern Burgundy, while his daughter Susanna10 married a certain Wulfard, who fathered yet another Wulfard, count of Angoulême, and finally Adalard, Adelaide's father.11 Adelaide also had a brother, also named Wulfard, who became abbot of Flavigny and was a member of the royal chancellery.12 The family also branched out into southern Lotharingia13 and had relatives in Aquitaine.14 Louis' marriage to Adelaide thus tied him to an important family with close links to his father. It also ensured the continuing influence of the faction of nobles Adalard belonged to, even after the death of Charles.

However, the circumstances of the two marriages gave rise to questions concerning their validity and thereby also the legitimacy of the offspring produced from each. Hincmar's *Annales Bertiniani* refrain from even mentioning the subject. Neither the divorce nor Charles' birth is mentioned and while the archbishop refers to the queen, he does not use her name, thus giving the impression that nothing had changed.15 The archbishop did not entirely succeed in his endeavour to hush up the affair, however. When treating Louis' meeting with Pope John VIII at Troyes in 878, he reports how the pope crowned Louis as king (Charles the Bald had died the year before), yet refused his explicit request to do the same for his wife, still not mentioning her name.16 John's own position on the matter is less clear than it appears at first sight. While he refused to crown Adelaide, he nevertheless stayed with Louis and his wife for several weeks. His refusal, therefore, seems not to have been taken badly, and appears more to have been the result of an unwillingness to take a clear position in the issue.

Louis himself, as the incident with John VIII shows, did his best to establish Adelaide as queen. Yet, when considering his own succession, he did not disinherit


<sup>8</sup> The Girart de Roussillon of the famous Chanson de Geste.

his sons by Ansgarde. In the treaty of Fouron, made with Louis the Younger, king of Eastern Francia, and negotiated with the consent of the magnates, he considered them equal to any male children he might have in the future.17 His sons' personal importance to him can also be deduced from a letter to Archbishop Hincmar in which the king asked for the cleric's aid for them.18 Furthermore, their political relevance is underlined by the fact that the king used Carloman to strengthen his ties to the powerful Boso by means of a marriage alliance. However, on his deathbed, Louis suddenly changed his plans for the succession and appointed his oldest son Louis III as his sole heir.19 This gave rise to a complicated situation, and thereby provided ample opportunity for those looking to gain an advantage to impose their own view of things. The legitimacy of Louis and Carloman was questionable to those who considered Louis the Stammerer's marriage with Adelaide valid. On the other hand, if this second marriage were uncanonical, the same doubts would be cast on any male offspring from that union.20

Indeed the legitimacy of the boys was soon openly questioned. Not even a year later, in October 879, Boso of Vienne proclaimed himself king,21 arguing, as Regino reports, that the throne was vacant due to the absence of legitimate heirs to King Louis,22 thus claiming that neither of Louis' marriages had been valid. However, an inspection of the circumstances of this new claim reveals the purely opportunistic character of this reasoning. With Louis dead, yet without any drop of Carolingian blood himself, Boso needed an argument explaining why he himself could now claim the throne. He constructed a claim by simply casting away that of Louis' sons, to one of whom, Carloman, he had engaged his daughter only a year earlier at Troyes.23

17 MGH Capit. II, N° 246 c. 3, 169: *Si autem vos mihi superstes fueritis, filios meos Hludowicum et Karlomannum et alios, quos divina pietas mihi donare voluerit, ut regnum paternum quiete tenere possint, similiter et consilio et auxilio, prout melius potueritis, ut adiuvetis rogo.*


<sup>18</sup> Flodoard, HRE III, c. 19, 261.

#### **I.1.2 The question of Louis' succession**

As we can see, due to the questionable status of Louis the Stammerer's marriages, Charles' legitimacy certainly offered possibilities to attack his claim. Nevertheless in 893, and in contrast to Boso fourteen years earlier, Odo's supporters chose not to exploit this issue but to question the paternity of Louis. So did the questionable legitimacy of Louis the Stammerer's sons carry any political relevance at all? It has been argued that, due to the questions over the validity of both of Louis the Stammerer's marriages, the decision to crown the sons from his marriage with Ansgarde invalidated any rights Adelaide's offspring might have had, thus explaining why Charles was passed over in the royal succession at least after the death of his brothers.24 An analysis of the royal successions between 879 and 898 will help to cast some light on this question.

It is possible Charles was even passed over in 879, before his birth. After Louis the Stammerer's death in April, the leading nobles around Hugh the Abbot did not follow his last wishes and crown his eldest son Louis, but called in an assembly at Meaux to discuss future proceedings.25 Yet before Hugh and his allies were able to put any plan into action, a rival group around the former archchancellor Gauzlin had already assembled bishops, abbots and other important noblemen at Creil, inviting Louis the Younger, the partner of the treaty of Fouron, to come to the kingdom.26 According to Hincmar, Gauzlin acted out of a thirst for revenge and ambition; yet his real motive appears to have been his rivalry with Hugh the Abbot. 27 While the latter has been made out as the driving force behind Louis the Stammerer's decision that only his eldest son should succeed him, Gauzlin appears to have favoured a partition of the realm between both Louis and Carloman. 28

Faced with Louis the Younger invading the realm, in May, Hugh's faction sent envoys to the king at Verdun. In the subsequent treaty, Louis the Younger received Lotharingia in return for his withdrawal from the realm and a guarantee that the rest of the Western kingdom would remain with Louis the Stammerer's sons. 29 This agreement indicates that Hugh's and Gauzlin's factions had also come to terms by this point. However, whether this meant a return to the original arrangement of Fouron, securing the succession of all of Louis' sons and thus including the possibility of rule for another prince from the pregnant Adelaide, or whether


<sup>24</sup> Hlawitschka, Lotharingien, 235−240, argues strongly in favour of this reading. The argument has been revived by Schneidmüller, Karl III., 25, although he does not add new points to the discussion.

the succession was now restricted to the older sons of Ansgarde is not clear from the sources. In any case, Hugh and his allies failed to put the terms of the agreement into practice and Gauzlin again summoned Louis the Younger. The latter, occupied with the succession to his brother Carloman of Bavaria, promised only to return to the Western kingdom as soon as he could.30 At this point, and just before Adelaide gave birth to Charles, Hugh and his allies sent for Archbishop Ansegisus of Sens to anoint and crown Louis and Carloman. 31

As Thilo Offergeld has rightly noted, Hugh's faction could have delayed the coronation for a few more weeks to wait for Adelaide's delivery.32 The threat posed by a potential second intervention on the part of Louis the Younger was certainly crucial in provoking Hugh to act, yet Offergeld is right to wonder whether Adelaide's imminent delivery might not also have contributed to Hugh's decision.33 Unusually, Charles' birth was noted by the sources, a fact that points towards its political importance.34 Moreover, the boy certainly had at least some support: his closest relatives did try to further his rights. As Regino notes, he was named by his mother after his grandfather, Charles the Bald. 35 During the 9th century, a strong tendency had developed to give possible heirs to the throne a name from a very exclusive corpus: Pippin, Louis and Lothar, Carloman and Charles.36 The choice to name the boy Charles therefore was not random, but followed a political program, emphasising the legitimate birth of the child and his right to the throne, further strengthened through linking him to his ancestors Charlemagne and Charles the Bald. After Adelaide, one should also consider her brother Wulfard, abbot of Flavigny. Wulfard had risen under Louis the Stammerer, becoming first notary− chancellor, then archchancellor, succeeding Gauzlin in early April 879.37 Yet, for almost two years following the death of Louis, the office of archchancellor seems to have remained vacant, and the first diploma of Carloman II was witnessed only by the notary Norbert. 38 Only at the beginning of 881 does Wulfard appear again as archchancellor,39 at the moment when Carloman was staying near his abbey in the


Nivernais between the first and second campaign against Duke Boso. 40 It therefore seems likely that the party around Adelaide, amongst them Wulfard, lost its position at court with the death of Louis and subsequent assumption of power on the part of Hugh the Abbot. Wulfard apparently only regained his office after the revolt of Boso, when Hugh and Carloman needed the Nivernais as a base for their campaign against him. He may have become part of the inner circle around the king very quickly, as two intercessions in Carloman's diplomas show.41 If Wulfard indeed had lost his office due to his support for his nephew, it seems safe to deduce that a part of the price for reassuming his old position at the court would have been to renounce Charles' claim.

Certainly, at the moment of Louis' and Carloman's coronation, Charles' claim was a mere possibility. Yet, given the political tensions at that moment, it does not seem unreasonable to assume that Hugh wanted to avoid the possibility of any further complications by creating a *fait accompli*. The coronation of Ansgarde's sons, however, did not solve the problem of Gauzlin's faction. Since both boys remained with Hugh, the spirit of the agreement with the abbot and his allies—as well as Louis the Younger—had been violated, for the realm had not been divided between the brothers. Louis again invaded the realm and joined forces with those of Gauzlin, forcing Hugh to confirm the treaty of Fouron. The partition of the realm followed only a couple of months later, in March 880, at Ribemont. Louis, under the tutelage of Gauzlin and his allies, would rule over Francia and Neustria; Carloman, guided by Hugh and his party, Burgundy and Aquitaine.42 Charles, no longer a hypothetical figure but an actual claimant, was not mentioned.

Did the questionable legitimacy of Louis the Stammerer's sons play any role in these events? As already noted, Boso was the only one to deny their claim to the throne, at the moment he himself reached out for the crown. His was a general attack, not differentiating between the sons of Ansgarde and Adelaide—indeed, how could he if his aim was to create a situation in which he would not appear as a usurper denying a legitimate heir his right to the throne? The behaviour of the other leading nobles is more interesting. Did they cast aside Charles because they considered him illegitimate in comparison to his brothers?43 Louis the Stammerer had done his best to strengthen Adelaide's position at court while at the same time also promoting his succession by his sons from Ansgarde. 44 The most important document in this matter is the treaty of Fouron, securing the succession of sons from both marriages. As the leading nobles of the realm had been included in this

44 See above.

<sup>40</sup> Bautier, Recueil Louis II, Louis III et Carloman II, LXVIII.

<sup>41</sup> DCmII 61, 1st December 880−before 6th September 881, (together with Hugh the Abbot and Count Wido) and DCmII 59, 29th August 881.

<sup>42</sup> Annales Bertiniani 880, 241−242; Annales Vedastini 880, 47.

<sup>43</sup> Hlawitschka, Lotharingien, 235. According to him, once the nobles had opted for Louis and Carloman and therefore acknowledged Louis the Stammerer's marriage with Ansgarde as valid, they would in turn have automatically considered Charles as illegitimate.

decision, the treaty seems to reflect the official position of the court.45 Of course, this might have changed with Louis the Stammerer's death. Eduard Hlawitschka has argued that in fact Gauzlin and his allies leant towards Louis the Younger because of their doubts concerning the legitimacy of Louis and Carloman, thus denying their right to the throne. Hugh's faction, on the other hand, focused on the preservation of the independence of the Western realm.46 Yet, as Karl Ferdinand Werner has argued, Gauzlin was most probably one of the driving forces behind the treaty of Fouron and had supported Ansgarde earlier on,47 with no source mentioning such doubts being the motive for his turn to Louis the Younger.

It indeed seems that, for the most part, pragmatic reasons determined the courses of action of the leading factions around Hugh and Gauzlin. Hugh had opted for a single king, or, if necessary, for two, with both of them being under his control. Gauzlin urged for a real partition of the realm that would leave one of the boys in his care. Had Charles been born earlier, before or just after his father's death, he would undoubtedly have played a role in the political rivalry. Yet then, in September 879, none of the factions was in need of another claimant, especially since, given the high mortality rates among infants, supporting Adelaide's new baby might soon have left them without a candidate.48 The final solution to the conflict, the treaty of Ribemont and the partition of Amiens, agreed in March 880, was the natural consequence of these early decisions: Given the political situation with two rival factions, there was also no need for a third candidate. It was therefore a mere question of political expediency as to whose claims were to be recognised or be disposed of.49 Charles, in any case only a baby, simply did not fit into the plans of the policymakers and thus was passed over in the succession for the first time.50

#### **I.1.3 Passing over Charles**

Over the course of the next ten years, three more occasions arose in which Charles could have become king and yet was passed over in the succession. On 5th August 882, his older half-brother Louis III died, yet the Frankish nobles turned towards Ansgarde's younger son, Carloman II. 51 Two years later, after the latter had also passed away, their choice fell on Emperor Charles III the Fat, son of Louis the


German and by then ruler of the East Frankish kingdom, Lotharingia and Italy. Another four years later, after the death of the emperor, Charles was the last surviving male Carolingian, yet Odo, a powerful noble from the Robertian family, seized power. The transition to the Robertion family did not go unchallenged. A group of nobles from Francia first invited Wido of Spoleto to take the crown, but then turned to Arnulf of Carinthia before the former had even crossed the Alps. In none of these cases is Charles mentioned as a possible candidate by the sources.

Why did the nobles chose to ignore Charles and to opt for other candidates instead? Archbishop Fulk in 893 argued in a letter to Arnulf that he had not supported the young Carolingian in 888 because of his youthful age, which made him unfit to rule, especially in the face of the threat posed by the Northmen.52 This argument was persued in older research53—yet as the examples of Louis the Blind in Provence and Louis the Child in the Eastern kingdom show, boys of Charles' age could be crowned kings if the nobility agreed on them.54 Eduard Hlawitschka, on the other hand, argued that Adelaide's son Charles lacked legitimacy once the nobles had declared his brothers as kings. According to him, only one of Louis the Stammerer's marriages could be valid, and the nobility had decided that it would be the one to Ansgarde, thereby making Charles an illegitimate child.55 However, as discussed, no sources indicate that the nobility had ever seriously questioned Charles' legitimacy. It might have been used as an argument against him, yet was certainly not the main issue which prevented him from being chosen as king.56 Rivalries between the Rorgonid group close to the emperor and Adelaide's family have been made out as the reason behind the call to Charles the Fat in 884,57 but again, no sources indicate that this was the case. Concerning this succession, it has recently also been claimed that the magnates preferred a weak and mostly absent ruler, allowing them to follow their own interests.58 Yet, as Simon MacLean's study of the emperor has shown, Charles the Fat's policies towards the nobility in no way differed from those of his predecessors59 and this argument can be dismissed accordingly.

The absence of Charles the Simple from the sources concerning the question of the subsequent successions is striking and appears to indicate that he simply was


not seriously considered as a potential ruler. In 882, as well as in 884, the realm had been under the immediate threat of Viking attacks, and the sources reflect a consistent unity amongst the nobility when they decided to ask first for Carloman and two years later for Charles the Fat. Louis III had won a battle against the Northmen at Saucourt in 881 that had forced that group of Vikings into Lotharingia60 and had concluded a treaty with Hasting and the Northmen along the Loire in 882.61 Just before his death, he appears to have prepared a campaign against the group now devastating the neighbouring *regnum*; and the nobles now invited Carloman to take over the realm and the army as fast as he could.62 Their haste demonstrates that, despite Louis' last measures, the realm was far from secure: Hasting and his group returned to haunt the coast63 while the other group also returned and took position in the Condé from where they then devastated the countryside.64 Considering the previous rivalry between the groupings around Hugh and Gauzlin, the decision to unite the realm under Carloman after his brother's death may seem surprising, all the more so as Gauzlin appears to have played an important part in it.65 Hugh's influence at court, despite his rival's return, remained untouched, at least for the time being. Gauzlin stood back for a moment, yet soon resumed his office as archchancellor and became bishop of Paris.66 Facing the Viking threat, the two factions had seemingly found an agreement acceptable to both sides. Carloman would rule the entire kingdom, led by both groups in concert.

The situation two years later was not much different. Carloman, following fruitless efforts to defeat the Northmen, had finally concluded a treaty with them, paying a huge tribute for them to leave the realm. Yet, according to Regino of Prüm, when he suddenly died, the Vikings returned, claiming that the agreement had lost its validity with the king's demise. Faced with the option of either paying them again or fighting them, the magnates chose to send for the emperor, Charles III the Fat. 67 Once again, the sources emphasise the unity with which the decision to invite the emperor to the crown was taken.68 As before, we find that


Hugh and Gauzlin played key parts in the process, with other important magnates from the Paris region and Neustria acting alongside them.69 The former factions now being at peace with each other, only one candidate was needed, and Emperor Charles appeared as a logical choice considering the Viking threat.

Charles the Fat had already been in contact with the West Frankish magnates in the summer of 879, when Archbishop Hincmar of Reims had asked him to adopt one of Louis the Stammerer's sons and to take care of the boys and the affairs of the realm, a matter in which Hugh was also involved.70 Hincmar's letter was followed by a meeting in October between Louis, his brother Carloman and Charles at Orbe. However, the question of whether one of the boys really was adopted by Charles remains open.71 The archbishop sent another letter to Charles, probably at the end of 879 or the beginning of 880,72 making the same request and asking him to provide tutors for the adolescents.73 Whether Hincmar's successor, Fulk, pursued these plans any further is not known, but he was also in contact with the emperor as early as June 884.74 Charles thus had connections with the highest ranks of the West Frankish nobility long before Carloman's untimely death.

The emperor became quickly and widely accepted in the Western realm. As the diplomas he issued during his two visits to the West show, he possessed a wide-ranging network of support.75 This unity he inspired amongst the leading magnates of the realm, especially in view of the past rivalries between the different factions, was without a doubt the biggest point in his favour. If Charles the Simple, still a mere five years old, had been made king, a divided nobility fighting for control over the child and therefore a regency might reasonably have been expected. Charles the Fat, on the other hand, had proven his capability to integrate

*capiunt consilium et Theodericum comitem Italiae dirigunt ad imperatorem Karolum, uti veniat in Franciam.* In contrast to Regino, the annals do not make a direct connection between the Viking threat and the decision for Charles the Fat.


the other parts of the Carolingian empire under one ruler and it could be expected that he would also succeed in doing so in West Francia.76 The equilibrium among the magnates was new and potentially fragile and, in the face of the Viking threat, no-one seems to have been ready to risk a new civil conflict. Finally, this unity also points to another issue: the emperor seems to have had a claim to the West Frankish throne strong enough to go uncontested. His succession to the throne appears to have come so naturally that Archbishop Fulk of Reims, when explaining to the emperor Arnulf in 893 the reasons why he had not previously supported Charles the Simple's claim to the throne, referred only to the situation after the emperor's death, not to that after Carloman's.77

#### **I.1.4 A Robertian on the throne**

Three times had Charles been passed over in the succession so far and three times political reasons appear to have been predominant. In 879/880, the factions had already settled on their positions before Charles had been born and therefore had no need for him. In 882 and 884, the Viking threat created an atmosphere that enforced political unity amongst the leading nobles. The sudden deaths of Louis III and Carloman II were thus followed by quick decisions concerning the succession question, and in each case the candidate behind whom the nobles could rally the most easily appears to have been chosen. This in turn implies that Charles lacked sufficient support. By way of explanation, his youth and the problem of the legitimacy of his birth have been brought forward. Both may have played a role, yet other examples demonstrate that they were certainly not a hindrance.

This was even more the case in 888 after the death of Charles the Fat. As Regino of Prüm famously noted: "After his death the kingdoms which had obeyed his authority, just as though a legitimate heir were lacking, dissolved into separate parts and, no longer waiting for a natural lord, each decided to create a king from its own guts."78 The death of the emperor had created a vacuum which was now used by a number of powerful nobles to seize the crown: Arnulf of Carinthia, illegitimate son of Carloman of Bavaria, in the East Frankish realm; Berengar, son of Eberhard of Friuli in Italy; and Odo, son of Robert the Strong in the West Frankish kingdom, to name but three. Over the preceding years a "crisis of authority"79 had

<sup>76</sup> MacLean, Kingship, 120−122. See also the positive accounts of Charles the Fat by Regino, Chronicon 888, 128−129, and Abbo, Bella I, 18, v. 48−52 and II, 98, v. 442−443. On Abbo's view of Charles see MacLean, Kingship, 55−62.

<sup>77</sup> Flodoard, HRE IV, c. 5 380−383: *De hoc etiam, quod idem rex in culpa trahebat, quare non id ante fecissent, reddit rationem, quod, quando Karolus imperator decessit, et idem Arnulfus regimen huius regni suscipere noluit…*

<sup>78</sup> Regino, Chronicon 888, 129: *Post cuius mortem regna, que eius ditioni paruerant, veluti legitimo destituta herede, in partes a sua compage resolvuntur et iam non naturalem dominum prestolantur, sed unumquodque de suis visceribus regem sibi creari disponit.*

<sup>79</sup> Airlie, Nearly Men, 26.

developed, a legitimacy deficit of the Carolingian dynasty resulting from the continuous threat generated by Vikings, Hungarians and Saracens as well as the quick series of successions since the deaths of Charles the Bald and Louis the German. When Boso had initiated his own coronation in 879, four Carolingian rulers had allied themselves to crush this attempt.80 Their alliance demonstrates how serious they considered the threat to be.81 Boso's case had unlocked the door for all those who wanted to challenge the Carolingian monopoly on kingship, and the death of Charles the Fat, which left his realms without an heir,82 pushed it wide open.

One of the new kings, and in our case the most interesting one since he claimed the West Frankish throne for himself, was Odo, son of Robert the Strong, 83 who had risen from count of Paris to *marchio* of Neustria under Charles the Fat. Odo had proven himself a more than capable military leader during the siege of Paris in 886 and, due to his family ties and his proximity to Charles, he had acquired the inheritances not only of Gauzlin, 84 but also of Hugh the Abbot after their deaths. As such, Odo united the military command over Neustria with the counties Tours, Angers and Blois, as well as the abbeys of Saint-Martin of Tours and Marmoutiers, with the county of Paris and the abbey of Saint-Denis. 85 In his person, therefore, the Western kingdom now had a magnate powerful enough to dominate the other nobles and be able to legitimise himself through military victory.


However, this same situation also offered new opportunities for Charles the Simple. If the problem of his mother's marriage to Louis the Stammerer had thus far been an issue for the nobility, now, in the face of the lack of other Carolingians, his claim by blood was stronger than that of any others, and his position thus changed dramatically. It is therefore hardly coincidental that Charles reappears in the sources at this very moment. Once Odo had consolidated his rule in the North and overcome the initial opposition, he moved to Aquitaine, where he met with Ramnulf II of Poitiers.86 The latter also brought Charles to the meeting and swore an oath not only for himself, but also for the Carolingian, "lest anything evil be suspected of him."87 Ramnulf himself might have attempted to take the crown for himself, although this is only reported by the *Annales Fuldenses*. 88 It is clear, however, that Odo considered Charles a threat, and this threat could only stem from Charles' Carolingian blood.

This is also emphasised by other sources. "And when at the day mentioned above they gathered at Reims, they placed him on his father's throne, consecrating him king"89 the *Annales Vedastini* report for the year 893 when the rebellious faction around Archbishop Fulk finally crowned Charles. This reference to his Carolingian descent undoubtedly reflects the official standpoint of the group: it is the same argument used by Archbishop Fulk in his letter to Arnulf from the same year, in which he named Charles the last descendant of the royal family, whose predecessors and brothers had been kings.90 Fulk's argument however should not be taken out of context. The archbishop was defending himself against the accusation of unjust rebellion. Charles' claim served to justify his own actions while he presented Odo as "a stranger to the royal family [who] had tyrannically abused the


royal power."91 The argument of Charles' legitimacy was therefore not only used to emphasise his claim, but also to deny Odo's right to rule.

This last point mattered more to Fulk than enforcing the rights of the Carolingian youth. The opportunistic nature of his argument becomes clear from the archbishop's behaviour in 888. After the death of the emperor, a group based in Francia and Burgundy, centred on Archbishop Fulk, 92 invited Fulk's relative, Wido of Spoleto93 to the realm, only to abandon him even before he arrived.94 Wido was still crowned as the choice of a number of Burgundian nobles by Bishop Geilo of Langres, but returned to Italy on the news of Odo's coronation.95 Odo, for his part, seems to have been able to bring the resisting nobles from Francia over to his side using promises and threats;96 however, he failed to quickly integrate them under his rule. Soon after, Fulk, together with Count Baldwin II of Flanders and the latter's cousin, Abbot Rodulf97, turned towards Arnulf, king of the East Frankish kingdom, and invited him to the Western realm.98 Negotiations were still in progress when Odo won a decisive victory against the Vikings at Montfauçon.99 On hearing this, Arnulf refused Fulk's offer "without council or conciliation."100 Instead he opted for the victor of Montfauçon and not only invited him to a meeting, but also sent him a crown soon after. Fulk's party immediately disintegrated,


leaving him no other option but to submit to Odo for a second time.101 The "primacy of the royal blood" the archbishop used as an argument five years later102 thus appears not to have played a role in the succession crisis of 888; Fulk first opted for his own relative, Wido of Spoleto, and then turned to Arnulf of Carinthia.

Fulk, of course, was aware of this hole in his argument when he wrote his letter. In his defence he claimed that in 888 Charles had been too young in body and wisdom to govern the realm and that the Viking threat made it too dangerous to choose him as king.103 Yet Boso's son Louis had succeeded his father in 887 at an even younger age and so too would Arnulf 's own son Louis only a few years later. While the Viking threat might be more credible, the question arises whether, given his young age in 888 as well as in 893, it would have been Charles who commanded the army and not his advisers around Fulk. His half-brother Carloman, for example, remained dependent on the groups around Hugh the Abbot and Gauzlin even after he had been given the arms of manhood104 and in the years following his own coronation the sources emphasise the role of those "who were with Charles."105 The archbishop's arguments are thus revealed as mere afterthoughts, excuses to Arnulf for his own behaviour five years earlier.

This, however, does not answer the question why Fulk and his allies turned to Wido of Spoleto and Arnulf instead of elevating Charles. The cause may have been partly practical: Charles simply might not have been available, since he was with Ramnulf II of Poitiers.106 Yet other reasons probably carried more weight. In contrast to 893, Fulk's party seems to have been rather weak in 888. Apart from the suffragan bishops of Reims, Fulk was supported by only Baldwin II of Flanders and Baldwin's cousin Abbot Rodulf. 107 He still lacked the support a man as powerful as Count Heribert I could provide,108 as well as connections outside Francia.109 Fulk's party was therefore limited to candidates who could not only serve as rallying points for other discontented magnates, but who also brought with them their


own connections, like Wido of Spoleto, or had a power base sufficient to fight Odo, like Arnulf. Since Charles could provide neither of them, he was disqualified as a candidate.

#### **I.1.5 Carolingian blood matters**

We can now return to our initial question: why did Odo's supporters in 893 deny Louis the Stammerer's paternity instead of attacking Charles where he was more vulnerable, the problem of the legitimacy of his birth, as Boso had done in 879? Four times Charles had been passed over in the royal succession until he finally became king. In the first three cases, the political circumstances—the rivalry between the factions of Hugh the Abbot and Gauzlin and the Viking threat—could be made out as the main reasons. Charles fell victim to his late birth when the political decisions concerning the succession of his father had already been taken. Once marginalised, he simply did not possess enough political support to have his claim enforced. Whether this was due to the problem of the validity of Adelaide's marriage to Louis the Stammerer has to remain an open question, although the political realities appear to have carried enough weight on their own for him to be cast aside time after time.

Be that as it may, the death of Charles the Fat in 888 changed the field dramatically. With Odo, a man now reached for the crown who drew his legitimation not from his blood, but from his victories against the Northmen.110 Does this mean that by the end of the 9th century Carolingian legitimacy had lost its importance? The emperor's death had created a vacuum due to a lack of heirs and there being no agreed-upon succession, while Boso's example from 879 had already demonstrated that the Carolingian monopoly on kingship was not set in stone. Odo, making use of his prestige, won by defending the realm against the Northmen, now went down the same path and attempted to fill the vacuum. This new legitimation by victory allowed him to declare his claim while his political connections and his resources enabled him to enforce it. Yet, the immediate challenges also reveal the limits of his support and the fragility of his position. These challenges also show how weak the power of Carolingian legitimacy had become at that point, for Fulk and his allies decided against supporting Charles—who was now passed over for a fourth time—and instead opted first for Wido and then for Arnulf.

It would be wrong, however, to underestimate the force of Carolingian traditions. They remained an essential part of royal behaviour and representation: the carefully phrased language of Odo's diplomas and *promissio* sets him in line with his Carolingian predecessors and attempts to negate the different nature of his

<sup>110</sup> On the importance of the prestige won from warfare for Carolingian kingship, see Scharff, Gott, 265−276.

kingship, thus trying to create the impression of uninterrupted continuity.111 He had himself crowned at Compiègne,112 the palace that had become so important under Charles the Bald that it was called Carlopolis by later writers.113 His *promissio*114 was based on those of Louis the Stammerer and Carloman II, 115 assuring the audience that in fact nothing had changed. Immediately after his coronation he took possession of the Carolingian insignia stored at Saint-Denis. 116 In short, Odo did his best to appear no different from his predecessors. He used the force still inherent in the Carolingian tradition to strengthen his rule. Having no Carolingian blood to legitimise himself, he at least wanted to appear as a Carolingian since Carolingian traditions offered a legitimation more durable and reliable than that based on victory, which depended on the fortunes and misfortunes of the battlefield. How strong the Carolingian argument still was can also be deduced from Fulk's letter to Arnulf. 117 The archbishop certainly did not crown Charles out of deeply held principles of Carolingian legitimism—we will return to his motives later—yet his use of the Carolingian card in opposition to Odo's "strangeness" shows how much weight Charles' blood still carried in political discourse.118

This was Charles the Simple's opportunity. Having been passed over in the royal succession four times and lacking political support he could not have expected to leave his political backwater soon. Now, however, Odo's lack of Carolingian blood and the absence of other male Carolingians—apart from the also questionable Arnulf and Louis of Provence, Boso's son with the Carolingian princess Ermengard—improved his position considerably. Because of his claim, Charles was the ideal candidate, not for a normal succession, but for any opposition challenging the Robertian's rule. His blood made him an ideal rallying point for all those discontented with the Robertian. This was why Odo's supporters did not use the issue of the legitimacy of Charles' birth against him but brought forward doubts about Louis the Stammerer's paternity. In doing so, they attacked Charles where he had his advantage over Odo: his Carolingian blood that, even considering his possibly questionable legitimacy, made him superior to his opponent. Because Carolingian traditions remained strong, simply questioning Charles' legitimacy did not


<sup>111</sup> Schneidmüller, Tradition, 109. On the continuities in form see Mersiowsky, Urkunde, 214−223. See also Airlie, Élites, on the takeover of Carolingian practices by important nobles.

suffice. Charles' Carolingian blood had to be denied entirely, since it was the pillar on which his claim rested.

#### **I.2 Close supporters and distant allies**

Even relatively recently, the members of the alliance elevating Charles to the throne have been described as lacking constancy and principles, driven mainly by self-interest. 119 Yet the war which ensued between the Carolingian's supporters and Odo lasted more than four years and, in the end, Charles not only kept his crown, however small his realm may have been, but it was also agreed that he would succeed the Robertian upon the latter's death. If this network was indeed so fragile, why had the conflict lasted so long and ended with such a positive result for Charles? It would appear that the motives of the alliance's members in opposing Odo and supporting Charles were strong enough to create a certain level of group cohesion, a cohesion that proved sufficient to withstand even great pressure from the Robertian. However, scholars rarely produce their arguments out of thin air. What, then, are the reasons for the image of the alliance as unprincipled and inconstant?

#### **I.2.1 Archbishop Fulk of Reims**

The best starting point for an analysis of this network is one of its principal agents, the oft-mentioned Archbishop Fulk of Reims, who we saw acting against Odo in 888. Fulk first appears in the sources accompanying Charles the Bald to Rome in 875. Two years later he became abbot of Saint-Bertin and was among those to whom the emperor entrusted his realm and his son Louis in the capitulary of Quierzy.120 He seems to have spent the following years at the royal court121, yet his actual influence is hard to determine since he neither figures in royal diplomas nor is mentioned by the narrative sources. After Hincmar's death he became archbishop of Reims, which seems to indicate closeness with the dominating factions around Hugh the Abbot and Gauzlin. 122 Despite his new position, his influence at court remains hard to trace. Soon after his ordination, he wrote a letter recommending King Carloman to Pope Marinus I. 123 In summer 884, he visited Charles the Fat

<sup>119</sup> Offergeld, Reges pueri, 425: "Die folgenden Jahre sollten zeigen, daß die Großen sich ihre Unterstützung von Fall zu Fall neu abringen ließen; eine verläßliche oder gar gesinnungsfeste Gefolgschaft hat Karl nie besessen." Similar Schneider, Erzbischof, 112.

<sup>120</sup> Schneider, Erzbischof, 22−23.

<sup>121</sup> Flodoard, HRE IV, c. 4, 379.

<sup>122</sup> Schneider, Erzbischof, 25−27.

<sup>123</sup> Flodoard, HRE IV, c. 1, 363−364.

in Metz to ask for the restoration of possessions in Thuringia,124 which indicates that he had connections outside the realm, but again does not shed light on his actual influence in the politics of the West Frankish court. Once the emperor had become ruler in the West as well, Fulk sent him a letter asking for his help in the defence of Paris against the Northmen,125 yet, as before, this letter says more of Fulk's ambition to gain a preeminent position than his actual influence.126 Instead, Charles' reign saw the rise of Odo, who favoured the church of his relative, Archbishop Walter of Sens. Fulk's opposition to Odo might thus be seen in the context of the struggle between the sees of Reims and Sens for the primacy of the Gallic church.127 Yet this solution seems too narrow, as it fails to consider the archbishop's political ambitions, demonstrated by his invitation to Wido of Spoleto to take the crown after the death of Charles the Fat before turning to Arnulf of Carinthia. Fulk, as archbishop of Reims, seems to have claimed a central position in West Frankish politics that had been denied to him during the reigns of Carloman II and Charles the Fat. Odo's power, on the other hand, was firmly rooted in his Neustrian possessions and *honores*, while the position at his side was occupied for now by Count Theoderic of Vermandois. 128 Under the rule of the Robertian, Fulk could not hope to gain the influence on royal affairs which Odo himself had wielded under Charles the Fat. It was a lack of proximity to the new king which frustrated his political ambitions and provoked his resistance.

Aware of the archbishop's ambitions, Odo followed a policy of appeasement and integration, trying to deal with the root cause of the resistance initially shown by Fulk and the nobles around him. Count Baldwin II had already submitted to Odo before the Robertian met with Arnulf. 129 After the meeting, in a well-enacted ceremony, Odo was crowned for a second time at Reims,130 probably by Fulk himself,131 who also received a diploma for his church.132 Odo then moved on to spend Christmas at Saint-Vaast, one of the abbeys of Abbot Rodulf, one of Fulk's allies when he had turned to Arnulf. In 890, the abbot also received a diploma through the intercession of Odo's wife, Theodrada. 133 Another diploma went to


Bishop Dido of Laon,134 a suffragan of the church of Reims. Dido, as well as Honoratus of Beauvais and Riculf of Soissons, accompanied Fulk to a large assembly at Verberie in 892.135 Whether Count Heribert I, who was to become one of Charles' most important allies,136 had supported Odo or had joined the ranks of Fulk's allies cannot be answered, but just after Christmas 889, he appeared in one of the kings diplomas as "one of our great men."137 Odo's reign certainly does mark Heribert's ascent: from 888 onwards he became count of Meaux, the Vexin, the Mézerais and Soissons, while his brother Pippin acquired the county of Beauvais.138 Odo's network in the Francia is dealt with in the next chapter, so for the moment we shall simply note that, among those the Robertian tried to integrate into his rule, we can find at least two—Fulk and Heribert—who were to become the leading members of the revolt against him.

For the next few years however, nothing indicates that the two were fundamentally opposed to Odo. Heribert, who was close to the king at the end of 889, unfortunately disappears from the sources. Fulk, on the other hand, remains much more visible. After his final submission to the king, he mostly appears to have cooperated with Odo. The *Visio Karoli,*139 probably written at Reims, endorses the right of Louis of Provence to the imperial crown and may be a sign that Fulk tried to promote Louis as a candidate for the West Frankish throne, which would indicate his continuing ambition to claim a central role in the realm's politics. With that said, the question is still not resolved satisfactorily.140 More important for the political situation in the realm was the struggle for the episcopal see of Langres. After the death of Bishop Isaac, the clergy and people of Langres had chosen the deacon Teutbald and asked Pope Stephen V to consecrate him, against the will of the metropolitan Aurelian of Lyon, who should have been the one to consecrate the new bishop.141 Stephen refused the demand, as he did not want to interfere with the archbishop's rights, and instead advised Aurelian to consecrate Teutbald, unless he happened to have any personal objections to him. If that were to be


the case, by papal order no other bishop might be appointed in the meantime. Furthermore, Stephen sent Bishop Oirannus of Senigallia as his legate to mediate between the two parties.142 On Oirannus' arrival at Lyon, Aurelian sent him on to Langres with the promise to proceed there himself as soon as possible, which he then failed to do. When his refusal to follow the pope's commands became apparent, the clergy and people of Langres sent another letter to Rome, once more asking Stephen to consecrate Teutbald. The pope, still unwilling to interfere with Aurelian's rights, now sent a letter to the archbishop to enquire about his reasons for refusing Teutbald. 143 Once again Aurelian did not respond; instead, he, together with his suffragans and Archbishop Bernuin of Vienne, now consecrated the cleric Argrim as bishop.144 Only then did the pope send instructions—this time to Fulk—to consecrate Teutbald, 145 in turn prompting Odo to send an embassy to Rome. Fulk, for his part, informed Stephen that he would not act before the return of the embassy.146 Early in 891, Teutbald was finally consecrated and installed as bishop of Langres. Eduard Hlawitschka has interpreted Fulk's role in the affair as that of a leading actor who tried to install his protégé, Teutbald, in Langres, acting against Odo's candidate, Argrim, and finally successfully depriving the king of an important foothold in northern Burgundy.147 However, as Rudolf Pokorny has shown, neither Fulk's purported backing of Teutbald nor Odo's for Argrim can be supported by evidence.148 Whether and how the archbishop of Reims finally acted is indeed unclear; from his only appearance, it seems that he did indeed cooperate with the king.

The picture is clearer in a number of other cases in which Fulk seems to have supported Odo's policies. In 892, the king participated in a synod concerning a dispute between the abbess Hildegard and a certain Hermingard. 149 Likewise, the archbishop and his suffragans Dido of Laon, Honoratus of Beauvais and Riculf of Soissons were present at a royal assembly at Verberie, where they sat in judgement over the case of the monk Aginus with the archbishops, Walter of Sens and Wido of Rouen.150 Finally, Fulk sided with the king against Baldwin II of Flanders in the conflict over the abbeys of Saint-Vaast and Saint-Bertin: after taking counsel from four of his suffragans, Fulk ordered Bishop Dodilo of Cambrai to dissuade the count from continuing with his course of action and threatened his followers


with anathema if he did not give in.151 When Dodilo failed to obey, Fulk held a synod to deal with the matter and wrote directly to Baldwin, threatening him with excommunication.152 Until late in the year, Fulk seems to have supported Odo's rule—and even profited from it. When the monks of Saint-Bertin, having refused Baldwin as their new abbot, received Odo's permission to make their own choice, they elected Fulk as abbot.153

However, it is worthwhile to note another dispute, also related to the death of Abbot Rodulf. Rodulf had not only held the abbeys of Saint-Vaast and Saint-Bertin, but also Cysoing and the relics of Saint Calixtus. Before his death, Rodulf had placed both under the protection of the church of Reims. Yet, once Rodulf died, his sister's husband, Hucbald, one of Odo's closest allies in Francia,154 took the abbey under his control. Fulk wrote to the pope to ask him for advice on the matter, to demand a confirmation of the donation and to threaten all opponents with excommunication.155 In his answer, reported by Flodoard, Pope Formosus did indeed confirm Cysoing as the possession of the church of Reims; however, there is no mention of any threats of excommunication.156 This may indicate that Hucbald had given up his claim on the abbey by the time of Formosus' reply. However, relations between the archbishop and the count remained tense even after the affair was resolved. When Fulk instructed Dodilo of Cambrai to fetch Calixtus' relics from Arras and transfer them to Saint-Quentin, Dodilo chose to seize the relics, making Fulk fear that he would hand them over to Hucbald. 157 Both abbey and relics finally ended up in the possession of the archbishop, as Flodoard states,158 but whether this happened under Odo or under Charles remains open. It is nevertheless worth noting that Fulk, just before he decided to revolt against Odo, was in conflict with one of the king's main supporters in Francia. In general


however, Fulk cooperated with Odo in cases such as the succession of Langres and supported him against Baldwin II; while the *Visio Karoli*, despite all the problems of interpreting it, might be read as a sign of his continuing ambitions.

Yet does this mean that Fulk wholeheartedly supported Odo's rule? As he indicates in his letter, it was not until late 892 that he was approached by a group of discontented nobles for advice.159 The origins of this newest conspiracy against Odo seem indeed to have been quite recent, at least as concerns the archbishop.160 However, this passage also shows that Fulk was perceived as someone who would not betray such discontent—which would indicate that he was not part of Odo's most inner circle—and, in addition, that he was thought likely to support this new opposition. The archbishop, despite his cooperation with Odo, was apparently known for having had a negative attitude towards the king. At the same time however, the king himself seems to have continued to be oblivious to this, as he, on the advice of that very same group, decided to leave Francia and march to Aquitaine just before they openly rebelled against him.161 Fulk figured as a rallying point for the opposition and his cooperation with Odo was therefore probably motivated less by actual conviction than by necessity in the face of political realities. As the year 888 had proven, opposition against Odo needed stronger support than he could mobilise on his own. This, of course, does not mean that Fulk only bided his time, waiting for an opportunity to rebel. Cooperation could also be beneficial, as him becoming abbot of Saint-Bertin for a second time shows.162 Under different circumstances, cooperation might have given the archbishop the political influence he aspired to.

#### **I.2.2 The network elevating Charles**

This leads us to the network that was now formed to elevate and support Charles. According to Regino of Prüm, most of the Francian nobles renounced Odo to follow the new king,163 yet this statement should be read with great care. Several bishops from the archdiocese of Reims remained loyal to the Robertian and a number of counts from the area did not side with Fulk either, most prominently Odo's old ally Count Hucbald of Senlis.164 Within the group, the archbishop occupied the most prominent role. We have already seen him become a rallying


point for those discontented with Odo's reign. After the rebellion began, he did his best to further Charles' cause by using his contacts with popes and kings.165 Fulk's central position is also reflected by the *Annales Vedastini*, which repeatedly mention him as one of the leaders, being close to Charles166 and acting as legate to Arnulf. 167 After Fulk, Count Heribert occupied a key position. Having risen under Odo and now holding several counties, he was mentioned by Fulk as one of those who approached him before the rebellion.168 Regino of Prüm names him as one of the leaders of the rebellion, next to his brother Pippin and Archbishop Fulk169 and later he appears as a military leader170 and chief negotiator.171 His defection to Odo in 896 seems to have hit Charles especially hard. Relations between him and the count appear to have deteriorated so much that, a year later, Heribert was the only noble who had left Charles' side who the sources mention to be formally reconciled with the Carolingian.172

Unmentioned by the narrative sources yet at the same time prominent in both of Charles' surviving diplomas from the years of the rebellion is another key actor: his mother Adelaide. That Carolingian queens could possess political influence is well known yet hard to assess in individual cases.173 In Adelaide's case, she intervened remarkably often in Charles' diplomas—thirteen times174 in the 31 charters that have survived as originals or copies−until her death in 901.175 Only three times is she named queen (*regina*), but all of the diplomas name her as "our mother" (*genitrix nostra*), with epithets varying from "venerable" (*venerabilis*) to "dearest" (*carissima*) and "most beloved" (*dilectissima*) and even "sweetest" (*perdulcissima*).176 Her importance thus does not seem to have stemmed from


her position as queen. The special character of her appearances will be addressed again at a later point.177 For the moment, it suffices to note that all of the nobles she intervened with had connections to the court independent from her, as in the cases of Fulk, 178 his suffragan bishops Honoratus of Beauvais and Rudolf of Laon179 and the counts Ecfrid180 and Erkanger. 181 While she was certainly important, it does not automatically follow that her importance was based on her own political connections. Rather, it seems that, based on her role as his mother, she was Charles' closest advisor during these years, and that the diplomas reflect her influence on the young king.

As we have seen, Fulk and Heribert were both integrated into Odo's rule. Although not part of Odo's innermost circle, they were nevertheless in contact with the king. Another instigator of the rebellion was part of this inner circle, the man who led the group approaching the archbishop of Reims: Anskeric, the bishop of Paris, the king's archchancellor in succession to the rebellious abbot Ebolus of Saint-Denis. 182 Little is known about Anskeric. His brother was Count Tetbert of Meaux and he may have had relatives in the Langrois.183 A marriage alliance with Pippin, father of Count Heribert I of Vermandois, has also been suggested, which would explain the connection between the bishop and the count.184 In addition, a later source, the *Carmen de pontificibus romanis*, indicates that he was related to Charles the Simple.185 Politically, Anskeric seems initially to have been in competition with Gauzlin for the office of bishop of Paris, failing to secure it on the death of Angelwinus. He nonetheless appeared at the court of Charles the Fat in 885 as *episcopus vocatus.*186 Charles' support allowed him to succeed to the episcopal see upon the death of Gauzlin. 187 His proximity to the emperor and his position in Paris are both emphasised by the fact that, when the Vikings arrived at the city in 887 to demand the tribute promised to them in return for lifting the siege the year before, it was Anskeric who went to Charles,188 before negotiating their passage to

75: *dulcissima genitrix nostra Adeleidis*; DChS 39, 83: *dulcissima genitrix nostra Adeleydis*; DChS 41, 90: *dilectissima genitrice nostra Adeleide*.



<sup>177</sup> See chapter III.2.2.

<sup>178</sup> DChS 5.

<sup>186</sup> DChF 116.

Burgundy.189 Once Odo had become king, Anskeric joined his operations against the Vikings with his own troops.190 He appears shortly afterwards as abbot of Saint-Germain of Auxerre,191 intervened for Bishop Argrim of Langres before the king192 and finally became archchancellor in 892.193 Thus the bishop's political importance can hardly be overestimated, an importance he also possessed within Charles' inner circle after the rebellion had begun. He now became the Carolingian's new archchancellor, the same role he had held at Odo's court.194 However, Charles' only other diploma surviving from the years of the rebellion, dating to the summer of 896, names neither him nor any other as archchancellor,195 indicating that he had left Charles' side. When this happened, however, is not clear. It may have occurred at the very time that Heribert and Fulk defected from the Carolingian. In any case, this appears not to have had any lasting consequences for their relationship. After Fulk's death, Anskeric immediately resumed his old office.196

Taking the office as an indicator of importance, we can also add another count to the men close to the Carolingian. A certain Elduin appears as count of the palace in a private charter of Richard the Justiciar in 896.197 His presence there can be explained by the negotiations taking place between Charles and Richard, with Elduin acting as the former's legate. This would also emphasise his importance at Charles' court at that moment, furthered by the circumstance that Heribert, Fulk and others had submitted to Odo earlier during that year.

Other counts also participated in important negotiations. Treating with Odo in 895, Heribert was accompanied by Erkanger and Ecfrid. 198 Little is known about either of them. Ecfrid was one of those who first approached Fulk in late 892199 and


<sup>189</sup> Abbo, Bella II, 96, v. 411−415. Bautier, Recueil Eudes, XXVII, suggests that the Northmen were prohibited from using the Marne, as Anskeric's brother Teutbert was count of Meaux.

is also attested in the sources in connection with Baldwin II's rebellion after the death of the abbot Rodulf. At the time, the monks of Saint-Vaast first sent him to ask after the king's will, only then to choose Baldwin as abbot on the advice of a certain Everbert. 200 That Fulk chose to mention him in his letter nonetheless indicates that Ecfrid was of some political importance. Erkanger, generally identified as count of Boulogne, had served as Gauzlin's legate in 886 when he went to seek Duke Henry's help against the Northmen.201 Like Heribert, he left Charles' side in 896.202 Heribert's brother, the already mentioned Pippin, also belonged to this circle of men carrying some political weight while probably not belonging to the group of central decision makers.

Also interesting is the case of yet another count, Aledramnus. He was sent to Arnulf by Fulk before Charles' coronation, only to be excommunicated by Bishop Honoratus of Beauvais soon after.203 The choice of Aledramnus as emissary to Arnulf was by no means coincidental. Since the last decade of Charles the Bald's reign, the count had been a leading member of the Frankish nobility, appearing prominently in diplomas of Charles the Bald, Louis the Stammerer and Charles the Fat. 204 During Charles the Fat's reign, he and his brother, Count Theoderic himself a very important ally of Odo—led the advance guard during the emperor's march on Paris.205 Moreover, Aledramnus was related not only to Charles the Simple,206 but also to Odo, whose wife Theodrada was Aledramnus' daughter.207 The count thus was a political heavyweight, well connected in the highest circles. Choosing him as emissary to Arnulf sent a powerful message to the realm: Aledramnus' support tied the reign of the young Charles to the reigns of his father and grandfather and demonstrated that even Odo's closest supporters—even members of his own family—rallied around the young Carolingian.

The mention of the Lotharingian Count Reginar Longneck at Charles' side in 895 is difficult to assess. In his fight with Odo, Charles and his supporters had concluded an alliance with the Lotharingian king, Zwentibold. Having united their forces, they now besieged Laon, where, as the *Annales Vedastini* report, "Reginar Longneck, count, *marchio*, accepting bad advice, left Charles and went to Zwentibold."208 This phrasing indicates that he, although a Lotharingian noble,


had previously supported Charles. Reginar appears to have held possessions in the Western kingdom209 and had married Charles' sister Ermentrude around 888, the latter probably dying around 892, after the birth of her daughter Cunigunde. 210 Whether he had joined Charles at the beginning of the rebellion or only later cannot be established. It seems plausible that, due to his connections to both Charles and Zwentibold, he had established the link between the two, yet, lacking further evidence, this is mere speculation. Since, during the rebellion, he is not mentioned before nor after this incident, it appears best to assume that his importance within the network supporting Charles was limited to a shorter period around the conclusion of the alliance with Zwentibold.

Less problematic, but certainly of more peripheral importance within the network, were other men: Gauzfrid's sons, nephews of Gauzlin and relations of the Ramnulfids in Aquitaine,211 belonged to the initial group approaching Fulk; 212 the otherwise unknown Count Adalung was killed accompanying the archbishop to a meeting with Arnulf. 213 From the archdiocese of Reims the bishops Heitilo of Noyon, Herilandus of Thérouanne and, soon afterwards, Mancio of Chalons214 joined the ranks, along with Bishop Teutbald of Langres in northern Burgundy.215 The example of the archdiocese of Reims, however, also demonstrates the limits of this new network. A number of suffragans did not agree with the policy of their

*consilio accepto Karolum reliquerunt et se ad Zuendebolchum contulerunt.* For Baldwin and his brother see below.


metropolitan: Dido of Laon as well as Honoratus of Beauvais stayed loyal to the king216 as, possibly, did Otgar of Amiens. 217

We have so far described different layers of the network forming to rebel against Odo and elevating Charles to his father's throne. The innermost core consisted of Charles' mother Adelaide, Archbishop Fulk of Reims, Bishop Anskeric of Paris and Count Heribert of Vermandois. In the next layer were men like Heribert's brother Pippin, Count Aledramnus, the Count of the Palace Elduin, Count Erkanger, Count Ecfrid and possibly also Reginar Longneck if he indeed belonged to Charles' supporters. Undoubtedly to the outer rim belonged those primarily in contact with the men from the inner core or those whose base lay a greater distance from Francia: Count Adalung, Fulk's suffragan bishops Heitilo, Herilandus and Mancio, Bishop Teutbald and the sons of Gauzfrid. In the *Annales Vedastini* this group is most often referred to as "those who were with Charles,"218 indicating, as Thilo Offergeld has shown, their influence on the actions of the king. Whenever political decisions had to be made, the annals emphasise their role as a group and only mention Charles when they indicate simple geographical movement.219 "Those who were with Charles" are thus opposed to the *fideles* who were accompanying and counselling Odo. The Robertian made his own policies; Charles seems to have been a boy−king dominated by those around him.

#### **I.2.3 The motives behind the rebellion**

What was it that brought these nobles together?220 Adelaide, we may reasonably presume, wanted to further her son. Fulk had received at least two abbeys since


Odo had become king, Saint-Bertin and Avenay, yet, as we have argued, had not made it into the king's inner circle. In Fulk's case, it is indeed most likely that his political ambitions were the root of his opposition to Odo. A new king could certainly offer more than Odo would ever be able to, be it in *honores* or, more importantly, influence on royal decision making. From the beginning, Odo's rule had not been based on general consent, but rather on his Neustrian possessions and connections, as well as some support from the likes of Theoderic of Vermandois and Archbishop Walter of Sens, who thereby increased their influence over the king.221 Only after his victory against the Normans and his recognition by Arnulf was Odo able to impose his rule over the whole of the realm. That Odo did his best to integrate former opponents, such as Fulk, shows that he was well aware of their importance for the stability of his realm. However, they would never have been able to assume a position equal to that of the king's first supporters. Charles, on the other hand, was just a boy on the brink of manhood and so far without any significant support. Backing him would have meant that Fulk and Heribert would be able to exert that very same influence that was denied to them under Odo. 222

Yet, this focus on the assumed political ambitions falls short when it comes to explaining the motives of other members of the network. Anskeric and Count Heribert, for example, were powerful men in key positions and close to Odo, the former only recently having been promoted to archchancellor and the latter rapidly becoming one of the leading magnates in Francia. Others, like Heribert's brother Pippin, undoubtedly joined because of their personal connections to those in the inner core, while in the case of men like the counts Aledramnus and Ecfrid we simply do not know enough to make a proper judgement. There is Fulk's letter to Arnulf to take into consideration however. At two points Fulk makes remarks concerning Odo's behaviour. First, Anskeric, so Fulk claims, came to him asking for his advice concerning "insupportable" commands from the king. And second, that those who had approached him also asked him for advice about what to do concerning the "evil" Odo wanted to do to the sons of Gauzfrid. 223 These remarks may of course only have been pretexts, yet it is worthwhile pursuing this thought further.

be an overly large proportion. The image only changes if we add, as Schneider and Schneidmüller do, Baldwin II of Flanders and his brother Rodulf, Reginar Longneck and the descendants of Ramnulf II of Poitiers to the group's allies. However, we will later show that these men cannot be so added. Following Offergeld, Reges pueri, 425, we might in addition add that most of the nobles of the region were in fact related to the Carolingians, so that the (supposed) high concentration of them in the group can hardly be surprising.


During 892, the year leading up to the rebellion of Fulk and his allies, two conflicts broke out in different parts of the realm. In the Francia, on the death of Abbot Rodulf, Baldwin II of Flanders claimed his cousin's abbeys of Saint-Vaast and Saint-Bertin for himself.224 In the following months, not only did the monks of the former declare for the count, but Odo's own relative, Walker, whom he had entrusted with the castle of Laon, did so as well. Castle and city both fell soon enough and Walker was condemned to death by judgement of the nobles who were present. Odo refused to pardon him. In addition, and probably at the instigation of the king, Bishop Dido of Laon also withheld spiritual support from Walker during his last hours.225 The conflict with Baldwin continued nevertheless. Even deeper rifts in Odo's political order were caused by the succession of Ramnulf II at Poitiers.226 Ramnulf had an illegitimate son, Ebalus Manzer, who was well connected at court: his uncle, Ebolus, abbot of Saint-Denis, 227 served as Odo's archchancellor, a fact which probably helped secure Odo's recognition of his succession.228 In early 892, however, Odo decided to grant Poitiers to Count Altmar, 229 thus provoking the revolt of Abbot Ebolus and his brother Gauzbert, who took up their nephew's claim. Soon after, he revoked his earlier decision and installed his own brother Robert instead of Altmar at Poitiers, in turn leading to a revolt of the latter. By this point, the group of nobles surrounding Fulk of Reims had crowned the young Charles king, meaning that the rebellion would henceforth demand Odo's entire attention.

These different conflicts which broke out against Odo, in Francia as well as with Count Baldwin in Flanders and the Ramnulfids in Aquitaine, may have been to a great part independent of each other, but they may also have been a consequence of deeper problems with Odo's rule. The source of his legitimacy as ruler of the whole realm had been his ability to protect it against the Northmen. Yet, by 892, the symbolic capital of the victories of the siege of Paris and the battle of Montfauçon had been used up by his inability to more than temporarily contain the Vikings or divert their incursions by paying them off.230 Furthermore, the conflict between Baldwin and Odo reveals a strong mistrust231 of the count towards the king. Baldwin refused Odo's initial proposal to come to terms peacefully while

224 Annales Vedastini 892, 71*.* On this conflict, see also chapter VI.2 and Lößlein, Ressources.

225 Annales Vedastini 892, 72; Regino, Chronicon 892, 139–140; Flodoard, HRE IV, c. 6, 392.

226 For a detailed study of the conflict, see chapter VI.2.

227 Ramnulf and Ebolus were brothers. Regino, Chronicon 892, 140: *Post haec in Aquitaniam proficiscitur contra Ramnulfum et fratrem eius Gozbertum et Ebulonem abbatem de sancto Dionisio et alios nonnullos…* Annales Vedastini 892, 73: *At ubi fines attigit Aquitaniae, Ebulus, eius adventum praesciens, in fugam versus interfectus est iuxta quoddam castellum lapide; frater quoque eius Gozbertus post haec obsessus atque in brevi vitam finivit.*

228 On Abbot Ebolus see Bautier, Recueil Eudes, Introduction, XXI–XXV.

229 Altmar supported Odo even before he became king. He appears as the first witness in a charter Odo issued for Saint-Martin in 887 (DOdo 55).


later on, in 895, the king's determination to rebuild trust in him becomes visible when he returned Saint-Vaast to the count without concluding a peace treaty.232 This mistrust was probably rooted in Odo's actions as king. The conflict around the succession at Poitiers shows Odo acting against the interests of those close to him.233 In addition, also the king's decisions in the case of his own relative, Walker, whom he had entrusted with the castle of Laon but who had made peace with Baldwin, can be noted. Walker was sentenced to death, but then, according to the *Annales Vedastini*, the king did not pardon him.234 That the annals make note of this event reflects that, in this case, Odo did not act according to common expectations, but rather violated them. In this context, we can also safely assume that the denial of absolution to Walker by Bishop Dido of Laon was the king's doing as well.

Against this background, Anskeric's claim that the king had ordered him to do intolerable things, as reported in Fulk's letter to Arnulf, makes sense. Whether this had actually happened or not did not matter, nor did the actual content of these orders. Odo's own actions appear to have already created an atmosphere in which such accusations could easily be believed, as there were enough other examples of his intolerable behaviour to give them credibility. By the end of 892, Odo had not only lost his legitimacy as victor against the Normans, but may also have been deeply mistrusted by many important nobles. Next to the political ambitions of Fulk and some of his allies, this possible crisis of trust in the king may well serve as an answer to why so many nobles from Francia chose to oppose Odo.

#### **I.2.4 The cohesion of the network**

A brief account of the ups and downs of the rebellion will allow us to draw some conclusions about the strengths and weaknesses of the alliance which had formed around Charles. After Fulk and his allies had elevated Charles to the throne, they mobilised an army to march against Odo. The following encounter ended without any blood being shed after all parties withdrew. In autumn Odo managed to surprise Charles' group and force them out of the realm. Yet, in another surprising march, the allies returned to Francia in September and an armistice was concluded until Easter 894.235 The end of the armistice saw both sides gathering their forces, with Odo besieging Charles at Reims. Leaving the city garrisoned, Fulk and

<sup>232</sup> For a detailed study of the conflict, see chapter VI.2, were we further elaborate on these thoughts.

<sup>233</sup> See chapter VI.2.

<sup>234</sup> Annales Vedastini 892, 72: *Nam antea Walkerus eius consobrinus castrum Ludunensium, quod a rege perceperat, per tirannidem obtinuit, sed rex castellum obsedit ipsamque civitatem mox cepit. Et post paucos dies diiudicatus, sed rex non sibi praevidit, capite eum iussit truncari*; contrary to the annals Regino (Chronicon 892, 139−140) reports that Walker was executed because he had drawn his sword against the king in a public assembly.

<sup>235</sup> Annales Vedastini 893, 73−74.

his allies took the young Carolingian and moved to meet Arnulf, who acknowledged him as king and sent Lotharingian troops to support him. Encountering Odo, however, the Lotharingians refrained from fighting him and returned home, forcing Charles to seek refuge with Richard the Justiciar in Burgundy, where he was pursued by Odo, although without success. Deprived of their supplies in Francia and now in conflict with Richard, Charles' allies then went to devastating the region.236

In 895, events took a new turn, as Arnulf succeeded in installing his son Zwentibold as king in Lotharingia. In return for the promise of a part of the Western realm, Zwentibold agreed to an alliance with Charles and joined forces with him to besiege Laon, forcing Odo to withdraw behind the Seine. The alliance between Charles and Zwentibold, however, soon became fragile and those who were with Charles sent to Odo to open negations: Charles should be granted a part of the realm, they themselves should receive the king's peace. Odo's subsequent return to Francia forced Zwentibold back to Lotharingia. The negotiations between Odo and Charles' supporters continued until Easter 896,237 while Odo remained in control of almost the whole of Francia apart from Reims, leaving Charles' party short of supplies to get through the winter. Charles' allies now withdrew to the Moselle and warred against Baldwin of Flanders and his brother Rodulf. In 896, the negotiations finally came to an end and Odo was willing to agree to the terms proposed the year before, leaving Charles the part of the realm held by his supporters at that time. The decisive assembly, however, was disrupted by Count Rodulf, leading Heribert and Erkanger to abandon Charles and go over to Odo, soon, although reluctantly, followed by Fulk. Charles sought refuge in Lotharingia once more and aimed to create an alliance with a group of Northmen led by Hundeus when his remaining supporters reopened negotiations with Odo. This time, they succeeded: peace was restored, Charles was granted a part of the realm and "promised more"; in the context of the developments soon to come, this was undoubtedly the right to succeed Odo. 238

Several observations can be drawn from this account: a) After the initial success of Charles' party, Odo's military superiority soon became evident. He managed to drive his opponents out of Francia and brought their strongholds under his control, thus depriving them of their bases of supply. b) Their own weakness evident, Charles' supporters sought out potential allies from the beginning—here we have seen Arnulf, Richard, Zwentibold and Hundeus' Northmen. We will turn

<sup>236</sup> Annales Vedastini 894–895, 74−75.

<sup>237</sup> Annales Vedastini 895, 75−77.

<sup>238</sup> Annales Vedastini 897, 78−79: … *Verum post haec hi qui cum Karolo erant, videntes suam paucitatem et nullum tutum habere locum refugii, iterum ad Odonem regem dirigunt, quatinus ad memoriam reduceret, quod senior eorum filius esset sui quondam senioris, et partem aliquam ei ex paterno regno concederet. At vero rex cum consilio suorum respondit se illi velle misereri, si sibi liceret; et intercurrentibus nuntiis venit Karolus ad eum; quem ille benigne suscepit, deditque ei tantum e regno, quantum sibi visum fuit, promisitque maiora et remisit eum ad locum suum…*

to them and others in detail further below. c) Several of these allies soon turned into enemies. d) Despite his superiority on the battlefield, Odo appears to have had a fundamental interest in re-establishing peace with Charles and his allies. e) Negotiations appear to have been complicated, starting in 895 and taking most of that and the following year. The initial proposition from Charles' allies appears to have been the basis for the final agreement: a partition of the realm and the reentering of those who had supported him into Odo's peace. f) Only one moment occurred when defections actually took place: after the failure of the negotiations in 896, when almost all of Francia had been lost to Odo and no more potential allies were on the horizon.

Given all this, the cohesion of the group around Charles appears remarkable. Only after three and a half years of war, after having lost their bases and without the prospect of other potential alliance partners, and at the same time embroiled in conflict with Baldwin and his brother, did members of the network go over to Odo. Despite the constant setbacks, Fulk and Charles' other supporters demonstrated a notable ability to seek out allies against Odo. While their own strength evidently would not suffice to prevail against the Robertian, in alliance with Zwentibold they succeeded in forcing Odo out of Francia. Even by itself, this sufficed to pose a constant threat to Odo and to divert him from other duties, such as defending the realm against the Northmen, as the *Annales Vedastini* remark.239 Thus Odo was as interested in coming to terms with the rebels as they were—the only question remaining was what price would have to be paid.

#### **I.2.5 Allies sought within the realm**

Charles' network of supporters was certainly stronger than that which had tried to elevate Wido of Spoleto and then Arnulf back in 888, yet it still was not in a position to guarantee success in imposing a new king. Other possible allies had to be sought out, within and beyond of the realm. The most obvious course was to look to those already in conflict with Odo—the Ramnulfids and Baldwin II of Flanders. In its entry describing late 892, the *Annales Vedastini* mention "others" who joined up with the prospective rebels around Fulk to convince Odo to leave Francia.240 The Ramnulfides have often been linked to the rebellion241 and it might indeed be that they are the "others" here, yet, as no other source indicates a con-


nection between the two groups, this must remain mere speculation. In any case, the Ramnulfid leaders Ebolus and his brother Gauzbert died soon after,242 thus ending the possibility of further cooperation.

Baldwin, on the other hand, had already been allied with Fulk in the succession of 888, when they and Abbot Rodulf, the count's cousin,243 had approached Arnulf to take the West Frankish crown.244 After the death of Rodulf in 892, Baldwin, against Odo's will, claimed his cousin's abbeys, Saint-Vaast and Saint-Bertin, for himself.245 Fulk quickly became involved in the ensuing conflict. He and his suffragans threatened all those who communicated with the count with anathema246 and sent another letter to Baldwin himself in which they reproached him for his misdemeanours and asked him to repent, threatening him with excommunication if he did not comply.247 However, although Baldwin still offered resistance to Odo at the end of 892, in another letter composed at a synod at Reims the bishops seemed to be much more obliging. The count was informed that he had indeed been excommunicated for his deeds, but that this excommunication had been suspended for the time being, due to possible advantages for state and church.248 The conclusion that Fulk wanted to spare Baldwin in order to leave open the possibility for him to join the ranks of the conspiracy against Odo comes to mind249 and indeed the count and his brother are found at Charles' side in 895.250 Yet nothing indicates when they had actually allied themselves with the group around the Carolingian. The siege of Laon, however, certainly saw the end of their cooperation. Baldwin and his brother Rodulf went over to Zwentibold and fought Odo on their own when Charles' allies started negotiations with the Robertian.251 Over the winter of 895/896, when Charles and Odo had agreed upon an armistice, fighting ensued between Charles and Baldwin,252 indicating that the rift between the two groups had become serious.

In 896 Rodulf also appeared at an assembly which concluded long lasting negotiations between Charles' supporters and Odo. Odo had finally agreed to leave


<sup>242</sup> Annales Vedastini 892, 73.

Charles the part of the realm his followers had owned before the war, when Rodulf disrupted the assembly and caused the whole agreement to fail.253 The question arises: why did the intervention of Rodulf, who belonged neither to Charles' side nor Odo's, have such dire consequences? The only possible explanation is that the negotiations sought not only a solution for the conflict between Odo and Charles, but a general peace including Baldwin and his brother. This meant that an arrangement satisfying all three parties had to be found. When Rodulf derailed the negotiations, we can conclude that a decision was close to being made that would have been to his disadvantage. It seems safe to assume that, during their conflict with Charles' supporters, Baldwin and Rodulf had taken over *honores* belonging to or claimed by Charles' followers. This was definitely true for *honores* belonging to Odo's allies in the region: Saint-Quentin, now occupied by Rodulf, had been in the possession of the son of Count Theoderic, one of Odo's most loyal followers. A general peace would have meant that some or all of these *honores* would have had to be returned to their original owners—something that Rodulf would not agree to. The problem was solved over the following winter when Baldwin's brother was killed in battle with Heribert. 254 In 897, the different parties, including Baldwin, came to agreements similar to those of the year before and peace was finally restored. However, the underlying conflicts, Baldwin's desire to extend his influence in the North and his enmity with Heribert continued.255

When trying to determine Baldwin's and his brother's position concerning Charles, the central point is not their timing for joining the Carolingian's side, but that they had already been in conflict with Odo before the start of the rebellion. Their behaviour during these years does not point towards a fundamental opposition to Odo *per se*, but their aim to extend their influence in the north-east of Francia. The contentious point between Baldwin and Odo was the abbey of Saint-Vaast, which Baldwin had claimed for himself after his cousin Rodulf 's death.256 Furthermore, Rodulf had used the confusion in Francia to acquire Saint-Quentin, something that Odo could not tolerate either, since the abbey had been in the hands of the son of one of his closest supporters. Both Baldwin and Rodulf appear as independent actors in the political arena of Francia, their support for Charles


not aimed at disposing Odo but at increasing the pressure on the king to come to terms with them.

The character of Baldwin's support for Charles brings us to another group of nobles whose position has been the object of scholarly discussion ever since the publication of an article by Jean-Pierre Brunterc'h.257 Following Charles' coronation, the *Annales Vedastini* report that:

After Easter Archbishop Fulk and Count Heribert took King Charles and set out with all the army they could muster against King Odo. Richard, William and Altmar came against them with a considerable army. Against whom King Odo did not hesitate to come. And he sent [messengers] to those who were with Charles, ordering that by his guarantee they should make amends for whatever wrong they had done them and remember the oaths they had sworn to him.258

Until Brunterc'h's article, it had been assumed that Richard (the Justiciar), William (the Pious) and Altmar (of Poitiers) came to support Charles and oppose Odo, 259 a reading that would be supported by the circumstance that both William and Altmar had been involved in the conflict ensuing around the succession of Ramnulf II at Poitiers in 892.260 However, Brunterc'h argues that, in fact, the three magnates were allies of the king, having joined him in April or May 893 at the latest.261 He translates this passage of the annals as "À l'encontre de ces derniers (= Richard-Guillaume-Adémar), le roi Eudes vint sans tarder et dépêcha ses représentants à ceux qui étaient avec Charles, leur enjoignant de réparer par son gage tout ce qu'ils avaient commis contre eux (= Richard-Guillaume-Adémar) et de se souvenir du serment qu'ils lui avaient prêté." Odo, therefore, assumes the role of a mediator, offering his guarantee to Richard, William and Altmar for deeds done by Fulk and his allies.262 Brunterc'h further links the three magnates' alliance with Odo to a number of measures taken by the king: Richard was given the abbey of Saint-Germain of Auxerre, most probably in 893 or 894,263 while William received the abbey of Saint-Julien of Brioude by March 894 at the


<sup>257</sup> Brunterc'h, Naissance.

<sup>258</sup> Annales Vedastini 893, 73−74: *Post Pascha Domini Fulcho archiepiscopus et Heribertus comes assumentes Karolum regem cum omni exercitu disponunt [ire] contra Odonem regem, veneruntque contra eos Richardus, Willelmus et Hadamarus, habueruntque exercitum copiosum. Contra quos rex Odo venire non distulit. Misitque ad eos qui cum Karolo erant mandans ut quicquid in eis deliquissent per suum eis vadium emendarent et memores essent sacramenti quae sibi iuraverant.*

<sup>259</sup> For example Eckel, Charles, 11, Chaume, Origines, 376, Kienast, Vasallität, 471. This reading was supported by Altmar and William being in conflict with Odo during 892. See below.

<sup>260</sup> See chapter VI.2.

<sup>261</sup> Brunterc'h, Naissance, 81.

latest.264 Finally, Bishop Adalgarius of Autun became Odo's new archchancellor before 28th May 893.265

While this reading of the events has subsequently been accepted by scholars,266 Steven Robbie has recently brought up a new argument, challenging Brunterc'h's view that Adalgarius had been a close supporter of Richard. 267 According to him, in light of their previous behaviour, both were "more likely at loggerheads tha[n] natural collaborators."268 He proposes an alternative reading: Richard, William and Altmar indeed came up in support of Charles and Odo successfully made haste to dissuade them.269 Robbie's analysis can be used as a starting point for further thoughts: while Brunterc'h's reading does agree with the syntax, he may be overly precise with his translation—the *Annales Vedastini* are known to contain grievous grammatical errors.270 It is hard to imagine what wrongs Fulk's party had done to the three magnates that would require compensation and royal mediation; at the same time William and Altmar (and possibly Richard as well) were in conflict with Odo himself.271 In addition, why should the annalist emphasise that Odo did not hesitate to march against them, when he already had the support of these powerful men? Finally, while it is certain that Saint-Germain of Auxerre and Saint-Julien of Brioude came under the control of Richard and William respectively, these acquisitions cannot be dated precisely enough to determine whether Odo had already granted them before Easter 893. That these grants were related to these events is indeed possible; however, another explanation seems more plausible, especially considering that over the next few years neither William nor Altmar appear again in the struggle between Charles and Odo, while Richard subsequently behaved like "a loose cannon," as Steven Robbie put it.272 Richard, William and Altmar were not Odo's allies at that particular moment, but nothing indicates that they supported Charles. In fact, Fulk asked Bishop Teutbald of Langres to inquire into the attitude of Richard and the Aquitanians.273 If they had been allied, this would not have been necessary. It seems therefore most likely that they marched up north with an army as a third party, one that would offer itself to the highest bidder. Odo


then won the race by granting these magnates important abbeys, buying not their support, but their neutrality.274

Neither William, Altmar nor Richard can therefore be linked to Charles and his supporters at the beginning of the rebellion. At least two of them, William and Altmar, had already been in conflict with Odo and none of them appears to have fundamentally opposed his rule. Like Baldwin and his brother Rodulf, they used the chaos in Francia to enforce their claims. Faced with the rebellion of Fulk and his allies, Odo needed to pacify this conflict; it did not challenge his rule as such, but prevented him from concentrating his forces in Francia. While Altmar soon returned to Odo's side, the image is less clear for William. 275 A charter from Saint-Julien of Brioude dates to "the fifth year of the reign of King Charles,"276 corresponding to 897 and therefore the year of the final agreement between Charles and Odo. 277 Unfortunately, the charter, while naming William as lay abbot, does not indicate whether he was present at that moment or had had a hand in the proceedings in any other way. We might therefore suggest, although with great care, that in 897 William was in contact with Charles and thus increased the pressure on Odo.

Richard, despite the agreement with Odo, became repeatedly involved in the conflict. In 894, Charles and his allies were forced to leave Francia and seek shelter with Richard in Burgundy. Odo followed them, but failed to secure a victory in battle. It was undoubtedly during this campaign that he stayed at Flavigny, issuing a diploma for the abbey of Montiéramey.278 This was not only an attempt to prevent Charles' supporters from establishing a new base in northern Burgundy, where Adelaide's family had connections.279 It also affirmed Odo's control of the Troyes region, where Montiéramey, as well as the donated goods, were located. Furthermore, the intervention of Count Hucbald in the diploma was a symbol of Odo's own ties to and control over Francia, the heartland of Charles' supporters. Finally, his stay at Flavigny might also be taken as a gesture against Richard, who, after the death of Bishop Adalgarius of Autun, Odo's archchancellor, had installed


Walo, brother of his right hand man Manasses, in this see, although the community had elected Aquinus, the provost of Flavigny. 280

Whether Charles' party had originally intended to move as far south as Richard's lands remains an open question, but, without a doubt, Odo's campaign did not leave them much of a choice. Even if Richard actually agreed to an alliance with Charles' party, it did not hold for long. By the end of 894, Manasses had attacked Bishop Teutbald of Langres, who had opposed the installation of Walo at Autun, and blinded him.281 Since Teutbald had been close to Fulk both sides entered into conflict.282 Odo, at this point if not earlier, had changed his strategy from trying to overwhelm his opponents by force to depriving them of their bases of supply, taking their possessions in Francia.283 The situation of Charles' supporters turned desperate. Without supplies from Richard they were now forced to live off the land, leading to a wave of violence and devastation, emphasised by all contemporary witnesses.284

This did not mean the end of contact between Charles and Richard. In the winter of 895/6, a meeting in southern Lotharingia took place at Remiremont.285 Emperor Lambert and King Rudolf signed next to Charles, yet also present were Manasses and Rampo. Manasses was the right hand man of Richard the Justiciar and Rampo was involved in the blinding of Bishop Teutbald of Langres. In this affair, Fulk had not only tried to protect his relative Rampo, but had also done his best to delay the excommunication of Richard and Manasses demanded by the pope.286 Whether Charles gained any actual support during this time is not apparent from the sources, but he remained in contact with Richard. In late 896, Richard and Charles were still negotiating, as is signalled by the presence of the count of


the palace Elduin at Richard's court.287 Since the charter is dated by Charles' reign, we can conclude that Richard did not favour Odo at that moment.288 Once peace in Francia was restored, Richard also re-entered the king's favour. A diploma dating to October 897 shows him at Odo's court, honouring him as "Richard, our illustrious and beloved count."289

To summarise, we should note that, inside the realm, two leading nobles appear to have allied themselves with Charles' supporters: Baldwin of Flanders, together with his brother Rodulf; and Richard the Justiciar. Neither alliance lasted for long and at least for some time the rebels ended up in open conflict with both parties. This points towards the general character of these coalitions. They came into being because both sides hoped to profit from each other in their individual struggles against Odo. That they proved to be only short lived and were superseded by hostilities was the result of the fundamental differences in the nature of their opposition to the king. Neither Baldwin nor Richard aimed at replacing Odo with another king. Both fought to extend their influence. In the end, under which king they would achieve this did not matter. This freed their own policies from the restrictions the members of the network' suffered from. They could change sides whenever they deemed it profitable. Fulk and the others had to remain with Charles or abandon everything they fought for.

#### **I.2.6 Allies sought outside the realm**

Those who were with Charles not only sought allies from within the realm, they also looked for supporters from outside. Such support could come from the kings of Eastern Francia, Lotharingia (once it had become independent under Zwentibold), Italy, Upper Burgundy, Provence and the pope. By the end of 896, Fulk had turned to almost all of them, and in some cases more than once. The first choice was Arnulf, whose superiority had already been acknowledged by Odo in 888. Fulk's letter to him was not only meant to soothe Arnulf 's anger over the coronation and to prevent him from intervening in Western Francia. It was also aimed at winning his trust and getting him to renounce his support for Odo while recognising Charles, and, finally, obtaining material support. That this also meant accepting Arnulf 's superiority did not really matter at that moment, as long as

288 One could add to this a charter from Saint-Julien of Brioude (Grand Cartulaire, N° CCCXXXVII, 96), where William the Pious was lay abbot and which dates *anno V regnante Karolo rege* (corresponding with 897 taking in conseridation the terms of office of Provost Eldefred and Deacon Bernard, both named in the charter; Kienast, Vasallität, 466, n. 1654). The charter names William as lay abbot, but does not otherwise indicate whether he was present at that moment or had had a hand in the proceedings in any other manner. We may therefore assume, although only with great care, that also William was in contact with Charles or putting pressure on Odo at that moment.

289 DOdo 42 (21st October 897, Nanteuil-le-Haudouin), 180: *Richardus, illustris dilectusque nobis comes*. Richard asked for goods in Atuyer to be granted to a certain Gislebert.

<sup>287</sup> Cartulaire Montiéramey, N° 12, 18 (21stDecember 896).

Arnulf confirmed Charles as the legitimate king and thus demonstrated that his claim was just. In 893, Fulk's efforts to win Arnulf for Charles' cause proved futile, yet a year later the king finally changed his position and agreed to meet Charles at an assembly at Worms.290 Fulk and his allies brought rich gifts and Charles was benignly received. The Carolingian made a *promissio* of which nothing further is known, whereupon Arnulf granted him the kingdom of his father and even provided military assistance.291

Arnulf 's motivation to change his course so completely has been the subject of ample discussion.292 Most convincing so far is the pragmatic view of Arnulf 's situation offered by Thilo Offergeld. According to his argument, Odo's position had been strengthened by his recent successes against Charles. In contrast, the latter would have been, as his *promissio* shows, open to a more complete submission under Arnulf 's supremacy. Furthermore, it would have been in Arnulf 's interest to keep the conflict between the two rivals going, not least to secure the Lotharingian frontier. To shift his support to the weaker party would therefore have been a reasonable decision.293 However, a few additions can be made to this reading. Odo's position should not be overestimated. He had succeeded in appeasing Aquitaine and prevented Richard the Justiciar from joining his enemies' ranks, but he had failed in crushing the rebellion in Francia, as well as in solving the ongoing conflict with Baldwin II of Flanders. The peace talks at Reims might have brought these conflicts to an end, which would have left Odo his full strength to pursue whatever other goals he might have had at that moment. This outcome had the potential to pose a threat to Arnulf himself, who had only recently returned from Italy without achieving anything decisive. Even worse, on his march back home he had also failed to suppress King Rudolf of Upper Burgundy and almost led his army into disaster while crossing the Alps.294 How weak his position actually was at that moment is revealed by his subsequent failure to gain the consent of the Lotharingian nobles to the installation of his son Zwentibold as their king at a


meeting at Worms.295 Fulk's offer, that Arnulf acknowledge and support Charles in return for Charles' acceptance of his supremacy, would not only have prolonged the conflict in the West and secured the Lotharingian border from any possible threats, but would also have served to demonstrate Arnulf 's superiority to his own nobles, thereby strengthening his rule after a series of failures. Turning to Charles, therefore, was a pragmatic course of action not chiefly because of Odo's strength, but because of Arnulf 's own problems.

The military support provided by Arnulf appears to have been composed of the levies of some Lotharingian bishops and counts from the Meuse area.296 Once this host encountered Odo, however, the Lotharingians decided to keep their *amicitia* with the Robertian and refused to confront him.297 The leading members of Charles' party seem to have been well aware of this problem and did their best to keep the Lotharingians on Charles' side. A diploma of Charles issued at Attigny,298 an important royal palace on the Aisne, is a sign of their endeavours. At the request of Charles' mother Adelaide and Archbishop Fulk, Charles restored parts of the fisc of Arches in Porcien to Bishop Franco of Liège. It has long been assumed that this was done because Franco had been part of the Lotharingian forces which supported Charles, 299 but the actual reasons run deeper. In fact, as a diploma of Charles the Fat shows,300 Franco had been close to Odo's brother Robert. This new diploma, therefore, was issued to demonstrate to those other Lotharingians who had ties with the Robertians, that at least Franco would not leave Charles' side in the upcoming confrontation. Furthermore, it was a sign that the kind of fidelity shown by the bishop would be rewarded not only materially, but also with honour, as demonstrated by the intervention of Adelaide. 301 As we have seen, this did not prove entirely successful. Confronted with Odo's army, the Lotharingians refused to fight for Charles. However, they did not abandon Charles right away either. From the account of the *Annales Vedastini*, it seems that they treated with Odo. Before they left to return home, they secured Odo's departure as well, thus depriving him of the opportunity to crush the rebellion at that moment.

Arnulf 's next intervention in West Frankish affairs took place in the following year, when he responded to the devastation in Burgundy caused by the war by

<sup>295</sup> Regino, Chronicon 894, 142. The *Annales Fuldenses* do not note this event.

<sup>296</sup> Offergeld, Reges pueri, 435.

<sup>297</sup> Annales Vedastini 894, 74−75; Regino, Chronicon 893, 141.

<sup>298</sup> DChS 5, 26th September 894.

<sup>299</sup> E.g. Kienast, Vasallität, 474.

<sup>300</sup> DChF 105, dated to 884. In this diploma, Franco and Robert both ask the Emperor to grant a manse to Robert's *fidelis* Sanctio. The identification of this count with Odo's brother has been made by Wollasch, Gerard, 63, unfortunately without giving any further arguments. However, later on the only interventions made by Robert in Lotharingia during the reign of Charles the Simple are both connected to the church of Liège (DDChS 65 and 81), so that an earlier relation between Robert and the church—as demonstrated by DChF 105—seems very likely.

<sup>301</sup> On the role of Adelaide's interventions, see chapter III.2.2.

inviting both Odo and Charles to meet him.302 As in the previous year, his intervention may partly have been motivated by internal affairs. After his failure to install his son Zwentibold as king of Lotharingia, he now planned a second attempt. Having one or even both of the West Frankish kings acknowledging his superiority would aid his cause and their presence at Zwentibold's coronation would also help to secure his son's new position. While Odo responded, coming with rich gifts, Fulk's party decided not to accept the invitation.303 Instead, the archbishop sent a letter to assure Arnulf of the validity of the *promissio*, given by Charles the year before, and announced the preparation of an attack on Odo. 304 In consequence, Odo was received with all honours by Arnulf, returning home after the coronation of Zwentibold. On his way back, he encountered Fulk, who was finally on his way to Arnulf with further gifts. Fulk escaped but Count Adalung, who had accompanied him, was killed and the baggage lost.305 Why had Fulk's party not responded to Arnulf 's invitation earlier? It has been proposed that they refrained from doing so since the atrocities committed in Burgundy, which were the reason for Arnulf 's call to Odo and Charles, had been their doing.306 This may have been the case, but it is also possible that after the visit to Worms the year before, it had not seemed likely that Charles was in danger of losing Arnulf 's acknowledgment of his kingship. Only when the news spread that Odo had accepted the invitation and had been received amiably did it became apparent that a letter would not suffice and therefore Fulk hastened to Worms to limit the damage already done.

If Odo had hoped to deprive Charles of support from the East by his presence at Worms, he failed. Zwentibold, newly installed on the Lotharingian throne, did not hesitate to seize the opportunity when Charles' supporters offered him parts of the Western kingdom and allied himself with Charles. 307 Together they launched an attack on Laon to which Odo was incapable of responding. His forces exhausted, he withdrew over the Seine, leaving the defence of Laon to Bishop Dido. 308 Without pressure from Odo, however, the alliance between Charles' party, which by this point had been reinforced by Count Baldwin and his brother Rodulf, and Zwentibold did not hold. The count of Flanders went over to Arnulf 's son, and subsequent rumours of a plot to assassinate Charles then led to a final break. Zwentibold left the camp and later concluded a separate peace treaty with Odo. 309 The events at Laon, however, do not seem to have caused lasting damage to the


relations between Charles and Zwentibold. When in 896 Charles was forced to flee from Western Francia, he sought refuge in Lotharingia.310

Besides Arnulf and Zwentibold, Fulk also turned to Wido of Spoleto. His relative, whom he had invited in 888 to take over the realm only to abandon him shortly after, had briefly been considered as an alternative candidate to Charles yet again.311 This option seems to have been dismissed quite quickly, probably because five years earlier Wido had proven that he was not able to bring together sufficient support on his own and was unlikely to be acknowledged as king by a majority of the leading nobles.312 Charles, on the other hand, even if he could not provide any resources on his own, had Carolingian blood which could serve to legitimise the rebellion and was more likely to serve as a rallying point for other nobles than Wido. Instead, Fulk wrote a new letter to Wido, congratulating him on having been crowned emperor (two years earlier) and asking him to protect and support Charles. 313 Furthermore, he warned him of Arnulf preparing a campaign against him.314 This was undoubtedly meant as a sign of Fulk's good will, aimed to make Wido ready for an alliance. Whether this endeavour was crowned with success we do not know, since the sources do not mention a response on Wido's part. In any case, given his warning that Arnulf would soon invade Italy, it seems debatable whether the archbishop actually expected the emperor to intervene in Western Francia.315 It appears more likely that Fulk's true purpose was to gain Wido's acknowledgement of Charles and to demonstrate that Charles kingship was legitimate.

It is likely that, in late 895, when the negotiations with Odo entered their final stage, Fulk renewed his diplomatic efforts to win royal support for Charles. The young king met with Wido's son, Emperor Lambert, and King Rudolf of Upper Burgundy at Remiremont.316 For most of the year, Lambert had been targeted by the archbishop as a potential ally317 and the meeting at Remiremont was undoubtedly the result of these endeavours. Lambert himself probably hoped to win military support against Arnulf318 and the same was likely also true for Rudolf, who himself was on the defensive against Zwentibold in the north and the kingdom of Boso's son Louis in the south.319 The Remiremont meeting demonstrates the


range of Fulk's connections, but at the same time it also betrays that the only allies Charles could find at that moment were as much in need of help as he himself was. As Eduard Hlawitschka remarked:

Daß die […] bedrängten Herrscher […] sich bei ihrem Treffen mitsamt ihrem Gefolge in das Gedenkbuch in Remiremont einschreiben ließen, kann nicht verwundern. Erwartete man doch von dem mit der Einschreibung verbundenen Gebetsgedenken nicht nur zukünftiges Heil, sondern Gottes Hilfe in der Gegenwart. Und diese war ihnen bitter not.320

Pope Formosus was also deeply involved in Fulk's endeavours to win support for Charles' cause. He appears to have been the only one to quickly recognise Charles as king, congratulating him on his elevation, granting him the *panem benedictum* and giving him some counsel321 while at the same time trying to mediate a peaceful solution with Odo. Formosus proposed that both parties agree to a ceasefire, with Fulk travelling to Rome and the Gallic bishops working towards re-establishing peace.322 Fulk's plans were different: he asked the pope to win Arnulf 's support for Charles and to threaten Odo not to devastate the realm,323 without success. Arnulf seized the possessions of the church of Reims while Odo continued his efforts of suppressing the revolt.324 Fulk, nevertheless, redoubled his efforts of increasing diplomatic pressure on Odo after the negotiations for an agreement had begun. In another letter to Formosus he asked him again to write to Odo as well as all the nobles to urge them to keep the peace and acknowledge Charles' hereditary right or at least give him a part of the realm.325 As he had done in the cases of Arnulf and Odo, the archbishop also tried to use Formosus as a mediator between Charles, Wido and, after Wido's death, Lambert. 326 At least in Lambert's case, the pope appears to have indeed spoken for Charles and probably contributed to the realisation of the alliance. Formosus' death, however, brought an end to Fulk's influence in Rome. From the beginning, his successor Stephen VI, who exhumed Formosus to accuse him of perjury, reproached the archbishop for not coming to Rome and threatened him with the penalties of canon law if he chose not to be present at a synod later in 896.327


The key figure in finding support from potential allies outside the realm was, as we have seen, Archbishop Fulk, who used his contacts in the East and in Italy. Very much like the support Charles' allies won from nobles inside the realm, these coalitions resulted from short term goals on both sides and were not based on longterm common interests. Fulk was interested in gaining ideological and material support for the rebellion, Arnulf in demonstrating his superiority and Zwentibold in enlarging his realm. The contacts with Wido, Lambert and Rudolf appear not to have brought about any practical advantage, as their individual interests were too different. Lambert and Rudolf seem mostly to have been interested in joining forces against Arnulf instead of uniting against Odo. Pope Formosus' role should be assessed in the same way. His voice for peace appears to have gone unheard, his letters to Odo and Arnulf concerning Charles seem to have remained without effect. However, all of these contacts also tell us something about how Charles' coronation was perceived outside Western Francia: Arnulf, while initially reluctant, nevertheless acknowledged the Carolingian as king. Pope Formosus did so without hesitation, yet at the same time continued to call Odo *rex* as well. Wido, Lambert and Rudolf also had no problem with Charles being one of them, a king. We will return to the implications of that observation in our conclusion to this chapter.

#### **I.2.7 A network of different layers**

The network of men rebelling against Odo consisted, as we have argued, of different layers. In the centre were men like Archbishop Fulk of Reims, Bishop Anskeric of Paris and Count Heribert of Vermandois, all powerful nobles from Francia. The archbishop, as we have seen, appears to have been mainly driven by political ambition. The same may have been true for Heribert and the others as well, yet this has to remain an assumption. It is, however, remarkable that all of them appear to have been more or less well integrated into Odo's rule, the bishop of Paris even serving as the king's archchancellor. As an explanation, we have proposed that they were blocked by others already within Odo's most inner circle from achieving their goal of gaining more influence at the royal court, of entering this same circle. Since these others had already been there before Odo became king, they were able to defend their position and thus block the advance of the likes of Fulk and the others.

However, this explanation seems to be unsatisfactory in the case of those not belonging to the inner core of the network forming against Odo. While in some cases the participation of individual members can be explained by personal relations to those in the centre—for example Heribert's brother Pippin and the suffragan bishops of the diocese of Reims—in other cases their participation cannot be explained so easily. To assume that they were also driven by their ambitions would mean to fall back on thought patterns long discarded by scholarship that suppose that the relations between the kings and the nobles were characterised only by a constant struggle for power. In addition, taking the case of Bishop Anskeric into consideration, who was already at the centre of Odo's court, the motive of political ambitions seems to fall short of explaining the entirety of the rebellion. Seeking a more satisfying explanation, we have argued that another common trait of the rebels was their distrust in Odo, which was caused by a series of ill-received actions on his part. Whatever it was, it was a strong glue that held these men together. Only when everything was apparently lost, when their lands had been occupied by Odo and the negotiations seemed to have failed, did the network begin to disintegrate.

At the beginning of this chapter, we have referred to the perception of this network as being highly fragile. This image is not entirely wrong. Yet it does not apply to the core group of rebels but to the alliances they formed with men outside their own network. As they were too weak on their own to overcome Odo, they sought out potential allies. Within the realm these were men who had already been in conflict with Odo or were about to challenge him. In any case, while sharing a common enemy—the king—these nobles did not share a common goal with the rebels. They were defending their claims or trying to extend their influence, but they did not challenge Odo's rule *per se*. The same is also true of the allies Fulk sought outside the realm. Whether it be Arnulf and Zwentibold or Lambert and Rudolf, they all were driven by factors relating to their own advantage, not by the goal of overthrowing Odo. It was therefore in the nature of these alliances of convenience to break as soon as the common goal proved to be weaker than individual interests.

Consistent with this image is the fact that, at the very end of the rebellion, Charles and his remaining supporters made contact with yet another potential ally who did not fit neatly into the groups otherwise sought out within and beyond the realm. In 896, a group of Northmen under a leader called Hundeus had taken the opportunity presented by the ongoing fighting, between Rodulf and Baldwin on the one side and Odo and Heribert on the other, to return to Francia. Their strength quickly increasing, they installed themselves on the Oise and then turned towards the Meuse, from where they were finally driven back towards the Seine by the king. Charles now turned to them, baptising Hundeus at Easter. Over the summer, the Normans plundered the Seine region without encountering any further resistance.328 A letter from Fulk indicates his reaction when he discovered that Charles had planned an alliance with the Vikings: should Charles proceed with the alliance, the archbishop not only threatened that he would stop supporting him, but he would also encourage others to do the same, and, furthermore, he would excommunicate the Carolingian.329 No further evidence exists that this

<sup>328</sup> Annales Vedastini 896 and 897, 78.

<sup>329</sup> Flodoard, HRE IV, c. 5, 384−385. While Fulk mentions rumours about plans to forge an alliance between Charles and the Normans, the *Annales Vedastini* only report that Hundeus was brought to Charles to be baptised.

alliance was ever carried out,330 yet the plans appear to be the logical result of the policy Charles and his supporters had pursued during the rebellion: lacking sufficient strength by themselves, they took every opportunity to ally themselves with others who could aid them in bolstering Charles' claim, provide them with men and resources, or at least distract the forces of the Robertian.

#### **I.3 Conclusion**

In this chapter, we have addressed two questions: the reasons why Charles was passed over from his birth up until 893 and the composition and nature of the network that finally elevated him to his father's throne. Charles, as we have seen, was born just a few months too late, when the political decisions surrounding the succession to Louis the Stammerer had already been taken. His birth was certainly noted, yet it was not relevant to the royal succession for a long time. Once in his political backwater, Charles remained there after the deaths of his brothers and, in 888, when, after the death of Charles the Fat, non-Carolingian kings rose to take the crowns of the respective *regna*. Even then, Charles appears to have lacked substantial political support, as no source shows him being considered as a candidate by any faction.

Did the problem of his birth play a role in the decisions against his succession? Louis the Stammerer, forced by his father Charles the Bald, had rejected his first wife Ansgarde to marry Charles' mother Adelaide and thus created a situation in which the legitimacy of both marriages was questionable. Indeed, as we have seen, Pope John refused to crown Louis' second wife when they met at Troyes. Yet Louis himself, up until just before his death, did his best to ensure the equal succession of his sons from both marriages and the general consensus at court appears to have been in accordance with this policy. In fact, the only time when the legitimacy of Louis' sons was questioned was in 879 when Boso had himself crowned king. He argued that he could do so due to the vacancy of the throne in the absence of any legitimate heirs, thus claiming that neither Louis' marriage with Ansgarde nor that with Adelaide had been valid. However, only the year before, Boso had betrothed his daughter to Louis' younger son Carloman and, therefore, to one of the boys he now denied the right to rule.

Charles, as discussed, does not appear to have been a viable candidate in the subsequent royal successions. Nevertheless, and whether or not his legitimacy was doubted, he remained quiet, biding his time. The moment might have come in 888: Charles the Fat's deposition and death had left the question of succession

<sup>330</sup> If we assume that Hundeus' baptism preceeded Fulk's letter, the archbishop's threats may have been the reason why the alliance was dropped. While he was not with Charles anymore, he still possessed considerable political weight to either aid the Carolingian by influencing Odo and his allies or to hamper him by using his connections against him.

open and Hugh the Abbot and Gauzlin, who had dominated politics over the past decade, were dead. This time, however, the Robertian Odo took the crown in Western Francia. His claim was not uncontested. A party around Archbishop Fulk and Count Baldwin of Flanders invited first Wido of Spoleto and then Arnulf of Carinthia to be king. Odo prevailed, but another five years later was again challenged by a group around Fulk, who this time crowned Charles king.

The death of Charles the Fat, leaving the different realms without Carolingian heirs, created a vacuum that was used by the most powerful magnates to make themselves kings. Their claims were based on their resources and their political networks. Arnulf, additionally, could rely on his illegitimate Carolingian blood while Odo could also make a case because of his victories against the Northmen. The Carolingian dynasty might have come to an end at that moment, yet Carolingian traditions and the idea of Carolingian legitimacy remained strong constitutive factors. In Western Francia, Odo behaved like a Carolingian king, trying to negate the difference in blood that separated him from his predecessors. This strength of the Carolingian idea was Charles the Simple's chance. Whatever doubts there might have been about the legitimacy of his father's marriages, he was of Carolingian blood and, as Louis the Stammerer's son, he had an undeniable claim to the crown of Western Francia. All he needed was political support to claim his throne.

That support came in the form of a strong party based mainly in Francia and centred on Archbishop Fulk of Reims, Bishop Anskeric of Paris, Count Heribert of Vermandois and Charles' mother, Adelaide. Several layers of other group members who formed the core party now elevating Charles to the throne have been identified. This group distinguished itself from other groups also in conflict with the Robertian through its cohesion and its fundamental opposition to Odo. It was strong enough to challenge Odo's right to rule, yet too weak to prevail and thus in constant need of other allies. They found those within and beyond the realm, leading to a number of alliances of convenience which fell apart as soon as the common goal proved to be smaller than the individual interests. Some of the leaders of the party seem to have been motivated by political ambition, yet this argument appears to fall short when considering the entirety of the group now rebelling against Odo. We have proposed that another motive of the rebels was a lack of confidence in the new king, caused by a series of acts that were perceived as norm-breaking behaviour.

Carolingian legitimism, on the other hand, does not appear to have been a driving motive of the leading nobles. All of them had aligned themselves with Odo before the rebellion, with some even being part of his innermost circle. When Fulk and Baldwin had opposed Odo in 888, their choice had not been Charles but Fulk's relative Wido and then Arnulf. That is not to say that Carolingian legitimacy did not play a role in Charles' coronation. It did, and in fact it was the only reason why the party around Fulk chose Charles as their candidate. What Fulk's party needed was a claimant around whom they could rally and who would attract other potential supporters. This was where the interests of Charles and those of Fulk and his allies came together: Charles had the claim that they needed to legitimise their revolt against Odo, to demonstrate to the world the justness of their cause. There may or may not have been doubts about the legitimacy of Charles' birth, yet his Carolingian blood was undeniable and could serve as an argument against the Robertian Odo. Fulk's letter to Arnulf shows how this argument was used. The throne was Charles' by right of heritage, whereas Odo was nothing but a stranger to the royal family. How strong Charles' claim was and how well it served the rebels around Fulk is also demonstrated by the behaviour of the pope and the other kings of the Frankish world. Pope Formosus immediately acknowledged Charles as king and treated him as having equal standing with Odo. Kings like Arnulf may have taken a bit longer to come around, but in the end the result was the same. They accepted Charles as one of their own, as their peer, whose right to the crown was undeniable.

Thus, Charles' blood served to legitimise the rebellion of Fulk and his allies against Odo while giving himself the opportunity to leave the political backwater in which he had been stuck since his birth. How strong this claim was is also demonstrated by the willingness of the pope and the other kings to acknowledge him. This, however, did not mean that from then on they renounced Odo's right to be king. Western Francia, for the time being, had two kings who had to find a way to coexist. The result of the final negotiations between Odo and Charles' supporters is known: Charles was given a part of the kingdom and promised even more.331 To grant Charles some part of the realm had been one of the two key demands of the rebels from as early as 895. The "promise for more" was a new addition, which is usually interpreted as an agreement that the Carolingian would succeed Odo as sole king of the realm upon the Robertian's death,332 since Odo requested everybody to hold Charles' fidelity only a few months later when on his deathbed.333

Why would Odo, victorious after three and a half years of warring, agree to such "surprising"334 terms with the defeated Charles instead of promoting his own brother's succession? The Viking threat and his own weariness to continue the fight, possibly related to his sickness, have been brought forward, as well as arguments that Odo wanted to create a buffer zone against the aggressive Zwentibold


or to limit Arnulf 's influence.335 It might be added that Odo's own position in Neustria was not as stable as believed. At some point during his rule there,336 a certain Roger had usurped Le Mans, earlier held by Count Berengar, holder of a second march in Neustria.337 Odo's brother Robert had then, with the aid of the king's troops, laid siege to the city and installed a certain Gauzlin as count. Gauzlin, however, failed to hold Le Mans and soon Roger returned—only to be excommunicated by Bishop Gunter—without much success. Roger remained in Le Mans and the bishop spent the rest of his life without ever being able again to enter the city.338 Odo had not only failed to establish his candidate in his own backyard, but his position in Neustria was now threatened by a hostile Roger.

While Odo's position was vulnerable, the return of the great Viking army to the continent in 896 threatened to undermine his own legitimacy even more. Having focussed his forces on dealing with Baldwin of Flanders and Charles, Odo had neglected the defence of the realm, as was criticised by the *Annales Vedastini*. 339 A continuation of the conflict would mean risking the basis of his own legitimacy: his ability to protect the realm from the Northmen. Therefore, Odo needed to restore peace in the realm, a task he then addressed with vigour. Baldwin finally re-entered the peace of the king340 and Richard now appeared at Odo's court in Francia, where he was honourably received.341 Charles, on the other hand, may


have been beaten by late 896, yet he nonetheless remained as a rallying point for Odo's enemies. Furthermore, despite having made his peace with Odo, Archbishop Fulk continued to use his influence at court in support of the Carolingian and it seems possible that others did so as well.342 The core of Charles' party was characterised by a strong cohesion and Fulk in particular had proven to be energetic in his attempts to strengthen the network. Even if it was now ripped apart, its remnants still possessed enough political influence to exert pressure on Odo, whose own power had also suffered from the drain of the war.

Indeed, Odo had always been open to negotiations. In 895, when Charles' supporters sent envoys proposing terms, the Robertian "agreed most willingly"343 and when the talks came to an end a year later, Odo was about to grant the Carolingian a part of the realm.344 Therefore, it seems that Odo did—at least from 895 onwards—acknowledge Charles' claim as his father's heir and recognise him as a king. Another issue also appears to have been solved early on: when Fulk and his allies rebelled against Odo, they denied the Robertian's own kingship. The negotiations also included that the rebels were to re-enter into Odo's fidelity, meaning that they no longer denied his right to rule. In addition, that Charles would be given a part of the realm had been agreed early on and only the details remained subject to negotiations.345 More important, however, was the question of Odo's succession, since its solution also determined whether the partition of the realm would be permanent or only temporary. On Odo's death, his brother Robert, who had already taken over most of Odo's *honores* in 888,346 might have had a claim to the throne too. However, Odo himself, when he had reached for the crown, had been opposed by large parts of the realm. The symbolic capital of his Viking victories legitimised his new position, but in the end it had been his Neustrian *honores* which provided him with the necessary resources to impose his claim. Robert had no such legitimation and we should not assume that, just because Odo had


become king that he had created a new dynasty—Robert's claim as his brother's heir was thus likely to be challenged. In the end, it would have been up to the nobles to decide upon the issue,347 many of whom either followed their own interests like Richard the Justiciar and William the Pious, or supported Charles. Settling the question of succession in favour of Charles, whose claim was undeniable, was thus a pragmatic course of action, avoiding future conflicts.

When Odo died a couple of months later, this was the starting point for Charles' sole reign. In the absence of other Carolingian rulers, Charles' blood had become the crucial factor for his final elevation to the throne. It gave him a claim that was impossible to deny and thus distinguished him from other, non-Carolingian, candidates and claimants. In the end, it paved his way to the throne of Western Francia and was the basis for his recognition by his peers and the pope. Within the realm, the network around Archbishop Fulk and Count Heribert348 formed his old and future power base. This group, however, while cohesive and powerful, had proven too weak to be the sole pillar of his rule in times of conflict. During the time of the fight with Odo, the network had depended on alliances with other powerful nobles like Richard the Justiciar and Baldwin of Flanders. To integrate them under his rule would be the crucial task of the next years of Charles' reign.

<sup>347</sup> On 9th and early 10th century royal successions see Becher, Dynastie. Becher argues that successions were not simply a question of dynastic rights, but much more of the dominating influence of the nobles. The right of inheritance was used as an argument, but the final decision was to be made by the aristocrats (Becher, Dynastie, 198). Becher did not include the Robertians in his analysis; however, the situation in 892/893 ties in very well with his conclusions.

<sup>348</sup> Odo had reconciled Heribert with Charles. Annales Vedastini 897, 79.

## **II. Changes in the political landscape: From Louis the Stammerer to Odo**

Belonging to the inner circle around the ruler meant having influence on the political decisions taken there as well as being permitted a certain amount of control over access to the king and, thus, over the political agenda. This influence increased if a ruler's capacity to exert control over the political affairs of the realm was diminished, for example when he was too young to wield actual power or at moments when the old ruler had died and his succession was open to debate. The period of roughly 20 years after the death of Charles the Bald until Charles the Simple's accession as sole king of the Western realm was one of five very different rulers quickly succeeding each other. Each succession implicated alterations in the framework determining the relations between the ruler and the nobles around him. The ascents of new rulers to the throne were not the only factors causing shifts in the political landscape however. Also, the composition of the networks of nobles around the rulers changed. Rivalries opened up and were bridged, nobles once influential died or had to leave the court while others rose.

Hence, determining the possibilities and limits of royal power necessitates the exploration of these networks. Who formed part of these networks and how could each individual's position in these networks be described? As in the previous chapter, this approach reveals how the individual nobles interacted not only with the king, but also with each other, at times forming alliances and factions to influence the politics of the realm. These different interactions characterise two very different groups which should not be confused: depending on their proximity to the ruler; one formed layers around him which often consisted of nobles who were rivals with each other, the other aimed to increase influence on the ruler by allying itself to other nobles sharing the same interests and could stretch over the different layers. These groups could certainly be identical and the degree of congruency between them was crucial for the amount of control groups of nobles could win over the royal politics.

Analysing networks, of course, presents a number of problems. Sources for the period covered in this chapter are richer than during the 25 years that were to follow, yet are still far from abundant and allow us only small glimpses of the proceedings at court. Consequently, our results will always only partially reveal the actual networks. The nature of one of our major sources, the royal diplomas, poses another problem. As our sources indicate, many more diplomas seem to have been issued than have actually survived to the present, a loss that increases the problems of our analysis. In addition, while coincidence certainly played a role in regard to which diplomas have been preserved (and which have not), the chance of survival was much better in ecclesiastical institutions.1 This leads to a

1 Esch, Überlieferungs-Chance; Johanek, Herrscherdiplom; Merta, Laien. Koziol, Politics, 307 argues that chance played much less of a role than has generally been assumed so far, due to certain distortion of the overall image in favour of spiritual dignitaries within the royal networks. Furthermore, we need to consider how royal charters should be read. The study of royal diplomatic is a field that has seen massive change since the 19th century. They are no longer seen as mere remnants of day-to-day rule, a kind of bureaucratic paperwork, so to speak.2 Instead, their issuance is now considered as something special—to receive one was meant to have been, in the true meaning of the word, "a privilege."3 Expressions were carefully selected for their political meaning, its layout equally carefully chosen to transmit royal dignity and authority, while the handing over of the physical document was embedded in a well thought-out public ceremony. Diplomas, therefore, not only served to secure legal rights, but are also reflections of royal authority in the symbolic communication between the king and his nobles.4 This becomes even more true in the late Carolingian period as the spatial distribution of the recipients and temporal distribution of the issuing of diplomas became more and more concentrated around certain regions and moments. These concentrations are by no means random and reflect real political developments and events.5 In this context, royal diplomas gain another dimension of meaning: they are instruments of politics, issued at special moments. They are symbols of the political relations between the king and certain individuals, notably the petitioners and the recipients. Geoffrey Koziol recently argued that these diplomas should be considered as "performatives," a kind of memorial, reminders of already existing or newly forged alliances, of peace treaties, confirmations of honours or claims to another kingdom.6

However, overgeneralising this view comes with the danger of placing too much emphasis on the relations expressed by an individual diploma. While the intervention of an important noble for an ecclesiastical institution certainly reflects the importance of this noble and his relation to the king, the institution itself did not necessarily have a special bond connecting it to the ruler. Often, diplomas were also sought for more basic reasons, to have rights confirmed or possessions added.7 When using the evidence of the diplomas, therefore, we have to ask in each case what the nature of the relations between the nobles asking for them and the kings delivering them were. Cross-referencing diplomas with other sources helps to establish the importance of the individuals appearing in them and can

the importance of diplomas, ecclesiastical institutions had a special interest in preserving them. Critical Merta, Auctoritae, 300, n. 6 and Irmgard Fees in her forthcoming article.


aid in casting some further light on these relations. Titles and epithets used to describe individual nobles in royal charters point towards the special importance of certain individuals, as in the case of the word "beloved" (*dilectus*) or in the use of the superlative.8 These words do not always have to be taken literally, as a true indication of personal closeness, since they may equally be used to express the rank a noble took within the hierarchy at court. For us, however, this distinction does not matter too much, since even if the nature of their influence on the king was different, it was visible and real.

#### **II.1 Old elites: Louis IIthe Stammerer**

On the verge of leaving for his second Italian campaign, Charles the Bald held an assembly at Quierzy during which the famous capitulary of 877 was issued. The main purpose of the measures outlined in the capitulary were to constrain Charles' son, Louis the Stammerer, 9 who was installed as king, yet at the same time surrounded with Charles' *fideles* and other nobles of his choosing.10 Thus, for example, the count of the palace, Adalard, was entrusted with the royal seal and ordered to remain with the prince at all times.11 Adalard, given his office, was a politically influential man at court, whose importance is further emphasised by him being, at Charles' instigation, the prince's new father-in-law. He was not the only one mentioned by the capitulary, however. A large number of nobles were commanded to stay close to the prince at all times, while others were to join him should he move to certain regions.12 Three bishops were charged with keeping Charles informed about the state of the realm and his son's activities.13 Given the importance of these tasks, we can assume the nobles mentioned within the capitulary, like Adalard, represented a group possessing both considerable political importance during the emperor's last years and proximity to him. Therefore, the capitulary offers us an insight into the circle of nobles present at Charles' court during the last years of his reign and we use it extensively hereafter to point out certain continuities within the circles of leading nobles during the following reigns.

However, the system created to keep Louis in check appears to have been flawed. Some of the most important nobles, such as Hugh the Abbot, Count Boso of Vienne and Count Bernard of Auvergne, had not been present at the assembly of Quierzy.14 Hugh, figuring prominently in the emperor's last diplomas,15 held


the military command against the Northmen on the Loire.16 Even more important was Boso, 17 whose sister had become first Charles' concubine and then his wife.18 After Charles' first Italian campaign, he had been made the equivalent of a viceking in Italy, married the daughter of Louis II19 and from that point on intervened frequently in imperial charters with epithets and titles emphasising his proximity to the emperor. 20 Finally, Bernard of Auvergne was, next to Bernard of Gothia, one of the most powerful men in Aquitaine. Both of them had been assigned to Louis the Stammerer in 872, when he became king of Aquitaine.21 The capitulary which separated the nobles into different groups, of which one was to remain at all times with Louis and hence possessed significantly more influence on the affairs of the realm than the others, therefore might be expected to have created rifts among the nobles. And indeed, the new order did not hold for long. Charles, having arrived in Tortona and waiting for Hugh the Abbot, Boso, Bernard of Auvergne and Bernard of Gothia, who Charles had ordered to follow him, received the news that they had joined a conspiracy against him.22 None of these men had been assigned to remain with Louis at all times. Consequently, Janet Nelson proposes that their refusal to come to Charles' aid was meant to draw him back over the Alps and re-establish the status quo. According to this argument, the problem was not their proximity to Charles himself, but their position at Louis' court should the emperor die in Italy.23

This fear was not without reason. When Charles died upon his forced return to the North, the first thing Louis tried to do was to create his own power base by handing out *honores* to other nobles,24 thus provoking resistance from those who felt cast aside by him. A letter from Hincmar to the new king reveals the leaders of the opposition that now assembled at Montaimé: Hugh the Abbot, Boso, Bernard of Auvergne and Bernard of Gothia, who were now joined by two of Charles' key personnel, Abbot Gauzlin and Count Conrad of Paris.25 Gauzlin, abbot of Jumièges, Saint-Amand and Saint-Germain-des-Prés, had served as the emperor's archchancellor since 86726 and had probably also drawn up the


capitulary of Quierzy.27 Conrad had acted as Charles' legate in negotiations with the Northmen,28 although his importance is probably best captured by the epithets used to describe him in a diploma issued in the wake of the battle of Andernach. He was "our *fidelis*, associated with us by family propinquity, Conrad, most noble count" as well as "our dearest and most familiar."29 Interesting enough, these two apparently had no connection with the Italian affair. Quite the contrary, Gauzlin and Conrad had both belonged to the group originally intended to stay close to Louis. Therefore, with one stroke, the new king managed to upset not only those already in fear of losing their influence at court once the emperor was dead, but also those who were meant to stay close to him. Both groups accordingly united against him and forced him to give in. Louis was crowned king but had to agree to hand over even more *honores*30 and to give a *promissio* about his future conduct to the nobles.31 In the end, it was the nobles who made Louis king, not the designation by his father. 32

Hincmar's account in the *Annales Bertiniani* is full of references to leading nobles and councillors urging and advising the king. For example, according to him, Hugh the Abbot urged Louis to cross the Seine and come to his aid against the Northmen as well as against the counts Gauzfrid and Emeno. 33 Soon afterwards, the king was reconciled with Gauzfrid at the instigation of some of his councillors.34 Assessing these references however is rather problematic. Hincmar's annals were by no means an objective account of what had happened. The archbishop had a distinctive political position and phrased the annals accordingly, manipulating documentary evidence and omitting facts if it helped to further his case.35 Especially during his later years, he emphasised the necessity of consensus between the king and the nobles—no king should rule alone but be advised and counselled by chosen members of the nobility. The clearest evidence of his conviction is of course his *De ordine palatii*, but he expressed his thoughts also in his letters: "'The


general disposition of the realm' must depend, not on any one man, but on 'the judgement and consent of many'."36 Hincmar's references in the annals are therefore hardly surprising: they reflect the same thought pattern and were meant to show how kingship should work.

However, the question remains, whether the archbishop's account represents the actual relations between Louis and the nobles surrounding him. For example, Hincmar tells us that, while treating with Pope John at Troyes, Louis ordered the bishops Frotar and Adalgarius to ask John to confirm his father's diploma that handed the realm over to him. In turn, the pope asked for the confirmation of another diploma that granted Saint-Denis to the Church of Rome. According to Hincmar, this was a forgery produced by the very same bishops and other councillors of the king,37 which would indicate that the nobles around Louis were behind the initial request of the king in order to provide the pope with an opportunity to present his own charter. Equally, a visit of Louis to the pope was made at the instigation of some of the councillors, probably leading to the excommunication of Emeno. 38 When Louis met with Boso after the synod, he was in the company of several of his most important councillors, resulting in the betrothal of his son Carloman to Boso's daughter.39 Furthermore, the distribution of the *honores* of Bernard of Gothia is said to have been made with their counsel.40 In all of these cases, Louis is shown to have acted in close cooperation with those around him. As argued, this concords well with Hincmar's own convictions of how a kingdom should be run and does, in all probability, reflect the reality of Louis' court.41 There is, however, a second layer to this. In all of these cases Louis' actions directly reflect the interests of those around him, the most obvious case being the conspiracy with the forged diploma, meant to deprive Gauzlin of Saint-Denis. However, should we take this as a sign of the influence the nobles exercised over Louis? Depriving Bernard of Gothia of his *honores* certainly intensified the tensions between the king and the count and might not have been in Louis' best interest if he was to seek a peaceful solution.42 Yet at the same time, distributing Bernard's *honores* among those around him also served his purpose since it reinforced his relations with


the same men who were to take up their arms to subdue the count.43 Similarly, the betrothal between Louis' son Carloman and Boso's daughter tied the latter closer to him. Nevertheless, taking into considering the way Louis' reign had started, how these nobles had enforced their will on the new king and how the king's decisions concur with their own interests, we can deduce that they still possessed a distinctive political leverage over him.

Who were these men exercising such influence on Louis? Hincmar's alreadymentioned letter is a good starting point for an analysis of this inner circle. Of the six nobles he named as most important, three also appear in the king's diplomas. Most notable is Hugh the Abbot, who requested three diplomas for his abbeys, and was named "glorious abbot" and "our relative"; while another three charters, in which he did not appear, were issued for monasteries under his control.44 Considering that he held the command along the Loire, a diploma for the cathedral church of Tours may also have been issued at his request,45 although due to the lack of further evidence this must remain a mere possibility. Boso, apart from the betrothal of his daughter to Louis' son Carloman, intervenes in two charters, described as "our beloved *dux*" and "our dearest *dux*" respectively, for the church of Lyon and the abbey of Tournus under abbot Geilo. 46 A lost diploma for Saint-Martin of Autun may also have been related to him.47 Gauzlin, who served as archchancellor for most of the reign,48 intervened together with Queen Adelaide in a royal charter for the church of Paris, while the abbey of Saint-Denis, which came under his control at the beginning of Louis' reign, received another diploma.49 Given the offices they held at court, their appearances in Hincmar's annals and the epithets used to describe them in the royal diplomas, we can already tell that these men held important positions at court. The same was probably also true for Bernard of Auvergne since he was not only amongst those receiving the *honores* of Bernard of Gothia, 50 but was also charged with protecting Louis' son and successor, Louis, when the king felt his death approaching.51 Bernard of Gothia, in turn,



<sup>45</sup> DLS 39.

quickly lost his *honores* and was eliminated from the inner circle. Finally, during Louis' reign Count Conrad disappears from the sources, thus making it impossible for us to draw any further conclusions.

Bernard's loss of his *honores* points us towards a rift among the leading nobles that emerged almost immediately after Louis' accession to the throne. When Hugh the Abbot called on the king to cross the Seine, he did so as a consequence of two conflicts:52 one with Gauzfrid, Gauzlin's brother;53 and another with Emeno, the brother of Bernard of Gothia. 54 Thus, in both conflicts, the rebellious nobles were closely related to men who had, up until then, been key figures in the political affairs of the realm. Gauzlin was able to reconcile his brother with the king,55 something that Emeno might also have hoped for from his own brother. Bernard, however, had entered into conflict with Archbishop Frotar of Bourges56 around the same time and both he and his brother were excommunicated at the synod of Troyes in 878,57 at the same moment that Bernard lost his *honores.*58 Soon after, he also became the target of a military campaign.59 Thus, Bernard had clearly been eliminated from the circle around the king. Gauzlin's influence was also diminishing. At Troyes he became the target of a conspiracy to deprive him of Saint-Denis, 60 followed by the loss of the office of archchancellor in early 879.61 Linking of the cases of Bernard and Gauzlin has been proposed since both of them belonged to the Rorgonid family.62 While this link by kinship should not be overinterpreted in regard to both of them pursuing the same goals,63 the family ties between the two did play an important role as a means to mediate a solution to the conflict.64 Gauzlin himself was apparently not directly involved in either of the cases, since he was able to use his influence at court to mediate a favourable outcome for his brother. This he failed to do for Bernard, probably because at this point his position at court had already come under pressure. In any case, two parties appear to have formed, struggling for influence at court: Hugh the Abbot, Boso and Bernard of Auvergne on the one hand, and at least Gauzlin and Count Conrad on the other.

While the group around Gauzlin remains in the shadows, the one around Hugh is more visible due to its presence at court and hence we can adduce a number of the other nobles that were part of it. Next to Hugh, its most important member


certainly was Count Theoderic the Chamberlain, undoubtedly the Theoderic who was named in the capitulary of Quierzy as part of the group to remain with the prince. His importance at court becomes clear from a number of cases: like Bernard of Auvergne, he also profited from the distribution of Bernard of Gothia's *honores;*65 on at least one occasion he demanded a royal diploma;66 and finally, he belonged to those accompanying Louis' heir while conducting the campaign against Bernard. 67 Considering his affiliation with the group around Hugh, we must, nonetheless, be careful. While the sources do indeed indicate his cooperation with this group in the case of Bernard of Gothia and he certainly joined forces with them from that moment on, his earlier importance at court may also have been rooted in a personal relationship with Louis himself. In other nobles' cases, the connection with the dominating circle is more apparent, although their political importance appears to have been of a more limited nature. For example, Archbishop Ansegisus of Sens figured as the leader among the bishops paying homage to Louis the Stammerer after his coronation,68 indicating that he kept some of the influence he had gained during the last years of Charles the Bald, when he, Odo of Beauvais and Actard of Tours replaced Archbishop Hincmar of Reims at court.69 However, his association with the group around Hugh is only noted after Louis' death, when he crowned Louis III and Carloman II at Ferrières on their behalf.70 Archbishop Frotar of Bourges, who had been the target of Bernard of Gothia, together with Bishop Adalgarius of Autun participated in the conspiracy against Gauzlin at Troyes, where the two of them handed over a precept to the pope on the king's behalf.71 Adalgarius also received two diplomas for his church, one of them issued at the request of Count Theoderic, who by then had become count of Autun,72 thus not only marking his access to the king but also his connection to the inner circle around Hugh. The same was true for Archbishop Aurelian of Lyon, for whose church Boso demanded a diploma.73 Also closely connected with Boso was Abbot Geilo of Tournus, the future bishop of Langres, who received another diploma at the count's intervention.74 Finally, we can also add Bishop Walter of Orléans as well as the counts Goiram and Ansgar (probably the count of Oscheret) to the list. Walter was yet another of the nobles who had been assigned in the

65 Annales Bertiniani 878, 229−230.


capitulary of Quierzy to accompany Louis when he entered the Seine region and was now granted another royal charter,75 thus pointing to his continuing importance. More importantly, however, he and the two counts were sent to Louis the Younger to negotiate on behalf of the group around Hugh after Louis the Stammerer's death.76 Apart from Theoderic, all of these men appear to have been part of an outer layer of the dominating circle. Their influence at court remains more or less obscure since none of them appears in the direct vicinity of Louis. In the case of Ansegisus, his mention can also be explained by his importance to the Gallic Church77 while Frotar's and Adalgarius' appearances at court may as likely have been the result of the ecclesiastical proceedings of the synod of Troyes, as was probably also Aurelian's. 78 Finally, the two counts only appear after the king's death.

Louis' contacts were not limited to those belonging to the dominating groups of Hugh and Gauzlin. As we have already discussed in the context of Theoderic the Chamberlain, certain nobles also drew or might have drawn their position at court from a personal connection to the king. Examples of this mainly come from the family of Louis' second wife, Adelaide, who can be found intervening in a diploma for Saint-Médard of Soissons.79 Related to her was Count Aledramnus, who Louis granted a rich gift of property naming him "our dearest" and "our beloved relative."80 In his case, a connection with Gauzlin is also discernible which certainly did not hurt his position at court. Both were captured after the battle of Andernach81 and Aledramnus made a donation to Gauzlin's abbey of Saint-Denis in 879.82 Finally, just before Louis' death, Adelaide's brother Wulfard became the new archchancellor in the succession to Gauzlin. 83 While he was certainly appointed with the consent of the leading group around Hugh, his being chosen was undoubtedly more due to his relationship to the king: once Louis died, he appears to have lost the office again.84

Unrelated to the king but holding a position independent from Hugh and his allies were Bishop Odo of Beauvais, one of Charles the Bald's confidants who had been assigned to remain close to Louis in the capitulary of Quierzy, and a certain

81 Annales Bertiniani 876, 209.

84 Bautier, Recueil Louis II, Louis III et Carloman II, LXVII−LXVIII.

<sup>75</sup> DLS 40 (deperditum).

<sup>76</sup> Annales Bertiniani 879, 236−237; Annales Vedastini 879, 45.

<sup>77</sup> On the primacy of the church of Sens (and Ansegisus), see Schramm, König, 112−114. The importance of the archbishops of Sens within the Gallic Church becomes evident from their role in the royal coronation, a role that was heavily opposed by the Archbishop Hincmar of Reims and his successors.

<sup>78</sup> Aurelian's presence at Troyes is indicated by his subscription of the synodal acts. MGH Conc. V, N° 9, 104, 135, 140.

<sup>79</sup> DLS 30.

<sup>80</sup> DLS 28, 84: *…carissimum nobis comitem nomine Aletramnum…* and *…praefato dilecto propinquo nostro Aletramno…* On him see chapter I.2.2.

<sup>82</sup> Lot, Notes, 150.

<sup>83</sup> DLS 31, 29thMarch, 10thApril 879. On him see also chapter I.1.2.

Count Albuin. When he felt his death approaching, Louis entrusted the two of them with the royal sword, crown and the rest of the royal insignia, to be brought to his son Louis III, who was to be crowned and anointed. When they heard of the king's death, they handed over the insignia to Count Theoderic and quickly left.85 That Odo and Albuin did not remain with Theoderic seems to indicate that they kept their distance from the dominating circle. Difficult to place is Bishop Wala of Auxerre, one of the bishops assigned to advise Louis when entering the Seine region in the capitulary of Quierzy. At an unknown moment, the church of Auxerre received a diploma from the king86 which points towards a continuing connection between the royal court and the bishop. Yet, whether this connection was determined by the dominating groups around the king or was independent from them remains unclear. Once Wala had died, Wibald, a cleric who had been educated at the palace school of Charles the Bald, succeeded him at Louis' instigation.87 After the king's death, from one of his successors he obtained a royal charter concerning a private donation he made to the congregation of his church. In this diploma he established a prayer service for Louis commemorating his death day,88 which may be read as a sign of a closer connection between him and the king. In the context of episcopal succession, another see also needs to be considered. Early in 878, Bishop Willebertus of Châlons died and was succeeded by a certain Berno. 89 It seems likely that Berno was well connected at the royal court since Hincmar asked him in 879 to intervene on his behalf before the kings and Hugh the Abbot90 and since he later received diplomas both from Carloman II and Charles the Fat. 91 Had Berno, like other bishops of the time, previously belonged to the royal chapel? In 877, Charles the Bald issued a diploma for the abbey of Marchiennes at the request of "Berno, the venerable deacon of the palace, our beloved *ministerialis.*"92 It seems at least possible that the two Bernos were identical and that Louis managed to lift another cleric from the palace on an episcopal see. Whether in both of these cases this was entirely his own choice or whether the circle around Hugh had had a hand in their appointments has to remain unanswered.


Finally, there is Archbishop Hincmar of Reims himself to consider. During Charles the Bald's last years, he had fallen out of royal favour and, as we have seen, been replaced by a number of other men.93 After the emperor's death, Hincmar addressed a number of letters to Louis, offering his counsel.94 When Louis' efforts to create his own power base by handing out *honores* was resisted by the leading nobles, Hincmar advised him to refer himself to their leaders.95 This advice was less a sign of support for the rebels than a simple acknowledgement of political realities96 as well as possibly an attempt to gain a foothold within the group of dominant nobles by siding with them. The archbishop's importance was underlined when he consecrated and crowned Louis king.97 He continued to send Louis more letters on political affairs and also played an important role at the synod of Troyes.98 Yet despite all this, his actual influence at court remained limited and he appears not to have belonged to Louis' inner circle. All in all, he was someone who carried some political weight but was never in the actual centre of political decision making, although cooperating with the king and those close to him.99

So far, we have been able to identify different groups of nobles at the king's court. The groups around Hugh the Abbot and, to a lesser degree, Gauzlin appear to have dominated the inner circle around the king, using their influence to further the interests of their associates. However, another group can also be identified—although in this case this term may be misleading since there are no signs of an inner cohesion or even cooperation—which had independent access to the king and probably, at least in the case of his close relatives, like his wife Adelaide, also exerted some political influence. The existence of this last group demonstrates that the king was able to maintain a certain liberty in choosing those close to him and was not entirely dependent on the groups dominating the court. Royal contacts went even further, encompassing large parts of the realm. Shortly after his coronation, Bishop Arnaldus of Toul100 and soon thereafter Abbot Ansbald of Prüm101 received diplomas, representing royal contacts with Lotharingia. From Burgundy came Abbot Odo of Vézelay,102 from Aquitaine Abbess Ava of Sainte-Croix103 and the abbey of Solignac104 also received a royal charter. The synod of Troyes presented the chance for a large number of ecclesiastic dignitaries to


demand confirmations and grants. Among them were not only the bishops Abbo of Nevers105 and Lambert of Mâcon106 from Burgundy but also Archbishop Sigebod of Narbonne107 and his suffragan Frodoynus of Barcelona108 from Septimania and the Spanish March, while other diplomas went to the church of Girona109 and the abbeys Notre-Dame of Arles-sur-Tech110 and San-Esteban de Banolas.111 However, these were contacts that can easily be explained by regular day-to-day work related to the administration of the realm and therefore they do not betray any larger political influence of their recipients at the royal court.

#### **II.2 Rival factions: Louis III and Carloman II**

After the death of Louis the Stammerer, the tensions between the groups around Hugh the Abbot and Gauzlin rose and turned into open conflict. Since both of Louis' sons were under Hugh's control, Gauzlin decided to call Louis the Younger into the realm. At the same time, the composition of the two groups changed. In the case of Gauzlin, the contact with Bernard of Gothia—if it had ever been strong—now appears to have ceased completely. Instead, until the division of the realm at Amiens, the sources emphasise Gauzlin's cooperation with Count Conrad of Paris.112 By early 880, their base of support amongst the nobility appears to have weakened considerably.113 Meanwhile, the alliance around Hugh also began to crumble in the wake of the succession crisis. Early on the group still held together: Hugh mediated an exchange of *honores* between Boso and Theoderic, 114 while Hincmar also mentions the three of them acting together in the face of Louis the Younger's intervention.115 However, Boso broke away from them soon after, having himself crowned king at Mantaille on the 15th October 879, and taking with him a number of nobles belonging to the wider circle. Boso's *electio* is signed, among others, by Archbishop Aurelian of Lyon, Bishop Adalgarius of Autun und Abbot Geilo, 116 two of whom we have already identified earlier as having a close


<sup>105</sup> DLS 18.

connection with him. Adalgarius acted as archchancellor for Boso, 117 Geilo was granted a diploma shortly after118 and also became bishop of Langres in early 880.119 Furthermore, Bernard of Auvergne's position in relation to Boso is uncertain. In a charter issued by Boso "after the death of the most glorious King Louis," Bernard appears as one of the witnesses,120 but soon after he joined the fight against Boso. 121 In the light of this, it is very possible that the final settlement of Ribemont and the partition of Amiens was necessitated not only by the intervention of Louis the Younger, but also by the diminishing support both sides received from the nobility. Louis III, from then on with Gauzlin, was to rule over Francia and Neustria; Carloman, remaining with Hugh, received Burgundy and Aquitaine.122

The actual influence each of these groups had on the kings, before and after the partition, is hard to assess. Once crowned, both Hincmar and the *Annales Vedastini* emphasise the independent actions of both Louis and Carloman. 123 Yet once we turn from the narrative sources towards Hincmar's letters, the image changes, as Thilo Offergeld has shown.124 Just after the death of Louis the Stammerer, in a letter to Count Theoderic the Chamberlain, the archbishop argued for a broader basis to support the regency of the late king's sons.125 Moreover, two other letters reveal his idea of involving Charles the Fat in their tutelage. The first letter, addressed to Charles, describes Louis and Carloman as "orphan boys without a father" in need of a "mature, prudent and rational guardian."126 The second letter, prefacing the first, was addressed to Hugh the Abbot, trying to win his support for the plan.127 Hincmar, of course, aimed at regaining his lost influence at court and may therefore have exaggerated the young kings' dependence. Yet, the way he passes over them in these letters, disregards them as underage boys completely dependent on those around them, as children who still needed to be educated, seems nevertheless to reflect the common opinion. Otherwise his arguments would hardly have


been convincing. Thus, the letters are a strong indicator that, at least in 879/880,128 Louis and Carloman were entirely in the hands of those nobles around them.

Hincmar's own weak position at court put him in a difficult position. He had crowned Louis the Stammerer only two years earlier, yet in 879 it was the Archbishop of Sens who had consecrated Louis and Carloman at Ferrières. As Hincmar's letters to Theoderic and Hugh show, he did his best to influence royal politics, yet whether his efforts actually produced any result seems doubtful.129 The conflict surrounding the episcopal see at Noyon provides us with further insight. After the death of Bishop Raginelm, Hincmar, aiming at limiting royal influence to the confirmation of the final candidate, requested a free canonical election.130 Hincmar did not appear at court in person, but, apart from writing to the kings131 and Hugh the Abbot,132 he also repeatedly asked Bishop Berno of Châlons to intervene on his behalf.133 In the end, Hincmar had his way and Heitilo was elected according to the canons, as the archbishop had pressed for.134 It is interesting to note that the archbishop did not himself address the kings or Hugh the Abbot, but had one of his suffragan bishops mediate for him. Both letters sent to Berno indicate that the bishop was present at the royal court at the time. Furthermore, we can assume that Hincmar's choice in using Berno was not only because the bishop happened to be present at court, but because the archbishop considered him as a suitable intermediary. Following this thought, it seems plausible that Berno had certain connections at court that would allow him to better influence the decision makers in Hincmar's interest than Hincmar himself would have been able to at that time. If, as we have proposed above, Berno had been a member of the royal chapel before his appointment as bishop of Châlons, this would serve as an explanation as to why this was the case. In any case, his connections to the leading members of the court appear to have been better than the archbishop's, who therefore chose to involve him in the matter of the see of Noyon.

#### **Louis III**

Sources for Louis III's short reign are rare since hardly any diplomas have survived to the present day. When Louis moved south to suppress Boso's revolt, Gauzlin was charged with the command against the newly returned Vikings in Francia;135 he also became the king's new archchancellor.136 His importance as one of the leading magnates thus remained unbroken. Of Count Conrad nothing further is


<sup>128</sup> On the dating of the letters see, Hlawitschka, Lotharingien, 234−235, n. 34 and Offergeld, Reges pueri, 385−386.

known up until his death towards the end of Louis' reign.137 Either his position remained unchanged or he was replaced by a new face, another Count Theoderic, Theoderic of Vermandois. 138 Theoderic was the brother of the well-connected Count Aledramnus who we have identified as an important figure in the court of Louis the Stammerer. He may have owed Aledramnus his entrance into the circle dominating the court. Since he would not have been able to achieve such success at court had he opposed Gauzlin, he must have started to cooperate with him at a very early stage. It was probably also through his own and Aledramnus' influence, in connection with Gauzlin's, that the Robertian Odo, son of Robert the Strong and maybe already Aledramnus' son-in-law, was appointed count of Paris following Count Conrad's death.139 In the context of the quarrel between Louis III and Hincmar concerning the succession of Bishop Odo at Beauvais, the archbishop named Theoderic as one of the king's most important councillors, which indicates his ascent had occurred already before Conrad's death.140 Finally, in 882, when nobles from Lotharingia offered Louis the realm, the king turned down their offer but sent Theoderic with an army to aid them against the Northmen.141 Thus, the leading nobles at Louis' court appear to have been Gauzlin, Theoderic and possibly Conrad until his death.

At first, Hincmar appears to have cooperated with Louis and this group of nobles, but relations began to deteriorate, as is demonstrated by the conflict over the succession of the deceased Odo of Beauvais. At Hincmar's request, Louis granted the people and clergy of Beauvais the right to elect their bishop; later, again in accordance with the archbishop, he refused their candidate. At that point, Hincmar's and Louis' positions clashed. While the king insisted on his right to name the candidate—he opted for Audacher, notary in the royal chancellery142—the archbishop did his best to institute a new solution.143 He wanted to install a council of five bishops, chosen by himself, to decide upon the new bishop. Hincmar supported a certain Roger while Audacher was excommunicated. In the end, the conflict was only ended by Hincmar's and Roger's deaths.144 The archbishop certainly


<sup>137</sup> 22nd March 882. Chaume, Origines, 310 and Dümmler, Geschichte III, 133.

did not take his exclusion from an influential position at court well, and appears to have had a private feud with Gauzlin. 145 This explains why one of Louis' greatest successes, his victory against the Northmen at Saucourt, almost appears as a defeat in Hincmar's annals.146

There is only one more noble left to mention in Louis' network: Welf, abbot of Saint-Riquier, who received a diploma probably during the preparations for the campaign leading up to Saucourt.147 Welf, "venerable abbot and our dearest kinsman", also held Sainte-Colombe and had been assigned in the capitulary of Quierzy to constantly remain with Louis the Stammerer. He thus belonged to the important nobles of the realm—and he was Count Conrad's brother.148 He was related to Louis through Welf 's and Hugh's aunt, the Empress Judith, second wife of Louis the Pious and mother of Charles the Bald. His presence in the diploma is therefore unsurprising. He was not only related to the king, but also, and more importantly, to one of the nobles dominating the royal inner circle. His influence, however, did not last long. Like his brother, he died before Louis, in his case 13th November 881.149

#### **Carloman II**

Much better documented than the reign of Louis III is that of his brother, Carloman II. Carloman's years as ruler of the southern division of the realm saw the continuation of the alliance around Hugh the Abbot and Theoderic the Chamberlain. Hugh's dominant role becomes clear from the *Annales Vedastini*, which names him as the mediator of the treaty of Ribemont.150 Two of Carloman's royal diplomas saw him intervening, as "our dearest venerable abbot", for the church of Narbonne and simply as "venerable abbot" for the king's chaplain Arveus, while in a third he served as *ambasciator* for the abbey of Saint-Florent. 151 Theoderic appears in the same role in a diploma for the church of Autun152 and was probably also behind a diploma for Saint-Martin of Autun.153 The wider circle of their supporters also remains visible: Archbishop Frotar of Bourges intervened on behalf of

145 Werner, Gauzlin, esp. 447−450.


the abbey of Beaulieu154 and properties in Berry and the Nivernais were restored to Bishop Walter of Orléans. 155 Two other figures joined the royal circle: one was Abbot Wulfard of Flavigny, Adelaide's brother, who again became archchancellor.156 He intervened alone for the *fidelis* Rainardus157 and together with Hugh the Abbot for the chaplain Arveus in another.158 While the first diploma seems to indicate a close connection to Carloman, his return to the office of archchancellor was certainly more due to an alliance he concluded with Hugh and his circle,159 as seems to be indicated by their joint intervention. This last diploma also figures the other remaining important noble: Wido, "our beloved count"160, who completed the list of those intervening for the chaplain. This Wido may have been either the brother of Count Ansgar of Oscheret, who we have already met as an associate of Hugh; or another Wido from Burgundy who later appears to have been close to Richard the Justiciar. The first Wido followed his brother and Wido of Spoleto to Italy in 888, where he died in battle.161 The second accompanied Richard at the assembly of Varennes162 and also witnessed two of his private charters.163 Either way, the epithets of the diploma and his intervention with Hugh and Wulfard indicate Wido's political importance and his link to the circle around Hugh.

Apart from these remarks, the rich corpus of diplomas preserved from Carloman's reign allows us to make some further observations. As during his father's reign, Carloman was sought out by a number of ecclesiastical dignitaries from Septimania and the Spanish march. Archbishop Sigebod of Narbonne and Bishop Theotarius of Girona again received charters for their churches,164 as did the abbey of Notre-Dame of Arles-sur-Tech. 165 In addition, Saint-Polycarpe and Sainte-Cécile were now granted diplomas.166 Similarly, Bishop Abbo of Nevers and Abbot Odo of Vézelay again appeared at the royal court:167 along with Archbishop Frotar's intervention, further examples of the king's contacts in Aquitaine. Also interesting are a number of appearances closely connected to the advancement of the campaign against Boso: Bernard of Auvergne, Boso's brother Richard and bishops Adalgarius of Autun, Geilo of Langres and Ratbert of Valence, all of whom had


<sup>154</sup> DCmII 62.

supported Boso when he took the crown.168 Bernard received the county of Mâcon the moment it was taken from Boso, 169 thus indicating that, if he had indeed supported Boso, he now had switched to the Carolingian alliance that had formed against the usurper. His affiliation with the group around Hugh is less clear, although his son Norbert became Carloman's new notary.170 Over the following year, charters from Saint-Julien of Brioude, which, although he never was its abbot, appears to have been under his control,171 were dated after Charles the Fat instead of Carloman. 172 Richard and Adalgarius acted together in a diploma for the church of Autun, Carloman's first preserved charter, dating to 30th November 880.173 In the same diploma, Theoderic the Chamberlain served as *ambasciator*, thus linking together the three most important men of the area. Richard, addressed as count of Autun, was now either granted the county or at least recognised as its count, Adalgarius was its bishop and Theoderic held important abbeys in the area.174 Next in line was Ratbert of Valence, who joined Carloman's court between November 880 and July 881,175 followed by Geilo of Langres.176 Thus, Boso's own network is revealed to have been extremely fragile, with key figures such as his own brother and his archchancellors deserting him as soon as the united power of the Carolingian kings became apparent. However, not all of those who were now reaccepted into Carloman's peace remained close to the king. In particular, Richard's position is far from clear. As Hincmar reports, the new count of Autun brought Boso's wife and daughter to Autun after Vienne had fallen,177 yet nothing indicates explicitly that it was he who captured the city.178 Moreover, that he brought his brother's family to Autun is not a sure indication that he supported Carloman wholeheartedly. He may also have done so to keep them out of the reach of the royal court after Boso's death.179 Be that as it may, for the next few years Richard disappears from our sources, indicating that he did not play an important part in the affairs of the realm. However, his appearance in the royal diplomas, like those of the other nobles from Burgundy originally supporting Boso, marks the reintegration


of Boso's old powerbase into Carloman's rule. At least in the Autunois, Theoderic the Chamberlain played a key role in these proceedings and served as a direct link to the dominant group around Hugh the Abbot.

#### **The realm reunited**

Once Louis III had died, Carloman also took over the Northern kingdom at the invitation of the nobles from Francia.180 Hincmar's use of the expression "great men of the realm" (*primores regni*) in this context points towards a unified action of the leading magnates, which leaves us to conclude that Gauzlin, as well as Theoderic of Vermandois, were involved in the issue. In fact, Gauzlin does appear in the petition of the bishops to Carloman made at Quierzy181 where the young king renewed the *promissio* he and his brother had given two years earlier. Undoubtedly, both groups had made an arrangement to set aside their rivalry and embark on a course of cooperation. However, Hincmar also tells us that Carloman, after an initial campaign against the Northmen, was not able to muster sufficient forces to fight them due to the absence of Hugh the Abbot and the withdrawal of several powerful nobles of Francia from the new king.182 Shortly after Carloman had been called to the North, the old frictions appear to have surfaced again. This does not mean however that the new king found himself completely isolated in Francia. Hincmar himself, for example, was among those who continued to support Carloman. 183 This is hardly surprising: as soon as Louis died, the archbishop had once again renewed his efforts to gain influence at court, composing the famous *De ordine palatii* for Carloman. 184 However, his death shortly afterwards deprived him of any possible profit.

This period, marked by the absence of nobles from Francia from Carloman's court, lasted for over a year. Efforts were certainly made to overcome this situation, as the succession of Fulk, abbot of Saint-Bertin and member of the royal court since the times of Charles the Bald, 185 to Hincmar at Reims demonstrates. Being well connected both at court and among the Frankish nobility, he might have been installed as a mediator between the king and the withdrawn nobles. Yet, the integration of the North only succeeded after Gauzlin had been made


archchancellor around August 883, an office that had remained vacant since the death of Wulfard almost two years earlier.186 His appointment certainly helped to ease tensions and create links to other nobles, thus providing the court with an influential lever. The final agreement seems to have been reached at the assembly of Ver about half a year later. Carloman's capitulary,187 a renewal of the one he had issued in February 883,188 placed the king in line with his predecessors and emphasised his claim to rule over the entire realm.189 In the wake of the assembly, he issued a diploma at Compiègne, the most important Carolingian palace of the Western realm, at the request of three of the North's leading nobles, Bishop Berno of Châlons, Bishop Angelwinus of Paris and Count Theoderic of Vermandois, 190 demonstrating the newly won support of Francia for the king. That Hugh's party now agreed on allowing other nobles into the circle around the king was undoubtedly also due to the threat posed by the Northmen. Before the assembly of Ver, the leading nobles had already gathered at Compiègne, where they decided to open negotiations with the Vikings. According to the *Annales Vedastini*, they finally settled on a tribute of 12.000 pounds of silver, a sum to be collected from the entire realm.191 Participation in the decision making at court was undoubtedly the price the nobles from Francia asked for contributing their part of the tribute.

The takeover of the Northern realm does not seem to have weakened Hugh the Abbot's position at court. In November 882, it was he who went to meet Charles the Fat at Worms to negotiate the return of parts of Lotharingia to Carloman, a fruitless endeavour.192 His dominance in the royal diplomas—after the unification of the realms Hugh intervened six times193 and received another two diplomas for himself194—has led to the belief that he wielded more or less absolute control over the royal chancellery.195 This seems to be an exaggeration, since most of these charters were issued after Gauzlin had become archchancellor. However, according to the Annals of Sainte-Colombe of Sens, he held the *monarchiam clericatus in* 


*palatio,*196 a phrasing that more or less applies to the office of archchaplain.197 Setting offices or control over the royal chancellery aside, Hugh's importance during this period is probably best captured by two royal diplomas, one calling him "our tutor and great protector of our realm"198, the other installing a prayer service at Rethondes not only for Carloman's brother Louis, but also for the "most venerable" and "most reverend" Abbot Hugh: 199 a singular honour during Carloman's reign. Thus, his role at court appears to have been more central than ever.

Despite his importance, we should not forget that other men also held influential positions at court. We have already mentioned Gauzlin's return to the office of archchancellor in August 883 and can add to this a diploma for his abbey of Saint-Germain-des-Prés200 as well as him serving as *ambasciator* in another royal charter.201 In addition, Theoderic of Vermandois entered into very close association with the king after the settlement with the Northern nobles. It was by no means a coincidence that he was among those appearing in the first diploma after the assembly of Ver, honoured as "our greatly beloved."202 This was undoubtedly a consequence of the central role he had assumed during the reign of Louis III. Like Gauzlin, he also received another royal charter, in his case for his abbey of Morienval.203 Theoderic's intervention for Bishop Berno of Châlons in the first diploma also indicates that, at least by now, Berno, who "stands out in the government of our realm and in very useful counsel and aid",204 not only possessed some influence at Carloman's court, but also had joined the circle of nobles originally dominating that of Louis III. The same diploma also saw the intervention of Bishop Angelwinus of Paris,205 another one of those assigned to remain with Louis


the Stammerer by the capitulary of Quierzy, signalling that he had also joined the group around Gauzlin and Theoderic.

Nobles from the southern regions who had belonged to the group around Hugh remained influential. Theoderic the Chamberlain appeared again in connection with the church of Autun, serving as *ambasciator*, 206 and was also granted a diploma for his abbey, Saint-Martin of Autun.207 The position of Archbishop Frotar of Bourges, who intervened together with Hugh the Abbot for the church of Marseille,208 appears to have remained unchanged. Bishop Walter of Orléans received two more diplomas for his own church, both of them at Hugh's intervention.209 Count Wido continued to be important, being described as "our highly beloved and illustrious count."210 Count Ansgar of Oscheret, who had belonged to the group around Hugh three years earlier when he had been part of a legation to Louis the Younger, 211 now

This, however, would leave us with a dead bishop appearing before the king. Therefore, two dates have to be examined: that of the diploma and that of the translation. The former date reads as *Data III idus martii, indictione II, anno II regni Karlomanni regis in Frantia* (DCmII 76, 198). We can agree without any hesitation with Bautier's conclusion that the second year of Carloman's reign in Francia corresponds with the year 884 (his brother Louis having died on the 5 August 882), even more so as the date also matches the year of the indiction. Alternatively, setting aside the indiction and the *in Frantia*, we could also read the date as 13th March 881, calculating from his first coronation at Ferrières. This, however, would mean that Carloman would have been at Compiègne at that moment, since the diploma was issued there. According to Hincmar however, at this moment Carloman was in Burgundy, fighting Boso (Annales Bertiniani 881, 243−244: *Anno incarnationis dominicae 881 remanente Karlomanno cum suis contra Bosonis seditionem, Hludouuicus, frater eius, reuersus est in partem regni sui contra Nortmannos.*), which also corresponds with Carloman's diplomas. The date 13th March 884 therefore appears to be correct.

The text of the *Translatio* on the other hand is less clear: it does give 884 as the year, but only for the moment at which the priest Theodebert approached Gauzlin to propose the translation of the relics. But in fact, between this moment and the actual translation, some time seems to have passed, so that the *die depositionis S. Mederici, in qua praefatus sacerdos id facere disposuerat*, could also have taken place the following year. This also leads to another problem, the *die depositionis S. Mederici*, which could either mean the day of Saint Merry's death (29th August) or the day of the translation of the body (which according to Mabillon, Acta Sanctorum Benedicti III,1, 14, n. c was celebrated 22nd January).

We therefore propose to date Angelwinus' death not to have been 8th December 883 but 884. If Gauzlin succeeded him within a couple of weeks, this would still have left enough time for Theodebert to initiate the translation, which then would have taken place in 885, either early in January or late in August. However, this would mean that Gauzlin was by no means appointed bishop of Paris by Carloman (as Werner, Gauzlin, 216 assumed) since the latter had already died 5th or 6th December 884 (Bautier, Recueil Louis II, Louis III et Carloman II, 210), but succeded him during a period of *interregnum*, before Charles the Fat arrived in the realm. This solution, of course, assumes that the account of the *Annales Vedastini* is wrong.


succeeded Angelwinus as bishop. This reading is also backed by the account of the *Annales Vedastini* (884, 54) which read *Anno Domini DCCCLXXXIIII. Per idem tempus mortuo Engelwino Parisiorum episcopo Gauzilinus abba subrogatur in sedem eius.*

reappeared in two diplomas, in which he was connected to Theoderic the Chamberlain and Count Robert of Troyes. 212 Finally, we may also add Bishop Adalgarius of Autun, who had rejoined Carloman after having supported Boso's rebellion, to this circle, although his connection to the court probably depended on his relationship with Theoderic and Ansgar, who served as *ambasciatores* in the royal charter granted to his church.213 Despite the integration of the North, the circle around Hugh thus continued to be extremely present at the royal court.

Thus Carloman's united court saw the continuing dominance of the groups that had formed years earlier around Hugh the Abbot and Gauzlin. Yet, again, access to the king was not limited to the nobles belonging to these groups. The aforementioned Robert of Troyes appears to have had connections to Ansgar of Oscheret, yet his own position at court—he demanded three diplomas for the abbey of Montiéramey during Carloman's entire reign214—was probably more determined by his relation to the king himself, since he was married to Gisela, Louis the Stammerer's daughter from Ansgarde and thus Louis' and Carloman's sister.215 Another interesting case is that of a certain Erifons, another "illustrious *fidelis*" of the king.216 He only appeared in Carloman's last diploma, probably issued just after the king had been fatally wounded. In the face of his inevitable death, Carloman wished to make donations to Saint-Crépin of Soissons in atonement for his sins.217 The choice of words and Erifons' presence at the king's side just before his death, as well as his being charged with the implementation of the measures laid down in the diploma (he also served as *ambasciator* at Gauzlin's order218), indicate that he, like Robert, enjoyed a more personal connection to the king than those of the other nobles dominating the court. The same is also possible for Abbot William of Croix-Saint-Ouen, who is described as "our highly beloved and illustrious *fidelis*  William" when he demanded a royal charter for his abbey.219 Nothing further is known about this abbot; however, as the way he is described shows, he appears to have been held in some esteem by the king, either because of his political importance, or his personal relationship with him. In two other cases, those of the


counts Hilduin220 and Alardus, 221 due to the loss of the diplomas that recorded their appearances at court, statements about their relation to the king or their connections to the dominating circle cannot be made at all.

However, while the likes of Robert and Erifons appear to have been close to Carloman, their political influence must have been limited. They had connections to the circle around Hugh and Gauzlin, but either only to second-rank nobles, as in the case of Robert's connection to Ansgar of Oscheret, or in a subordinate role, like Erifons to Gauzlin. The leading members of the circle, on the other hand, distinguish themselves by their frequent interventions for churches, abbeys and individuals, in some cases for those which were close to the regions where they themselves held important *honores*, such as Hugh's interventions for the church of Orléans,222 while in others there was no connection at all. Institutions far from the royal court, like Saint-Sauveur of Alfa or the church of Marseille, now sought out the inner circle to gain access to the king.223

These latter institutions point us to a last observation concerning Carloman's reign over the united realm. While he remained within Francia from September 882 onwards, his connections with the regions south of the Loire remained intact. We have already shown the presence of nobles from Burgundy at his court. To these we can add a number of appearances from Aquitaine and Septimania. Archbishop Frotar of Bourges was certainly the most important of the king's contacts in Aquitaine, but Carloman was also sought by representatives of the abbey of Solignac224 and the abbess Adalgard of Sainte-Croix of Poitiers. 225 In Septimania next to Saint-Sauveur of Alfa also Archbishop Sigebod of Narbonne, who had been present at court in 881, was in contact with the king, sending a legate to demand a further diploma for his Church.226 Contacts with the south of the realm are thus shown to have remained intact.

#### **II.3 New faces on the rise: Charles the Fat**

After Carloman's sudden death early in December 884, the West Frankish nobles sent for Emperor Charles the Fat to take over the realm.227 Up until his deposition by East Frankish nobles four years later, Charles had visited the West only twice. His initial voyage to receive the homage of the Western nobles brought him


from Italy via Lotharingia to Ponthion.228 In 886, he undertook a campaign to relieve Paris from a Viking siege, an endeavour that ended in him negotiating the payment of tribute and allowing the Northmen access to Burgundy.229 However, contacts with the West Frankish elite were not limited to these periods. On various occasions, he was sought out by individual nobles while absent from the realm,230 visits that signal special relations between them and the emperor.

According to the *Annales Vedastini*, the decision to call Charles into Francia was made by the Franks, who sent Theoderic of Vermandois to the emperor.231 Of course, the wording of the annals does not have to be taken literally and there may indeed have been nobles who opposed the choice. However, since there are no further reports of any such opposition, the leading men seem to have backed the call, a supposition evident from the fact that in Theoderic they chose one of their own as leader of the delegation. During the interregnum, both Hugh the Abbot and Gauzlin continued to exercise considerable influence: Hugh sent envoys to the Northmen now returning to the realm,232 while Gauzlin secured the vacant see of Paris for himself233 and installed his nephew Ebolus as abbot of Saint-Germaindes-Prés. 234 Given the unity shown by the Frankish nobles, both appear to have favoured the emperor as Carloman's successor. In the case of Gauzlin, it is also worth pointing out the connections he had made after his capture in the battle of Andernach to nobles from the East Frankish kingdom235 and which he already had used after the death of Louis the Stammerer. That he did indeed exert his influence on Charles' behalf may also be indicated by the speech Abbo attributed to him in his famous poem, praising the emperor's qualities.236


*Urbs mandata fuit Karolo nobis basileo, Imperio cujus regitur totus prope kosmus Post Dominum, regem dominatoremque potentum, Excidium per eam regnum non quod paciatur, Sed quod salvetur per eam sedeatque serenum.*

While the part these men took in the decision to call the emperor seems to be clear, assessing their influence at Charles' court poses considerable problems. Hugh the Abbot appears only once in the sources before his death when his absence from the emperor's first campaign against the Northmen due to an illness is noted.237 Whether this illness was indeed the real issue or just a pretext to cover a loss of influence must remain an open question. Posthumously, however, he is mentioned several times. The *Annales Fuldenses* describe Hugh and Gauzlin both as "abbots and leading commanders of Gaul, on whom the Gauls had placed all their hopes against the Northmen," whose whose deaths had encouranged the enemy and led to the new attack. 238 However, this note clearly confuses the chronology—Hugh and Gauzlin only died at the time of the siege of Paris—and reads more like a *laudatio* on both their lives rather than on their actual political importance during the year preceding their deaths. Similarly, Hugh's posthumous appearances in Charles the Fat's diplomas239 may do him honour, but only reflect the memory of a man who had shaped West Frankish politics for a long time and are, moreover, linked to abbeys that had been under his control for a long time. However, there is one event that possibly casts some light on Hugh's position at the imperial court: when the second campaign against the Northmen was organised, raising levies in Neustria and Burgundy, regions where Hugh had been most influential, the command went to Ragenold, the *dux* of Maine (*dux Cinomannicus*), instead of Hugh. 240 This may of course have been due to his already mentioned illness, although it seems just as likely that Hugh's influence was finally on the wane.

In the case of Gauzlin, who Simon MacLean considers one of the key figures in what he labelled the "North Frankish circle",241 we encounter the same problematic situation. His prominent role in the defence of Paris against the Northmen is well attested by the *Annales Vedastini* and Abbo's *Bella Parisicae Urbis*, yet is easily explained by his leading role as bishop of the city and does not necessitate any proximity to Charles. 242 More promising is a phrasing in the *Translatio S. Mederici*, stating that Gauzlin could not take part in the translation because he was occupied with various duties in service of the realm,243 although this may relate to


his participation in the campaigns against the Northmen in 885. His influence in Francia best becomes visible in the letters he sent to Count Erkanger, asking him to fetch help from the emperor.244 Yet, while this episode certainly underlines his good relations with leading members of the Frankish nobility, it does not allow us to draw any conclusions on his standing with the emperor. Did Gauzlin use an intercessor because he feared a direct approach would not yield the intended results?

Finally, Theoderic and his brother Aledramnus appear as military leaders in the context of the siege of Paris. Aledramnus took over command of the fortress built at Pontoise but was soon forced to withdraw to Beauvais.245 In Abbo's poem, both he and Theoderic are named as leading a force of 600 men to a victory against the Northmen in the course of the siege, most likely the same force that had been sent by the emperor to the aid of Paris.246 At least in the second case, it seems clear that they had received their orders from the emperor. Aledramnus can also be seen at the emperor's court, the only one among the nobles mentioned so far who came to meet Charles on his first journey to the West, intervening together with a certain Bishop Warnulf in favour of the church of Chalon-sur-Saône as "our beloved count" at Gondreville.247 Warnulf 's see is unknown,248 although his description as "our most beloved bishop" indicates his importance for Charles. Aledramnus' connection to Warnulf further emphasises his own role in the emperor's network in the Western realm. So far, the two brothers emerge as the only Westerners who appear to have been close to the emperor.

In contrast, to most of those who had formed the dominant circle at Carloman's later court, Bernard of Auvergne, who had been an important ally to them although never visible for a long period at the royal court, appears to have quickly entered into the circle around the emperor and received a very honourable welcome. He intervened on behalf of the church of Lyon together with the emperor's closest advisor, Bishop Liutward of Vercelli. He is described as "most illustrious *marchio,*" a very rare title that further underlines his importance.249 The diploma also seems to acknowledge Bernard's control over Lyon,250 appearing to be another concession made by the emperor to tie the most powerful man in Aquitaine to him and likely the result of their long established contact if the charters from Saint-Julien of Brioude can indeed be interpreted in this way.


<sup>244</sup> Annales Vedastini 886, 59.

<sup>245</sup> Annales Vedastini 885, 57−58. The report of the annals led to the belief that Aledramnus was count of the Vexin, where Pontoise is situated (Kalckstein, Geschichte, 472), or of Beauvais (Favre, Eudes, 19). Lot, Notes, 150, considers him as count of Laon, due to DLS 28, in which he was given property in the Laonnais. MacLean, Kingship, 106 names him as count over all three.

Among those seeking out the emperor, we can also find two others who belonged to the circle dominating Carloman's court. In the aftermath of the siege of Paris, both Bishop Berno of Châlons and Count Ansgar of Oscheret requested diplomas.251 Ansgar's appearance points towards a wider circle of nobles in Upper Burgundy on which Charles now relied. He intervened together with Count Milo of Langres for the provost of the church of Langres, the see of Bishop Geilo. The latter was the single largest recipient of the emperor's diplomas within the West Frankish kingdom,252 a key figure in the emperor's control over Upper Burgundy.253 It is not only the number of royal charters, however, that characterises the importance of see and bishop for Charles. Geilo was one of the first to meet the emperor, intervening on behalf of Saint-Bénigne and Saint-Etienne of Dijon,254 both of which indicate his influence in the region. His requests were also supported by Liutward of Vercelli255 and even Charles' wife, the empress Richgard. 256 Finally, Geilo was one of the few who did not wait for the emperor to come to the West, but sought him out on two occasions when he was in the East.257 The relation between Geilo, Milo and Ansgar, as demonstrated by the intervention of the latter two for the provost of Langres, appears to have been a close one. After the emperor's death, all three of them belonged to the supporters of Wido of Spoleto. 258

Contact with nobles from Upper Burgundy was not limited to these three, however. Robert of Troyes, whose contact with Ansgar we mentioned above, had died in the defence of Paris but was succeeded by his nephew Adelelm, 259 who continued to support the besieged Parisians.260 Considering Robert's close relations with the abbey of Montiéramey, it does not seem too far-fetched to assume that it was at Adelelm's request that Charles now issued another diploma for the monastery.261 Two more royal charters were petitioned for by a certain Count Rudolf and his son Pippin. 262 In both diplomas, property in northern Burgundy, Bar and


the Lassois, was given to *fideles*, which points towards them being local counts. Finally, Bishop Agelmar of Chalon-sur-Saône263 and the new abbot of Saint-Martin of Autun, Gregory, belonged to those in contact with the king.264

Two groups can be distinguished among the nobles mentioned so far. The old core, which had formed the inner circle around Carloman and his predecessors, was still intact and cooperated with the emperor. However, at the same time their influence at Charles' court is impossible to determine. Of its members, only Aledramnus appears in the emperor's diplomas while the connections of Hugh, Gauzlin and Theoderic with Charles remain in the dark. All of them were strongly involved in West Frankish affairs, most notably in the fight against the Northmen. Yet, at the same time, their influence in royal politics appears to have diminished, probably caused in large part by Charles' absence from the realm, depriving them of the opportunity to influence the decision-making process at court. On the other hand, for Bernard of Auvergne, always more an associate than a constant member of the old circle, not much appears to have changed. His position had increasingly become more distant from court, and apparently mainly focused on affairs in Aquitaine. Here, he acted for the emperor as a local power broker, channelling the royal will into the region south of the Loire via his own connections.265

At the same time, we can note the rise of other nobles, such as Geilo of Langres, who now became a key figure in securing Upper Burgundy for the emperor. Geilo, like Ansgar of Oscheret, had already been in contact with the dominating circle, yet now his position appears to have been much more central and, most of all, independent of them. And there were other men who rose through the emperor's protection. The most interesting example is Ragenold, according to the *Annales Vedastini* the *dux* of Maine.266 When he was killed, the newly assembled army made up of contingents from Neustria and Burgundy dispersed, indicating that he had been its leader. That the death of the commander led to the end of the entire campaign was not extraordinary.267 In 866, Robert the Strong's death in combination with Ramnulf 's injury lead to the dispersal of their force268 and similarly in 887 Duke Henry's death in battle was followed by his army returning home.269 All of these nobles appear to have held key positions at the head of their respective armies in which they could not easily be replaced. In the case of Ragenold however, the question arises as to how he had become the leader of such an army. No sources indicate that he belonged to the circle of nobles dominant at court. It has been suggested that he might have been a relative of Gauzlin, 270 yet how close


<sup>263</sup> DChF 119.

<sup>264</sup> DChF 122.

<sup>265</sup> See also MacLean, Kingship, 69−70.

this relationship was remains unclear. The only thing we know for certain about him is that he was in conflict with Bishop Lambert of Le Mans, who excommunicated him during 883−885 due to him having appropriated Church income.271 He was certainly a powerful noble based in Neustria, possibly the leader of a march against the Bretons and Northmen.272 Although our sources do not mention any influence of the emperor in the assembling of the army, it seems very unlikely that he was not involved at least in the appointment of its commander. As the annals of Saint-Vaast report, the army consisted of forces of "all who were in Neustria and Burgundy,"273 pointing to a large-scale operation. Given that Hugh the Abbot, who held honores in both regions, appears to have been passed over as commander, it seems likely a central authority—the emperor—was behind its organisation. Ragenold thus would have been Charles' choice, undoubtedly based on his position, experience and, if indeed correct, his relation to Gauzlin. But most of all, it was also a choice to appoint someone new to a key position in the defence of the realm.

Ragenold was not the only new noble Charles appointed—or had to appoint following the deaths of Hugh the Abbot and Gauzlin. Anskeric, who now became bishop of Paris, is another example. He was the brother of Count Tetbert of Meaux and appears to have competed with Gauzlin for the see of Paris upon Angelwinus' death.274 He was also one of the first to meet with Charles upon his initial visit to the West, intervening with Count Rudolf and his son Pippin on behalf of a *fidelis* from Upper Burgundy,275 thus establishing contact at a very early moment. Hence, Anskeric appears to have recommended himself to Charles at a very early point, while his nomination also shows the interest of the emperor in installing trusted men in key positions in the realm. The connection between the two also became apparent in 887, when Anskeric met Charles to take over the tribute due to the Northmen.276

However, the most important man promoted by the emperor was without a doubt Odo, son of Robert the Strong. Olivier Guillot, in an article dedicated to the Robertian's early career, claimed that, after the siege of Paris, Odo was exercising royal prerogatives such as distributing *honores* in the Western realm despite Charles still ruling.277 This, however, was convincingly refused by Simon MacLean, who in turn argues for a close cooperation between the count and the emperor instead of a political rivalry for power,278 a view we also follow. Odo had succeeded Conrad as count of Paris some years earlier but remained invisible during

271 Actus pontificum Cenomannis, 339−340.

273 Annales Vedastini 885, 57: *Inter haec omnes qui morabantur Neustria atque Burgundia adunantur et collecto exercitu adveniunt quasi debellaturi Nortmannos.*


<sup>272</sup> Guillotel, Marche, 12.

<sup>274</sup> See chapter I.2.2.

that time. He appears to have been politically close to Gauzlin279 and they acted together in the defence of Paris, as becomes clear from Abbo's poem.280 Besides Gauzlin, he also enjoyed the support of Theoderic and Aledramnus since he married the latter's daughter Theodrada, probably during the early 880s.281 His role became of central importance only after the bishop's death, when he took over responsibility for the city.282 Contact with Charles may have been established even before the emperor took over the West, since Odo's brother Robert appears to have held a countship at Namur283 and had intervened in one of Charles' diplomas two years earlier.284 Robert's importance at court becomes even clearer from the phrasing used to describe him in the charter, "our most loyal noble man count Robert", the use of a superlative generally being reserved for special cases.285 After the siege of Paris was lifted, Charles granted Odo the *honores* of his father Robert the Strong, 286 namely the counties along the Loire together with the abbeys Saint-Martin of Tours and Saint-Aignan. A number of diplomas issued on this occasion, confirming an exchange for Saint-Aignan and confirming the property and immunity of the church of Tours,287 further document Odo's newly won importance. His key position in the emperor's West Frankish affairs is underlined by a diploma for Saint-Martin issued at his request at Kirchen in June 887.288 It was a confirmation of a charter from Odo two months earlier, in which he returned property to the monks.289 However, the parts copied by the royal chancellery contain the prayer services for himself and his father Robert the Strong, 290 an almost unique concession in the emperor's diplomas.291 The choice to appoint Odo—apart from


his brother's connection to the imperial court—is not surprising. When Charles the Bald had installed Odo's father in the region, he had also settled some of Robert's supporters there.292 As Karl Ferdinand Werner has shown, these same men were still in place when Odo took over and supported him.293 Charles' diploma for Saint-Martin of Tours, issued after Hugh the Abbot's death, tells us that Odo had given property to the monks294 and his intervention with Bishop Adalard of Tours for the latter's church295 indicates that the two, at least after Odo's appointment, were close as well. Hence, Odo appears to have been well connected in the area and was therefore an ideal choice for the vacant position along the Loire. In addition to the bishopric of Paris, Charles thus secured another key position in the Western realm for himself.

Charles tried to use all opportunities presented to him to set up men of his choice in central positions and to tie them to him. In the case of the see of Auxerre, the emperor tried to install his candidate Teutbert, yet failed when Bishop Walter of Orléans intervened.296 Instead, Herfrid, one of Walter's relatives,297 was elected. The *Gesta Pontificum Autissiodorensium* do not know of Teutbert, but report that Herfrid was the royal candidate.298 Of course the *Gesta* are not an entirely reliable source.299 Nevertheless, it would seem possible that, encountering Walter's resistance, Charles altered his decision and settled for a candidate who was also agreeable to the bishop of Orléans. In this case, it seems likely that the emperor's lost diploma for the church of Auxerre was issued at Herfrid's request300 and that, like Anskeric, Herfrid would then have been one of the emperor's men. Concerning the appointment of bishops, it has also been suggested that the elevation of Walter's nephew, another Walter, to the see of Sens was also initiated by Charles.301 Finally, at Paris, Charles seems to have favoured Abbot Ebolus of Saint-Germaindes-Prés, Gauzlin's nephew, for whose abbey he issued a diploma.302 It also appears that, upon Gauzlin's death, Ebolus was given control over the deceased's other abbeys, namely Saint-Denis and Jumièges,303 indicating that Charles tried to tie the abbot even closer to him. In addition to these men, the emperor also strengthened his relations with other nobles. The fight against the Northmen had left a number


<sup>292</sup> Annales Bertiniani 866, 126: *Karolus Rotberto comiti abbatiam Sancti Martini ab Engilwino ablatam donat et eius consilio honores qui ultra Sequanam erant per illius complices dividit.*

of counties vacant and thus provided Charles with various *honores* to distribute among men of his choice, which he did before returning to the East,304 although unfortunately nothing further is known about these measures.

In addition to these more intense contacts, we can also make out a number of other nobles and institutions seeking out Charles. Archbishop Fulk of Reims had already communicated with the emperor in 883, when he asked him to intervene on his behalf with the pope to be given the pallium and a confirmation of the privileges of his Church.305 A year later, he successfully requested the restoration of goods in Thuringia and Lotharingia to the abbey of Saint-Remy of Reims when the emperor was visiting the western regions of his realm.306 Once Charles had taken over *Francia Occidentalis*, however, Fulk does not appear to have belonged to those enjoying imperial favour. Flodoard only records one letter from this time, probably dating to between February and July 886 due to its references to the siege of Paris. In this, he urged Charles to protect the realm and painted a picture of the consequence of the loss of the city in the darkest colours.307 According to Gerhard Schneider, this was a sign of his interest in the wider affairs of the realm and his claim for the primacy of the church of Reims.308 While this letter may indeed have contributed to the emperor's intervention, it is more likely that the pleas from Gauzlin and Odo carried more weight.

Apart from Fulk, Charles was also sought out by other churches and abbeys throughout the realm. From Septimania came Bishop Theotarius of Girona to receive a confirmation of the properties and rights of his church.309 Charles restored property alienated from the Aquitainian abbey of Saint-Agricol and Saint-Vital. 310 In Burgundy, he did the same for the canons of Saint-Vincent of Mâcon at the request of their bishop, Letard. 311 Another bishop from Burgundy, Emmensus of Nevers, received a confirmation for his new foundation at Cusset,312 as did Abbot Bonifatius of Sainte-Seine for his own abbey.313 Finally, in the North, Saint-Maur of Fossés314 and Saint-Médard of Soissons315 were granted diplomas. This last case



<sup>309</sup> DChF 148.

appears to have been more of a personal matter, the emperor donating a *villa* to the abbey for the sake his soul as well as those of his ancestors and predecessors.316 The abbey was also the burial place of Duke Henry, one of his closest advisors and his most important commander, who had been killed the year before,317 which may also have been a reason for the emperor's choice to grant the diploma. This is a rather limited number of contacts, yet considering that Charles only visited the West on two occasions and for very short periods of time, these diplomas nevertheless demonstrate that the emperor, like his predecessors, was in contact with all parts of the realm.

#### **II.4 Resistance, integration and rebellion: Odo**

The death of Charles the Fat had left a vacuum at the centre of West Frankish politics. For the first time since the death of Louis the Stammerer, nobles formed rival factions, each trying to establish their own candidate on the throne. Odo's main supporter is identified by the *Annales Vedastini* as Count Theoderic of Vermandois, 318 which is hardly surprising considering their relationship. Theoderic also served Odo shortly afterwards as an ambassador to Arnolf,319 his last appearance in the sources. By 895, he was dead and his *honores* had been passed on to his son.320 Odo's brother Robert was also a key figure in these events; Odo appears to have passed on the majority of his *honores* to Robert upon becoming king. No contemporary source actually mentions this procedure, but the *Annales Sancti Germani Parisiensis* note that, in 888, Robert became count of Paris,321 he furthermore appears as abbot of Saint-Martin of Tours around that time322 and finally is addressed as "our *senior*" by Viscount Garnegaud of Blois. 323 Moreover, the other viscounts of the Robertian counties along the Loire, Hucbert, Hardrad and Hatto were now in close contact with the king's brother, as a number of private charters from Angers and Tours show.324 The *honores* having been passed on, Odo nevertheless maintained contact with the Neustrian bishops. Raino of Angers received two diplomas,325 the church of Orléans, where his notary Throannus



<sup>322</sup> DOdo 19.

became bishop after Walter's death,326 received another327 while a judgement favoured the archbishop of Tours.328 Robert's power was further increased when Odo installed him as count of Poitiers in around 892329 and although he does not appear in any of the datable diplomas before 893330 nor in the major narrative sources, he was certainly one of the pillars his brother's rule rested on.

Among the others supporting Odo in those early days was an old acquaintance of his, Count Hucbald of Senlis. The two of them are named by a widow complaining to the archbishop of Sens that they had interfered with her rights and possessions.331 Hucbald's central role also becomes evident from three of Odo's diplomas, in which he intervened for various important abbeys, namely Saint-Hilaire of Poitiers, Montiéramey and Saint-Maur of Fossés.332 In these, his proximity to the king is further underlined by expressions such as "our *procer*" and "our *fidelis* in all affairs." Also close to Odo at this time was his kinsman333 Count Altmar, who before 888 had signed one of Odo's charters for the church of Paris next to him and Robert334 and who now also witnessed one of Robert's private charters.335 A few years later, after the death of Ramnulf II of Poitiers, Odo installed him as count of Poitiers336 as part of an attempt to tighten his grasp on northern Aquitaine. Altmar rebelled soon after when Odo withdrew the county from him to hand it over to his brother Robert. He took Poitiers by force337 and allied himself with William the Pious and Richard the Justiciar.338 Odo and Altmar appear to have come to an agreement soon afterwards, since Altmar appears alongside Robert as *marchio* in a diploma handing over the abbey of Saint-Hilaire of Poitiers to Bishop Egfred. 339 According to Eduard Favre, this Altmar was the same who held the abbey Saint-Médard of Soissons in 900.340 His identification however seems to be partially based on a forgery indicating that, in 893, Odo had donated property to the abbey.341 The Altmar in control of Saint-Médard of Soissons is generally thought to have held Arras, where he took over Saint-Vaast in 899,342 while


the Altmar close to Odo appears to have been mainly active south of the Loire.343 While we therefore have to refuse Favre's identification, Altmar, nevertheless, like Robert and Hucbert, belonged to a group close to the new king whose members owed their influence to their earlier connections to him.

Other allies came from the circle of men elevated by Charles into central positions. The new archbishop of Sens, Walter, carried out his first coronation at Compiègne344 and became royal archchancellor in 894.345 Whether two other men, Bishop Anskeric of Paris and Abbot Ebolus of Saint-Denis, belonged to Odo's very first allies remains unknown. From June 889 onwards, however, they belonged to the group of his most important supporters. Anskeric, "our most beloved", a phrase that emphasises his proximity to the king, received two diplomas for his abbey Saint-Germain of Auxerre.346 Thus, his presence at Odo's side not only meant that Paris was tied even closer to the king, but also that he had a foothold in Burgundy, where local nobles and bishops had originally opted for Wido of Spoleto. Anskeric's Burgundian connections were probably also the reason for his intervention on behalf of Bishop Argrim of Langres, whose appearance at Odo's court in December 889 marked an important victory for Odo. Argrim's predecessor Geilo was one of those who had opposed the Robertian: it was he who had crowned Wido king and therefore Argim's support for Odo meant that the king had obtained the support of Langres.347 Furthermore, Anskeric was the brother of Count Tetbert of Meaux and thus strengthened Odo's relations with the nobles of Francia even more. It may have also been due to his influence that, after Tetbert's death, the county passed to Heribert I, who was probably related to the brothers.348 Finally, Anskeric's proximity to the king is emphasised by his serving as *ambasciator* in a diploma for the church of Narbonne349 and becoming archchancellor in 892 in succession to Abbot Ebolus. 350

Ebolus appears as archchancellor in Odo's first preserved diplomas, dating to 13th June 889.351 Next to Gauzlin and Odo, Ebolus had been one of the leaders of the defence of Paris, particularly after Gauzlin's death and in times of Odo's absence from the city.352 Ebolus' role in Odo's bid for the throne is difficult to ascertain. His brother Ramnulf II of Poitiers certainly did not side with Odo and only submitted early in 889;353 it is even possible that he himself had aspirations


for the crown.354 When Odo went to Aquitaine to meet with Ramnulf, he took only a few men with him. The meeting between the two was undoubtedly preceded by negotiations in which Ebolus, due to his connections to both parties, must have played a key role. Whether he had already become Odo's archchancellor must remain unclear, yet his appointment ensured that the king had a direct link to the most important man in Aquitaine. After the death of Archbishop Frotar of Bourges, Ebolus was also given control over Saint-Hilaire of Poitiers, further strengthening Odo's connection to the most important family in Aquitaine. By the end of 889, his petition in favour of the abbey was supported by the intervention of the counts Hucbald and Heribert, 355 thus publicly placing him at the core of the king's network of close associates.

We have already mentioned Heribert's close association with the king and can thus refrain from going into any further detail.356 By furthering Heribert, Odo increased his grasp on Francia even more. Control over this region was decisive since it not only held the rich remnants of the Carolingian fisc, but also the palaces and abbeys closely tied to the memory of their family, most notably Charles the Bald's Compiègne, which Odo chose as the place for his coronation. But Francia was also the region where the resistance against Odo had been most fierce: Archbishop Fulk had led the party in favour of Wido and shortly afterwards allied himself with Abbot Rodulf of Saint-Vaast and Saint-Bertin and his cousin Baldwin II of Flanders.357 This party started to dissolve only after Odo's victory over the Vikings at Montfauçon and his acknowledgement by Arnulf, when Baldwin submitted himself to the king.358 Fulk and Rodulf followed soon after. Odo's second coronation at Reims indicates at least the archbishop's consent, even if not his participation in the ceremony.359 Fulk later received two diplomas for his church360 while his overall behaviour towards the king, as argued, shows general cooperation.361 Rodulf was soon integrated into Odo's reign. His request for a confirmation of the monk's *mensa* was supported by Queen Theodrada, her only appearance in the royal diplomas.362 The moment of the charter's deliverance may have been motivated by the Northmen's return to the Seine-Oise region in 890,363 necessitating the abbot's cooperation with the king. Securing Francia was probably also the motivation behind


<sup>354</sup> Annales Fuldenses (Ratisbon continuation) 888, 116: *…deinceps Ramnolfus se regem haberi statuit.* See also chapter I.1.4.

<sup>355</sup> DOdo 16 (14th December 889). The identification of the Ebolus in the diploma with Ebolus of Saint-Denis is disputed. We follow Bautier, Recueil Eudes, XXIII and 74−75. Different on this matter Favre, Eudes, 33, n. 10.

<sup>356</sup> See chapter I.2.1.

<sup>357</sup> Annales Vedastini 888, 64−65.

<sup>358</sup> Annales Vedastini 888, 66.

<sup>359</sup> Schneider, Erzbischof, 60–62 and Guillot, Étapes, 215−216.

<sup>360</sup> DDOdo 31 and 47.

<sup>361</sup> See chapter I.2.1.

the appointment of Odo's relative, Walker, as commander of the castle of Laon,364 as was a diploma for Bishop Dido of the same city, guaranteeing the security of the immunity of the Church's buildings and houses from intrusion by royal agents.365 By placing men closely associated with him in central positions, as well as by appeasing former opponents, Odo thus secured Francia for himself.

#### **Burgundy and Aquitaine**

Burgundy, as mentioned, was, besides Francia, the region where Odo's rival for the crown, Wido of Spoleto, had found most of his support. Geilo of Langres crowned him, the counts Milo and Ansgar followed him back to Italy. Winning the acknowledgement of the Burgundian nobles and forging links with its most important representatives were therefore vital when it came to securing Odo's rule over the realm. By June 889, the first steps had been taken. At Odo's first major assembly at Orléans, the king issued a large number of diplomas for recipients from Burgundy, Aquitaine and Septimania, marking his acknowledgement as king over the entire realm. Among those coming from Burgundy were Abbot Odo of Vézelay, 366 who had already requested diplomas from Louis the Stammerer and Carloman II, and Abbot Biltgerius, who now received a royal charter for his abbey of Saint-Philibert of Tournus.367 Even after Geilo had become bishop of Langres, he had preserved strong links with his former abbey, as his intervention on its behalf in front of Charles the Fat demonstrates.368 Biltgerius' presence at Odo's court therefore, like that of Geilo's successor Argrim half a year later,369 is a sign that the core region of Wido's support had now chosen to submit to Odo. A final key figure in the area was Bishop Adalgarius of Autun, who, not unlike Odo of Vézelay, we have already seen at the courts of Louis the Stammerer and Carloman II. While under Carloman his role in firmly establishing royal control over the region was undoubtedly important, especially during the efforts to suppress Boso's revolt (like many others, Adalgarius had originally been one of Boso's supporters), and at Odo's court he seems to have risen quickly to an even more central role. Property was restored to his church370 and in 893 he replaced the rebellious Anskeric as royal archchancellor,371 thus providing the king with a direct link to the Burgundian nobility. Considering the endeavours of Richard the Justiciar, who around the same time had formed an alliance with two of Odo's enemies, William the Pious of Aquitaine and Altmar of Poitiers,372 this measure was surely meant to provide Odo with a finger on the pulse of events in Burgundy.


<sup>364</sup> Annales Vedastini 892, 72.

<sup>365</sup> DOdo 29.

<sup>366</sup> DOdo 10.

In the case of Aquitaine, we have already noted the importance of Odo's relative, Count Altmar, as well as that of Abbot Ebolus, the royal archchancellor and brother of Ramnulf II of Poitiers, who, according to Ademar of Chabannes, from then on entertained close relations with the king and probably also introduced Count Gerald of Aurillac, one of his close supporters, to the royal court.373 After the king had come to an agreement with Ramnulf, abbots Theoderic of Solignac and Gerulf of Beaulieu appeared at the royal court to ask for diplomas.374 Again, like Odo of Vézelay, both had been regular visitors to the royal court, acknowledging the rule of Louis the Stammerer and Carloman II, thus demonstrating that Odo had now been accepted as their successor. However, more importantly, both cases also demonstrate that Odo could now also rely on the approval and support of another political heavyweight in the area, Archbishop Frotar of Bourges. In the case of Solignac he acted as *ambasciator*, while we have already seen him closely associated with Beaulieu in the time of Carloman II. 375 Further support appears to have come from a certain Count Hugh, to whom Odo gave the *honores* of the rebellious William the Pious in 892.376 In Aquitaine, as in Burgundy, we can therefore observe the king's attempts to establish links allowing him to remain in touch with the local nobility. These could be via important local nobles whom he tied to himself by giving them offices and *honores*, or men from the circle already surrounding him, who he now tried to install in key positions in the region.

Finally, at the assembly at Orléans in July 889 we can make out a large number of contacts in Septimania and the Spanish March, most notably Bishop Ermemirus of Girona377 and Count Sunyer of Ampurias, 378 representatives of one of the most influential families in the coastal Pyrenees,379 both of whom were intervening together on behalf of the abbey Saint-Polycarpe. The latter was not the only monastery in the South receiving a diploma on that occasion. In the case of Saint-Paul of Fontclara,380 Odo pulled out all the stops to demonstrate to the realm that he was willing to act like a king. With the consent of his closest *fideles,*381 as the charter makes special note of, the king granted the abbey the complete scope of royal attention: confirmation of its property, protection, immunity, the liberty to choose its abbot and some parts of the fisc in the area. In turn, he asked for a regular prayer service to be held for him, his family and the realm, thus not only taking care of his eternal soul, but at the same time ensuring that his rule was remembered in the South. Contact with the far South remained strong after this, with regulars such as Archbishop Theodard of Narbonne and Ememirus' successor at


Girona, Servus Dei, petitioning on behalf of their churches at the royal court382 as well as other new appearances such as the abbots Sunifred of La Grasse, Aduvirus of Notre-Dame of Amer and Tenericus of Saint-Martin of Mont-Redon. 383

Only very few of these men coming from so far away to seek out the king are likely to have wielded considerable influence at the royal court. Yet their presence is a good indicator that the nobles of the entire realm co-operated with the king. This image is also confirmed by the synod of Meung-sur-Loire and the assembly of Verberie in the early 890s. The synod saw the presence of no less than five archbishops and nine bishops, representing Neustria, Francia, Burgundy, Aquitaine and Septimania.384 The assembly appears to have been smaller: the original sentence against a monk from Montiéramey, which was confirmed in a later diploma,385 had been pronounced by the archbishops of Reims, Sens and Rouen as well as three bishops from Francia,386 two of whom, Dido of Laon and Honoratus of Beauvais, later remained loyal to Odo. At the same time, the synod of Meung, marking the acceptance Odo enjoyed in the entire realm like no other event, also appears as a turning point. From 890 onwards, the overall number of diplomas issued by Odo dropped. This, of course is hardly surprising since after this point the initial visits from nobles to acknowledge the new king and to receive confirmations and grants had taken place and diplomatic activity came down to a more normal level. However, it is worth noting that after 892 visits from the distant regions of the realm, as far as the evidence allows us to surmise, ceased entirely. The conflicts breaking out in 892 and the coronation of Charles the Simple early in 893 appear to have deeply affected the stability of Odo's reign. Nobles and ecclesiastical institutions seem to have laid low, waiting for events to unfold.

#### **The later years**

The same years also saw a massive change in Robert's role at court. During the first years of Odo's reign he certainly occupied a key role in his brother's policies. However, after 892, deep ruptures appeared in the circle of supporters the king had constructed to support his reign. Theoderic had probably already died and Aledramnus, Altmar, Ebolus, Anskeric and Heribert were in open rebellion, leaving only Count Hucbald, whose description as "our *fidelis* in all affairs"387 was undoubtedly not chosen by chance but reflected the contemporary problems of Odo's reign. The king renewed his efforts to increase his base of support, nominating Adalgarius of Autun as his archchancellor, but had to resort back to


387 DOdo 38 (894), 161: *…in omnibus fidelis noster…*

<sup>385</sup> DOdo 30.

<sup>386</sup> The bishops of Laon, Beauvais and Soissons.

Archbishop Walter of Sens when the former was poisoned. Walter certainly carried some political weight, but he did not offer the same connections that Ebolus and Adalgarius had possessed. The weight of influence within Odo's innermost circle therefore shifted massively in favour of Robert, who, in May 893, appeared for the first time in one of Odo's diplomas as "our beloved brother and illustrious count and *marchio,*" intervening on behalf of the abbey of Cormery.388 In half the diplomas issued from 893 onwards, Robert either intervened or acted as petitioner.389 When, in Neustria, Roger usurped Le Mans, it was Robert who led the army to liberate the city and installed a new count.390 Apart from in Neustria, Robert also acted on the king's behalf in Francia, where he played a key role in the king's dealings with Count Baldwin concerning the matter of the abbey of Saint-Vaast. 391 As the *Annales Vedastini* report, Robert acted as mediator between his brother and the count: in 895 he accompanied Baldwin's emissaries on their return to the king and received the keys to the abbey when Odo ordered his men to evacuate it.392 This indicates that he enjoyed the trust not only of his brother, but also of the count. And indeed, when Baldwin finally submitted to Odo, it was again Robert who had convinced him to do so.393 It may be that even earlier he had mediated between the opposing parties of a conflict, this time between his brother and Charles' allies. When, in 894, Odo had made his camp in front of Reims and negotiations were being carried out, it was not he who gave hostages to Charles' party, but Robert, 394 thus acting as a guarantor of Odo's good will and behaviour. During the second half of Odo's reign, Robert had become his brother's right hand.

Other contacts of Odo during the same period were—as far as we can tell mostly motivated by the political events unfolding at the same time. Altmar was reintegrated by Odo's acknowledgment of his position in the north of Aquitaine, as demonstrated by his intervention together with Robert on behalf of the bishop of Poitiers.395 William the Pious, involved in the conflict around the succession of Ramnulf 's son Ebalus Manzer at Poitiers, appears to have been neutralised by the grant of Saint-Julien of Brioude.396 The diploma issued at the request of the bishop of Clermont for the church of Saint-Agricol and Saint-Vital397 was probably also meant to strengthen the ties with the count of Auvergne. Another royal charter issued in July 894 at the request of Bishop Franco of Nevers398 may be read as an


<sup>388</sup> DOdo 33 (28th May 893), 146: *…dilectus noster frater atque illustris comes et marchio..*. DOdo 50 (15th October 893−3rd March 898) also ascribes him the title.

attempt to gain a foothold in the border region between Aquitaine and Burgundy, a contact which could be used to observe the ongoings of both William and Richard the Justiciar.

Richard the Justiciar, despite belonging to one of the most noble families of the realm, the Bosonids,399 brother to Richilde, Charles the Bald's second wife,400 and Boso, the emperor's closest supporter,401 had originally only inherited a few possessions in northern Burgundy. However, his birth had brought him close to the centres of decision making, knowing the most important people and in turn being known by them.402 A first step was his support for the Carolingian cause against his brother Boso, for which he was rewarded with the county of Autun and probably also the abbey of Saint-Symphorien, over which he appears as abbot in a private charter some years later.403 In 887/888, he further improved his political connections by marrying Adelaide, sister of the most powerful man in Upper Burgundy, Rudolf, 404 soon to be king of the same region. Up until 893, his focus appears indeed to have been on his new brother-in-law's *regnum* as well as Provence, where his nephew Louis now occupied the throne.405 A *judicium* held at Varennes, judging that the monks of Gigny were to be restored possessions they had originally received from King Rudolf, but which had been subsequently alienated by a vassal of the queen, shows Richard in the company of Louis' mother Ermengard. 406 It seems likely that on this occasion Richard mediated between the interests of his brother-in-law, Rudolf, and his sister-in-law, Ermengard. 407 Relations with the latter appear to have been equally good: in her son's Louis' coronation charter from Valence, Richard appears as the boy's guardian.408

Around the same time, Richard seems to have started to construct a network of support stretching over northern Burgundy. A number of charters and entries show him in the company of local nobles: a donation made by him, probably at a local assembly, to the abbey of Saint-Bénigne of Dijon409 was witnessed by Bishop Teutbald of Langres, the counts Girbald, Wido and Radulf as well as by Viscount


Madelgaud of Oscheret. 410 Count Wido also accompanied Richard at Varennes, together with Count Ragenar, Richard's nephew and brother of his most important supporter, Manasses. 411 The latter can also be found together with his wife in an entry dating to 890 in the *Liber Memorialis* of Remiremeont, along with King Louis and his mother Ermengard, Richard and his wife Adelaide, as well as William the Pious of Aquitaine.412 By early 893, Richard had become the most powerful noble in Burgundy, a fact that can be deduced from Flodoard referring to him as *princeps*  of Burgundy.413 Up until this point, however, leaving Teutbald aside, Richard's support consisted only of laymen—the bishops of the region seem to have preferred Odo, who they joined at the synod of Meung-sur-Loire. 414

Richard's big opportunity arose when the party around Fulk rebelled against Odo and crowned Charles king. His political connections first paid off just after the rebellion when Richard, William and Altmar joined their forces to see what could be won from the contesters.415 That they acted together at this moment can hardly be surprising. As we have seen, William and Richard already appeared together in the entry from Remiremont three years earlier. Fulk's letter to Teutbald, asking the bishop to inquire about their intentions,416 points to them forming a union during the early 890s. The subsequent agreement with Odo brought Richard the possession of Saint-Germain of Auxerre.417 This, however, did not put an end to his ambitions: in the wake of events in Francia, he now turned to more violent means to expand his influence in Burgundy.

The first victim of his new endeavours was Bishop Adalgarius of Autun, Odo's archchancellor and a key royal supporter in the region, who was poisoned and replaced by Manasses' brother Walo, thus ensuring Richard's control of this see as well as the abbey of Flavigny, previously held by Adalgarius. 418 Soon after, Teutbald of Langres was attacked and blinded by Manasses himself, acting on behalf of



<sup>415</sup> See chapter I.2.5.

Richard, because he had obstructed Walo's installation as bishop.419 Teutbald died soon after, upon which Argrim, who seems to have acknowledged Richard's dominance, returned to the see.420 Finally, Odo's cousin Archbishop Walter of Sens was imprisoned for nine months by Richard and Manasses, only being released upon the king's intervention and the provision of hostages by the people of Sens.421 It is probably at this time that Richard also gained control of the abbey of Sainte-Colombe of Sens422 and became count of Auxerre423 and Troyes.424 By the end of Odo's reign in 897, Richard had become one of the major players in West Frankish politics, a player who needed to be integrated into the king's rule if, after the years of conflict with Baldwin of Flanders and the alliance around Fulk supporting Charles, Odo wanted to restore a stable peace. Odo's last dated diploma from 21st October 897425 indicates that indeed a solution had been found. Richard not only obtained a rich grant of fiscal property for one of his *fideles*, but was also depicted as "our beloved and illustrious count", a combination of epithets that otherwise was only used for Robert. 426

#### **II.5 Conclusion: Continuities and changes**

Thus, the two decades following the death of Charles the Bald were marked by continuities and changes in the networks around the successive rulers. The reigns of Louis the Stammerer and his sons Louis III and Carloman II were shaped by a circle of very dominant nobles. The origins of this circle date back to the last years of Charles the Bald and formed when Louis attempted to create his own bases of support by handing out *honores* to nobles of his own choice. The key actors in this circle are the well-known Hugh the Abbot, Gauzlin, Boso of Vienne, Theoderic the Chamberlain, Bernard of Auvergne, Bernard of Gothia and Conrad of Paris. Soon, however, this circle began to fall apart due to rivalries for influence at court and control of Louis' sons; the rebellion of Boso weakened it further. Moreover, from 879 onwards, Bernard of Auvergne appears to have taken a more individual course, more detached from the royal court, where he was hardly present. This movement was halted, to some extent, by the


reconciliation of Hugh and Gauzlin after Louis III's death, a time that also saw the addition of Theoderic of Vermandois to the group. Nevertheless, by the time of Charles the Fat, the group had narrowed down to only three members, Hugh the Abbot, Gauzlin and Theoderic.

Indeed, by the middle of the 880s, the death or breaking away of almost all of its members—by 888 only Theoderic was still alive—led to a radical change in the political landscape. Even before the deaths of Hugh the Abbot and Gauzlin, Charles the Fat had shown his will to pursue a novel course by the appointment of Ragenold of Maine as commander of the army against the Northmen, a noble apparently not belonging to the previously existing inner circle at court. After the siege of Paris, Charles used the opportunity created by their and other important nobles' deaths to create his own new network of support by installing men of his choice in central positions and tying others to him by handing over *honores*. These were typically not new comers, but had started their ascent, like Geilo of Langres and Odo, under Charles' predecessors. He selected them because they represented the ideal choices to create the maximum effect with the least effort: they were already tied into the local politics and possessed their own networks and connections that the emperor could now use as conduits of his royal power without creating discontent or larger frictions among the West Frankish nobility.

After Charles' death, Odo's own claim was opposed by two different groups allying themselves against him. On the one hand, we find nobles from Francia who had been present at the emperor's court, although without playing any politically significant role, notably the new archbishop of Reims, Fulk, Abbot Rodulf of Saint-Vaast and Saint-Bertin, as well as the latter's cousin Count Baldwin of Flanders. The other group consisted of a network of nobles from Burgundy, centred on Bishop Geilo, one of the key figures in the late Charles' network. On the other hand, Odo's own basis of support was an alliance consisting of the last remnants of the old group around Hugh and Gauzlin like Theoderic of Vermandois, those installed and furthered by the emperor in the Paris region as well as relatives and old acquaintances from Neustria and Francia. This drifting apart of nobles formerly belonging to the inner circle of a ruler's predecessor is at first sight reminiscent of the rivalries breaking out during the reign of Louis the Stammerer. However, there appears to have been a crucial difference. Hugh and Gauzlin had known each other for a long time at Charles the Bald's court and learned to cooperate with each other. When their position was threatened by Louis the Stammerer, they quickly joined with others to form an alliance enforcing their interests. The groups opting for Odo and Wido of Spoleto consisted of key actors within Charles the Fat's network of support and therefore were a parallel to the rivalry between Hugh and Gauzlin. However, while the emperor's court was certainly a centre for political decisions, due to Charles absence from the Western realm it never became a rallying point for the West Frankish nobles. Thus, contacts between nobles from the different regions were much more based on a personal level and were certainly not as intense as before. The inner cohesion of the circle around Charles the Bald

was hence weakened. In addition, the question of succession, open due to the lack of a "natural heir,"427 meant there was no common candidate whose favour the nobles could seek, thus depriving them of another rallying point.

Odo himself did his best to integrate those nobles originally opposing him into his rule. By allowing certain nobles into his inner circle, he used them as a means to connect with other members of their families—his first archchancellor, Abbot Ebolus, for example, was the brother of Ramnulf II of Poitiers—or to gain access to their local networks—as in the case of his third archchancellor Adalgarius of Autun. However, by the crisis years of 892/3 at the latest, weight within this circle was shifting considerably. After the rebellion of several of its key members, the position of Odo's brother Robert became extremely dominant. Access to the king without Robert appears to have been hardly possible—or desired.

Those nobles with less political influence at court than the main actors we have treated so far are also interesting. Under Louis the Stammerer and his sons, this group consisted mainly of nobles cooperating with the circle around Hugh and Gauzlin, most of all Archbishop Frotar of Bourges, the bishops Walter of Orléans and Adalgarius of Autun as well as Count Ansgar of Oscheret. However, next to them, especially under Louis the Stammerer, we are also able to identify a number of nobles whose connection with the inner circle is less apparent. Count Aledramnus and Abbot Wulfard of Flavigny were related to the royal family, Bishop Berno of Châlons was probably a former member of the royal chancellery. These cases demonstrate that, at least on this level, the king also possessed certain liberties to shape the circle surrounding him. In particular, kinship opened the door for those who also had other connections, as is shown by the case of Count Robert of Troyes under Carloman II. This, of course, does not mean that these men formed a counter-weight against the influence of the key actors. Berno of Châlons, for example, seems to have been well connected to them and Wulfard later allied himself with them to enter their midst. In many of these cases, to whom they actually owed their influence at court is uncertain. If Count Wido, who rose under Carloman II, was indeed Count Ansgar's brother, he would have had a direct link to the group around Hugh and might have owed his influence to them as much as to the king.

This group of secondary nobles was more in flux than that of the dominating circle. Wido and Robert of Troyes, for example, disappeared again with Carloman, as did Count Rudolf with his son Pippin under Charles the Fat. However, there is also some consistency within this group: notably, Frotar of Bourges, Aledramnus, Ansgar, Berno and also Bishop Geilo of Langres appear time and again undoubtedly because their links to the court depended not only on the king, but also on their associations with the group around Hugh and Gauzlin. After the death of these two, however, this group also fell apart. Count Ansgar, for example,

<sup>427</sup> Regino, Chronicon 888, 129: *Post cuius mortem regna, que eius ditioni paruerant, veluti legitimo destituta herede, in partes a sua compage resolvuntur et iam non naturalem dominum prestolantur, sed unumquodque de suis visceribus regem sibi creari disponit.*

supported Wido of Spoleto and followed him to Italy where he became *marchio* of Ivrea. Count Wido, to give another example, if it was indeed the same Wido, associated himself with Richard the Justiciar soon after.

Finally, continuities and changes can also be found at the lowest level of contact, that is to say, visits of nobles and representatives of ecclesiastical institutions whose main interest appears to have been obtaining royal diplomas confirming their rights and possessions or to receive donations. Institutions belonging to this group include abbeys such as Solignac, Beaulieu, Vézelay and Saint-Maur of Fossés. All of these institutions seem to have had an especially strong bond to the royal centre and to the unity of the realm.428 This was undoubtedly also true for the visitors from Septimania and the Spanish March, who, due to their location on the border region with Muslim Spain, seem to have tried to found their identity on tying themselves to a larger political and historical entity, that is to say, the West Frankish realm.429 In the individual cases, however, very profound reasons may also have been behind their motivation to travel the huge distances to the royal court. Often local disputes about landed property and rights arose around new foundations in the area430 while the local nobility also tried to gain control over the episcopal sees. For example, in the case of Odo's early diplomas for Septimania, the presence of Count Sunifred and Bishop Ermemirus of Girona was the result of a conflict between the count and Archbishop Theodard of Narbonne, triggering in turn the appearance of the archbishop and his own candidate, Servus Dei.431 The Septimanians thus sought the king because they needed a monarch to defend their own interests against their rivals.432

While contacts with Septimania were strong, those with Aquitaine appear rather limited, focusing only on a small number of bishoprics and abbeys. This concurs with the evidence we have from the earlier 9th century. Aquitaine, despite Charlemagne's efforts to tie it more closely to the crown by installing new counts, bishops and abbots, had a strong tradition of independence, undoubtedly furthered by the practice of Carolingian kings of making it a sub-kingdom for their sons. Diplomas for Aquitanian recipients are thus rather scarce and the contact between the king and the local nobles was ever more attenuated.433 This points us towards another case, however. Burgundy, in contrast to Aquitaine, was among the regions of the West Frankish realm with the closest ties to the royal centre. Its nobles were some of the most influential at court and both churches and abbeys frequently received royal diplomas. The reign of Odo, however, marks a distinc-

429 Remensnyder, Remembering.


<sup>428</sup> Koziol, Politics, 92−93.

<sup>430</sup> Jarrett, Power.

tive change in these relations. The support Wido of Spoleto enjoyed in the area is a sign of the opposition of the local nobility to Odo. The king did his best to open channels of communication by bringing nobles with regional connections, such as Anskeric and Adalgarius, into his inner circle. Here, however, the Robertian had to compete with other men who tried to tie the local nobility to themselves. Wido of Spoleto, of course, was the first; yet more significant was Richard the Justiciar, who built up his own network of support around 890 and started to expand his influence by violent means in 893, thus installing himself as an intermediate layer between himself and the royal centre, a development that we will further explore in the following chapter.

## **III. Networks of royal power: Charles the Simple**

After his "restoration" as king (as his diplomas put it1 ), Charles was confronted with the effects of the struggle for the crown. According to the *Annales Vedastini*, Odo, when he felt his death approaching, called upon the nobles of the realm to serve Charles loyally.2 Odo's efforts to ease the transition of the crown from himself to the young Carolingian indicate that he was far from certain about how his newly established order would hold up against the shifts within the noble hierarchy brought about by Charles' succession. Odo's reign had been marked by the rivalry of the nobles who had supported his own coronation with the group around Archbishop Fulk. While Odo attempted to integrate these men into his rule after their rebellion, Charles' succession opened another window of opportunity for them to renew their rivalry and increase their influence in royal politics in West Francia. Even more, over the past years, Richard the Justiciar had risen in Burgundy, equalling William the Pious in Aquitaine in power. Their integration into the rule would be another difficult task for Charles. This leads to the first of the questions to be addressed in this chapter. How did Richard's new position affect Charles' rule in Burgundy? Or, more precisely, how did it affect Charles' contacts to the nobles of the region and his own movements? In a wider context looking for answers to these questions means to analyse Charles' itinerary as well as his diplomatic activity in terms of temporal and spatial distribution. The second set of questions deals with the problem of how Charles managed to integrate both his old supporters and Robert (as well as Richard and William) into his rule. Or to put it another way, this means addressing both Charles' own understanding of kingship and the importance he attributed to his Carolingian heritage as well as his ability to manipulate the noble networks around him. His rule lasting 25 years already indicates a remarkable capacity to deal with these problems and his ability to adapt to changes brought about by the deaths of key figures within the network, the necessity of integrating new nobles such as after the acquisition of Lotharingia or by shifts under the surface caused by changes of the relations between individual nobles.

#### **III.1 General overview**

A total of 120 diplomas are preserved from Charles the Simple's reign, 27 as originals, 81 in copies and another 13 as *deperdita*. 3 These numbers, although at first

<sup>1</sup> From 8th February 898 (DChS 10) onwards, almost all of Charles' diplomas date not only after his regnal years (*anno … regnante*), but also of those of his reinstatement to power (*redintegrante*). See also Wolfram, Herrschertitel, 71−72.

<sup>2</sup> Annales Vedastini 897, 79: *Qui dum languor per dies singulos incresceret, onmibus rogare coepit, ut Karolo servarent fidem.*

<sup>3</sup> For an overview of Charles the Simple's diplomas (and changes compared to Lauer's edition), see Lößlein, Urkunden.

**Figure 1:** Number of diplomas per year issued by Charles, CC BY 4.0.

glance impressive, are of course problematic in several ways. First, they represent only a fraction of those which were issued originally.4 Furthermore, quite a number of these diplomas lack dating formulae, making it impossible to allocate them to a specific moment or place. In other cases, the charters have been interpolated or even rewritten, meaning that they have to be read with even greater care than usual. Also, this ostensibly impressive number of diplomas appears much less so when averaged over the entire reign. Out of the 120, 116 can be attributed to the years 898−922, averaging 4.6 diplomas per year. In comparison, 36 diplomas5 of Louis the Stammerer are preserved, averaging 24 diplomas per year, 44 of Carloman II6 , resulting in an average of 7.8 and finally 50 in total of Odo or 5 diplomas per year. Louis' reign, of course, is exceptional due to its shortness, but the reigns of Carloman and Odo show that the average number of diplomas per year diminished over the last two decades of the 9th century.7 However, the average is perhaps as misleading as the overall total number of diplomas since neither number reflects the fact that there were concentrations of diplomas at certain moments or periods of time. As can be seen in the chart, there are high concentrations of diplomas during Charles' reign, most notable during the first three years of his rule, from the end of 911 to 912, from 917 to 919 and finally during 921 and 922.

7 This trend was continued under Charles' successors. Kienast, Wirkungsbereich, 546−547 and 552.

<sup>4</sup> See chapter II, introductory remarks.

<sup>5</sup> We omit DDLS 1, 2, 3 and 24 from our count, since the first two were issued under the reign of Charles the Bald, the third is actually Louis' *promissio* and the fourth the treaty of Fouron.

<sup>6</sup> Omitting DDCmII 64, 67 and 75, Carloman's renewed *promissio* and two capitularies while adding a diploma unknown to the editor. See Baudot, Abbaye.

Most of these concentrations seem, at first glance, to be linked to certain events: to Charles' accession to the throne in 898, to the acquisition of Lotharingia in late 911 and to the conflicts with Count Gislebert and the rebellions of the nobles from the West from 919 to 922. However, at second glance, these links are less clear. For example, during the first two years only five diplomas were issued for beneficiaries in Francia8 while thirteen went to Septimania and the Spanish March.9 Only in 900 can a significant rise in recipients from the North be noted10. We will return to this observation later; for the moment, it is worth keeping in mind that, like Odo ten years earlier,11 interest in Charles as king was actually rather limited during the first year of his reign and that contacts with nobles in Francia only intensified after two years had already passed.12 At the end of his reign, Charles' diplomatic activity was already visibly rising in 917 and 918 although the conflict with Gislebert only started in 919. A closer look at the beneficiaries in these cases reveals the issue: out of the thirteen diplomas issued during these years, six are closely linked to the passing of Queen Frederuna, confirming her donations or establishing memorial services for her and Charles. 13 Frederuna is also the cause for a sudden spike of diplomatic activity in 907, the year of their marriage, with two royal charters being closely linked to the event.14 Again, we will return to these diplomas at a later point.

Periods of low diplomatic activity can be distinguished from these high concentrations, especially the years from 902 until the end of 911 (with the exception of 907) and from 913 to 916. By comparing these two periods, we can mark a rise in the average number of diplomas issued per year. For the first period this gives us an average of 2,15 for the second of 3.3,16 an increase that can be attributed to the acquisition of Lotharingia which provided Charles with a greater base of support for his own network and increased the demand for royal charters. And, indeed, six of the diplomas issued during these years went to Lotharingian recipients while the rest went to those from the Western realm, thus giving a new average for the latter of 1.8 diplomas per year. While the acquisition of the *regnum Hlotharii* therefore


had an impact on the overall activity, relations with the West apparently suffered only very little.

#### **III.1.1 Charles' itinerary**

The same problem of having a very limited number of diplomas in comparison to the length of the reign also makes it impossible for us to establish a proper royal itinerary.17 Due to the lack of narrative sources for most of Charles' rule, there are necessarily large gaps in our knowledge about the whereabouts of the king at certain moments. This problem is increased by the fact that a number of diplomas, as already mentioned, cannot be dated properly or lack a mention of the place of their issuance; furthermore, there are also those cases where the name of the place given in the diploma could not be attributed to a modern location. While a study of the details of Charles' itinerary therefore becomes impossible, we can, nevertheless, take a look at the general picture. As before, it is worthwhile distinguishing between the periods before and after the acquisition of Lotharingia. From Charles' diplomas, we can see that he remained almost exclusively within the region between Seine, Loire and Lotharingia. There were, however, some exceptions: a visit to Saint-Benoît-sur-Loire brought Charles to the southern borders of Robert's power base in Neustria18 and two others to Troyes and to Melay close to Richard's in Burgundy.19 Within Francia, the most important palace is Compiègne, where we can make note of no less than nine stays.20 This is hardly surprising since Compiègne had become the most symbolically loaded royal centre in the Western realm, expanded by Charles the Bald and furnished with a chapel shaped after the one Charlemagne had constructed at Aachen.21 Next to Compiègne, the palaces of Verbery, Ponthion, Attigny and Corbeny as well as the *castrum* of Laon appear on Charles itinerary.22


**Figure 2:** Charles' itinerary based on places of issuance of royal diplomas (898–end of 911). (© Fabien Cerbelaud)

Charles can also be seen to have stayed in other important places, particularly Saint-Denis, Reims and Saint-Benoît-sur-Loire23 while the remaining stays were in different *villae*.

After the acquisition of Lotharingia the situation in the West remains unchanged. Charles' activities again appear to have been limited almost exclusively to Francia itself, apart from one exception in 914, when he issued three diplomas

23 DDChS 10 (Saint-Denis), 18 (Reims) and 34−35 (Saint-Benoît-sur-Loire).

near Le Mans.24 What is new is the extension of his activities east of the Meuse, where they cover most of his new realm. This second half of Charles reign saw a diminishing role of Compiègne in favour of other palaces, especially Attigny and Herstal.25 Within Lotharingia, Charles also made use of a number of other palaces, for example Thionville and Gondreville, but also—and more important in symbolic terms—Aachen and Nimwegen.26 Both in the West and in Lotharingia, a number of cities and important monasteries were visited as well, Laon and Metz most of all, but also Soissons, Reims, Noyon, Toul and Saint-Denis27.

Charles' frequent stays at the old Carolingian palaces tie in well with the practices of his predecessors. Compiègne, of course, had been dominant since the times of Charles the Bald. Louis the Stammerer and Odo were both crowned and anointed there,28 Louis III celebrated Easter and Christmas at the palace twice during his short reign,29 Carloman II made it his *sedes regni* once he took over the North30 and all four of them issued various diplomas there.31 Next to Compiègne, numerous other palaces were also in use,32 amongst them Quierzy,33 Gondreville34 and Verbery.35 The differences from Charles' reign, however, can easily be tracked by studying the global picture of his predecessors' diplomas. Louis the Stammerer started his short reign at Compiègne and in Francia, then moved via Paris to Tours from where he joined the pope at Troyes before finally directing himself to

24 DDChS 77−79. The first two of these diplomas, DDChS 77 and 78 (19th June 914), were issued *in villa Ruio*. Lauer proposes that this might have been a mistake made when the copies of the original diploma were produced, meaning that in fact the diploma did not read *Ruio* but *Rivo* which he identified with Rupt in the departement Haute-Marne (Recueil Charles III, 172, n. 1 and 174, n. 1). The third diploma (21st June 914), was issued in a *villa Collega*, identified by Lauer as Collège, departement Sarthe, commune de Sablé (Recueil Charles III, 176 and 321). Unfortunately, Lauer did not reveal his reasons for this identification. My own efforts to confirm the identity of the *villa Collega* with Collège (Sarthe) with the help of the various Dictionnaires Géographiques have revealed the existence of quite a number of places called Collège in the entire North of modern France, yet none show any connection to a *villa Collega* (nor the existence of a *villa Ruio*). Given that all three diplomas were issued within three days, the *villa Ruio* and the *villa Collega* must have been in close vicinity to each other (on average, the king and his entourage travelled 20–35 km a day, Reinke, Reisegeschwindigkeit), which would render the identification of Rupt as *villa Ruio/ Rivo* impossible if we put faith in Lauer's identification of Collège (Sarthe). Doing so would mean that in June 914 Charles left Francia for a journey to the very west of his realm.


**Figure 3:** Charles' itinerary based on places of issuance of royal diplomas (end of 911–923). (© Fabien Cerbelaud)

Lotharingia and back to Francia.36 Carloman II spent the first half of his reign in Burgundy and the Aquitanian borders, campaigning against the rebellious Boso. Once his brother died and he had joined the nobles in the North, he stayed in Francia to fight the Northmen.37 Odo's diplomas reveal him remaining mostly

<sup>36</sup> Bautier, Recueil Louis II, Louis III et Carloman II, XXIV–XXXI.

<sup>37</sup> Bautier, Recueil Louis II, Louis III et Carloman II, L−LVI. Only three diplomas were issued by Louis III (and only one, DLIII 43, of them mentions date and place), rendering the reconstruction of his itinerary based on his diplomas impossible.

within his Neustrian possessions, at Orléans,38 Tours,39 Paris40 and Chartres41 as well as in *villae* along the Loire.42 As becomes clear from these brief summaries, certain shifts took place between the death of Charles the Bald and the time of Charles the Simple. While Louis the Stammerer still travelled over large parts of the realm, Francia, Neustria and Burgundy, Neustria became excluded under his son Carloman II. Odo, of course, is an anomaly in this process. Due to his own power base being in Neustria, under him, Francia became less visited than before. Burgundy saw fewer and fewer royal visits after Carloman had moved to the North. Odo himself only came to the region in pursuit of Charles43 while the latter, once he became the sole king of the Western realm, did not travel further south than Troyes and Melay. The regions south of the Loire hardly saw any visits at all, leaving aside Odo's military campaigns in 889 and 892.44 From the death of Charles the Bald onwards, the royal movements were therefore more and more limited to Francia and only the acquisition of Lotharingia brought a change to this picture.

#### **III.1.2 Contacts with distant regions**

The limitation of the royal movements to a certain region, however, does not automatically point towards a loss of contact with the nobles from other parts of the realm. As we have seen in the previous chapter, contacts with nobles from the various regions remained on a constant level up until the second half of Odo's reign, when contacts with Burgundy in particular, which up until that point had been very strong, started to cease. At first glance, this process did not continue under Charles, with a number of diplomas going to Saint-Martin of Autun, Saint-Germain of Auxerre, Saint-Clément of Auxerre, Saint-Bénigne of Dijon, the churches of Autun, Auxerre and Langres as well as to Richard's right-hand man Manasses during the first half of his reign45 while after the acquisition of Lotharingia other diplomas went to Richard's son Hugh the Black46 and to Saint-Philibert of Tournus.47 At second glance, however, this image changes: with one or possibly two


exceptions,48 all the diplomas going to Burgundy were issued at the request of Richard himself or for members of his family. From the mid-890s onwards, Richard hence acted as an intermediary between the king and the local Burgundian nobility, taking over control of the access to the royal court. This process can be best studied using the *Gesta Pontificum Autissiodorensium*. Bishop Herfrid of Auxerre, who appears three times in Charles' diplomas and is indeed the only noble to request royal diplomas for a Burgundian institution without Richard's intervention, had been installed by Charles the Fat. 49 In 910, his successor Geran, belonging to a family based in Soissons, was the choice of Richard and his partisans,50 even if they still sought Charles' approval. Geran was followed by men originating not only from the area, but also coming from Richard's own abbeys, Sainte-Colombe and Saint-Germain of Auxerre. In their cases, the *Gesta* make no more mention of the king, while the *marchio's* role is increasingly emphasised.51 But Richard's growing power in the area was not only reflected in how bishops were chosen. Both Herfrid and Geran did their best to regain control over estranged possessions. Whereas the former did so by requesting diplomas from the king and bulls from the pope without any mention of Richard, 52 the latter only acted on the explicit permission of the magnate when he went to see Charles. 53

Contacts with Neustria had, as far as we can tell, been rather limited since the death of Charles the Bald, the main recipients being the Saint-Martin of Tours,54 Saint-Aignan of Orléans55 and the respective cathedral churches of both Tours56 and Orléans.57 This practice also continued under Charles the Simple, with one important change. As Richard channelled contact with nobles from Burgundy, in Neustria contact between the king and local elites appears to have been controlled by Robert. This was certainly not entirely new, since Hugh the Abbot had intervened frequently in the royal diplomas of Louis the Stammerer and Carloman II



<sup>55</sup> DChF 143.

on behalf of ecclesiastical institutions along the Loire.58 The diplomas of Charles the Fat for these abbeys and churches were all intervened for by Odo, although the emperor's charter for the church of Tours was also petitioned by its archbishop, Adalard. 59 The next step in this development then becomes evident under Charles the Simple, with direct contact between the king and the local elites ceasing completely. There are, of course, numerous charters for Saint-Martin and Saint-Aignan, yet all of them were issued at the request of Robert, the only possible exception being one diploma for Saint-Serge of Angers under Abbot Raino, which is only preserved as deperditum, neither mentioning Robert nor Raino. 60 The same applies for Charles' only known diploma for a cathedral church along the Loire, Saint-Maurice of Tours, which mentions Robert but not the archbishop—again, bear in mind that the diploma is lost and only referred to in a diploma issued by Louis VII in 1157 or 1158.61 The *marchio*, on the other hand, as becomes evident from his private charters, maintained close relations with the local bishops from Orléans, Nantes, Angers and, most of all, Tours.62 Close contact with Robert instead of the king, however, was not was not only held by the local bishops, but also by the viscounts of the Robertian counties along the Loire: Atto, Guarnegaud, Fulk, Rainald and later Theobald frequently appear witnessing Robert's charters.63 Neustria, or at least the counties and bishoprics along the Loire, seem to have been entirely under Robert's control. Yet, there is one important exception: a diploma for Saint-Évroult in today's western Normandy figures Count Hugh of Maine.64 Given the earlier conflict between Odo and Robert on the one side and Roger, Hugh's father, on the other,65 this is hardly surprising. We will return to this diploma later on, for now however, it is worth noting that, during the early years of Charles' reign, not all of Neustria was under Robert's direct control. After 911, this image changes. Next to bishops and viscounts, counts from the region begin to


<sup>65</sup> See chapter I.3.

appear in Robert's charters, among them Hugh of Maine,66 indicating that Robert had found even broader support by then. Hence, Neustrian cooperation with the king increasingly depended solely on a good understanding with the powerful *marchio* of the Loire counties.

Contacts with Aquitaine, as we have already seen, had always been limited and now became even more so with all of the earlier reliable contacts ceasing completely. The Archbishop of Bourges cannot be found at Charles' court,67 nor did the abbeys of Solignac, Beaulieu or Poitiers receive any royal charters. Two of Charles' diplomas indicate the presence of Aquitaine's most powerful noble, William the Pious, the son of Bernard of Auvergne, at court.68 The first one went to Count Gerald of Aurillac's newly founded abbey in the same town, with William serving as *ambasciator*. 69 Probably around the same time, according to the *Vita Geraldi*, William had, without success, tried to persuade Gerald to abandon royal service and to commend himself to him instead.70 This episode apparently did not affect relations between the two since Gerald continued to support William on various occasions.71 The other diploma sees William acting together with Robert of Neustria, restoring estranged property in the Limousin to the latter's abbey of Saint-Denis, 72 thus casting some light on the connection between two of the most powerful nobles of the realm but not providing us with any further insight into relations between Charles and the Aquitainian elite. Interesting in this context, however, is a lost diploma for the abbey of Conques, which supposedly confirmed the abbey's property at the request of its abbot Rodulf. 73 Abbot Rodulf also appears in the *Vita Geraldi* as being connected with Gerald, acting as his intermediary in a conflict with Count Raymond. 74 Furthermore, another diploma was issued in 919 at the request of Count Seguin of Nevers. 75 Seguin's predecessor, Rather, seems to have wavered between William the Pious and Richard the Justiciar76 while Seguin himself can be seen in contact with William's nephew and successor, William the


Younger: together they witnessed a charter for Cluny.77 Charles' contacts with Aquitaine therefore appear to have been dominated by a circle mostly in close contact with the Williamides. There is one notable exception however. A diploma dating to 90978 figures one of William's opponents at court—Raymond, at that point still count of Nîmes and Albi, but later succeeding his father as count of Toulouse.79 The Tolosan family was one of the Williamide's major opponents in southern Aquitaine, gaining supremacy over Septimania after William the Pious' death.80 Hence contacts with Aquitaine were, as before, tenuous and as far as the few diplomas we have tell us, limited to two factions, the circle around William the Pious and the Tolosan counts.

Septimania itself, apparently the target of these two factions, constitutes a complete contrast to Aquitaine, having, despite its distance from the royal court, one of the highest densities in diplomas over the two halves of Charles' reign. Septimanian elites had, as we have seen, a very particular interest in obtaining royal charters. The number of diplomas issued for nobles from the region is therefore hardly surprising, yet worth a couple of further thoughts. As in the preceding reigns, we find the churches of Narbonne81 and Girona82 among the recipients. Besides these institutions, a number of abbeys from the region also appeared repeatedly before the kings: Saint-Polycarpe, for example, received charters from Carloman and Odo; 83 Notre-Dame of La Grasse and Santa-Maria of Amer from Odo and Charles the Simple;84 while San-Esteban of Bagnolas was granted diplomas from Louis the Stammerer and Charles. 85 For others, their contact with the king appears to have dated back much further. Caunes, represented by its abbot Hilderic, and Saint-Chinian under Abbot Bera had last seen royal charters under Charles the Bald; 86 Elne under Emperor Lothar. 87


The large number of diplomas issued for recipients from Septimania can be explained in part by the interests of the local elites demanding them. Saint-Polycarpe and San-Esteban of Bagnoles, for example, were in a long-term conflict over some property originally given to the latter, but claimed by the former.88 In the case of San-Juan-de-las-Abadesas, Abbess Emma seems to have feared that after the death of her father Wifred, who had founded the abbey, her brother Sunyer might try to estrange its property. Charles' diploma was only one of several measures she took to protect her abbey's property.89 However, while these local interests certainly can serve to explain the individual diplomas, they fall short when it comes to explaining certain other anomalies. With some exceptions,90 all of these diplomas were issued in 898, 899 and 922. The ones from 898 reflect Septimanian elites paying homage to the new king and obtaining royal charters in return, similar to what happened during the reigns of Louis the Stammerer, Carloman II and Odo. In the first diploma in June 898, a certain Theodosius received parts of the fisc, comital rights and immunity in parts of the counties of Narbonne, Roussillon and Bésalu.91 In November, two more diplomas were issued, this time for the churches of Narbonne and Elne, with a confirmation of goods in the former case and generous grants in the latter.92 Both churches were represented by envoys, as the respective (arch-)bishops had not appeared in person.

In both 899 and 922, however, the diplomas for the distant South were issued over a short period of time, about a fortnight in the first case and a week in the second.93 Even more remarkable is that, on both occasions, the charters were given in the same place, a *villa* which had originally belonged to the royal fisc but had been given to the church of Tours by Louis the Stammerer, 94 situated south of Reims, the modern Tours-sur-Marne. The second occasion, in particular, raises questions. From 919 onwards, Flodoard's annals give us a detailed account of events in Western Francia, allowing us to track the king's movements through his realms. In June 922, Charles' reign was in crisis. The king was fighting both Count Gislebert and the Western nobles who had rebelled against him. Both sides gathered their forces on the Marne and were encamped some distance from each other. Flodoard then notes that Charles attacked Reims at Pentecost,95 in 922 corresponding to the 9th of June. In the midst of a military campaign that would prove decisive for his reign, the king then decided to issue a remarkable number of diplomas at the

90 DDChS 60 (La Grasse, dating to 908) and 85 (Bagnoles, dating to 916).


<sup>88</sup> Bautier, Recueil Eudes, 34−36.

<sup>89</sup> Jarrett, Power. On the family, see also D'Abadal, Comtes; Lewis, Development, 109−110 and 143−152 and Vones-Liebenstein, Katalonien, 491−501.

<sup>91</sup> DChS 13.

very place where he had already done so for recipients from the same area 23 years earlier. Why would he do such a thing?

By studying the diplomas issued on both occasions more closely, some further issues attract attention, especially concerning the diplomas for the churches of Elne and Narbonne. The two diplomas for Elne96 and three of those issued for Narbonne97 seem to form a particular group, which in the case of Narbonne also includes diplomas from Carloman II and Odo. 98 With the exception of Odo's charter, all the others appear to have been heavily interpolated when they were transcribed into the cartulary of the church of Narbonne during the archiepiscopacies of Wifred of Cerdagne (1019−1079) and Berengar of Narbonne (1156−1162), rendering them highly suspicious.99 All five of these diplomas, however, share a common phrasing, complaining about "the poverty of [the archbishop's] bishopric, since both his own see and nearly all the churches of his diocese are ruined, to the point that he can never restore them by himself."100 While this may indeed have been the case when Carloman II used the formula for the first time,101 by the late 890s, this period of impoverishment had ended long ago. As the *Vita Sancti Theodardi* notes, the archbishop requesting the charter from Carloman had managed to rebuild the cathedral only a few years later, with the construction work probably beginning in 885 or 886.102 At least in the case of Narbonne, by Charles the Simple's time, the formula was most probably unfounded. The use of the formula therefore must


have had another reason, which is most likely to be found not on the part of the bishops, but on that of Charles. In these diplomas, Charles could portray himself as the restorer and protector of the Church, as true king of the realm. This reading is further emphasised by the grants made in the various diplomas. The first charter for Elne grants, next to some properties, half of various incomes, among them the market and the salines—rights that were also confirmed to the church of Narbonne. The second diploma, issued one year later, was even more generous, granting the cathedral of Elne all the churches of the Roussillon and the Conflent, in addition to a number of other properties. In 899, Charles reinforced the immunity of ecclesiastical servants in the church province of Narbonne, prohibiting any officials from forcing them to attend lay courts and from levying taxes from their possessions.103 One of the most interesting phrasings, however, can be found in a diploma issued in 922 for the church of Girona, donating to the church "the little belonging to our royal power" because of the fidelity shown by its bishop, Guigo. 104 The message was clear: remain loyal to the crown and you will be rewarded with whatever is left. Given Charles' current situation, that was clearly a message aimed at those who so far had remained with him.

Perhaps this is the solution to the question of why Charles would issue such a number of diplomas for distant recipients at such a crucial point of his campaign against the rebel nobles. He did this to reinforce his links with his followers, to demonstrate to them that he was the true king, willing to act in favour of those who supported him, protecting the church and handing out generous rewards. Also present at Tours-sur-Marne were monks from Saint-Thierry of Reims,105 who could have relayed this to Charles opponents, thus ensuring that the message also reached those who might be willing to return to their rightful king. Both the place and the target of the demonstration were partly due to the contingencies of the situation. Tours-sur-Marne was close by; Bishop Guigo, who interceded for every single one of the 922 diplomas,106 had belonged to Charles' chapel before becoming bishop of Girona.107 But the place, time and recipients mirror the circumstances of 899 closely enough for it to be likely that there was more to the issuance of these diplomas. In 922, Charles used the opportunity offered to him by circumstance to reenact the assembly that had taken place in 899, to remind the realm of the beginnings of his reign.

If it is likely that Charles' initiative was behind the issuance of the 922 diplomas, it seems probable that the same is also true for the 899 assembly. Why, for example,

<sup>103</sup> DChS 24.

<sup>104</sup> DChS 120, 285: *Supra quae praefato episcopo Wigoni suaeque eclesiae […] pro remedio animae nostrae ac genitorum nostrorum, ob nimiam etiam fidelitatem quam illum erga nos cernimus habere, largimur perpauca nostrae regali jure competentia potestati…*

<sup>105</sup> DChS 115, 31st May 922.

<sup>106</sup> DDChS 116−120, with the exception of DChS 119 for the church of Narbonne, all name the recipients as Guigo's fideles. Two deperdita, identified by D'Abadal for the abbey of Saint-Quirze, D'Abadal, Catalunya II,2 384 and 472−473, probably also belonged to these diplomas.

<sup>107</sup> HGL V, col. 124−125. Schröder, Synoden, 178−179; Magnou-Nortier, Société, 322−333.

would Archbishop Arnust of Narbonne travel to see the king—from whom he had already obtained a confirmation of his possessions and rights half a year earlier by means of an envoy—to obtain another diploma guaranteeing the immunity of the clergymen of his church province from a king who was unlikely to enforce his order in such a distant province?108 We have already shown that, in 899, Charles used his diplomas to portray himself as the one true king. It, therefore, seems most likely, that Arnust did not seek out the king to demand further confirmation of his rights, but that Charles invited him and other local elites to join him at court. This would have provided Charles with the opportunity for a political *mise-en-scène* allowing him to demonstrate his royalty to the realm. Of course, many of the other diplomas—like the one for San-Juan-de-las-Abadesas—were undoubtedly issued because the petitioners had a viable interest in obtaining a royal charter. But at Tours-sur-Marne, they seem to have only provided a framework for the king's own message. He used the assembly of Tours-sur-Marne to strengthen his rule by demonstrating his willingness to act like a king.

#### **III.1.3 Using diplomas to demonstrate Carolingian legitimacy?**

The case of the Tours-sur-Marne diplomas leads us to another aspect of how diplomas were used to strengthen Charles' rule. Since the influential work of Bernd Schneidmüller on the power of Carolingian tradition,109 it has become the scholarly consensus that Charles was strongly attached to his family's past and especially to the memory of his grandfather, Charles the Bald. 110 Indeed, there are many indications that Charles did use his Carolingian descent to strengthen his rule by emphasising his legitimacy. Most notable are the changes to the dating clauses of his diplomas after the acquisition of Lotharingia, which refer to the event as "attaining [his] larger inheritance" (*largiore vero hereditate indepta*).111 Describing the new realm as his inheritance was, of course, a statement that was only loosely based on the political realities. Lotharingia had fallen to Charles because its leading nobles preferred his rule to that of Conrad I112 and not because of some


special Lotharingian attachment to the Carolingian dynasty.113 Nevertheless, this clearly shows that Charles used his heritage to his political advantage by stressing his Carolingian descent and the claims he derived from it. Moreover, in the *narratio* of his diplomas, Charles hardly ever fails to point out the dynastic links tying him to his ancestors. Direct links are thus established with Charlemagne, his grandfather's grandfather (*atavus*), Louis the Pious, his great-grandfather (*proavus*), Charles the Bald, his grandfather (*avus*), Louis the Stammerer his "begetter" (*genitor*), Louis III and Carloman II his brothers (*fratres*).114 Usually, such mentions are connected to the confirmation of earlier diplomas or services for the afterlife of his ancestors. However, there is a special case when Charles artificially inserted a mention of his ancestors and his relation to them into one of his diplomas. A charter dating to 917, issued for Saint-Corneille at Compiègne, introduces his ancestors Pippin, Charlemagne and Charles the Bald in the sanction clause, referring to sanctions laid down by them under similar circumstances.115 Making such references was, from a purely practical point of view, quite pointless. As Geoffrey Koziol has argued, its only purpose was to demonstrate Charles' descent and place him in a line with his ancestors.116

Yet, to what extent Charles really emphasised his Carolingian heritage is much less clear than has been generally assumed by scholars.117 Describing predecessors as ancestors—which, after all, they were—was a common practice among Carolingian kings and was equally used by Charles' brother, Carloman II, 118 and his father, Louis the Stammerer. 119 Charles' supposed close connection to Charles the Bald, which seems to be based mainly on the large number of his grandfather's appearances in Charles' diplomas—26 out of 120—also crumbles when placed under scrutiny.120 This is certainly a significant ratio and it is far greater than for other


kings mentioned in the corpus. For example, Louis the Stammerer, the second most cited ruler is mentioned fourteen times.121 Yet, given that only 36 diplomas of Louis are known, compared to about 460 from Charles the Bald, Charles the Simple's connection to his father seems to have been much more important than the one to his grandfather.122 It is helpful to compare these numbers to the practice of other kings. For example Charles' half-brother Carloman names Charles the Bald nine123 and Louis the Stammerer seven times124 in 44 diplomas while Louis the Stammerer in turn quotes Charles the Bald fifteen times.125 The ratio of mentions to the total number of diplomas known indeed seems to indicate that Charles' own practice of mentioning Charles the Bald (and his own father) was not more developed than that of his predecessors.126 This conclusion ties in well with another finding. While Charles' and his grandfather's diplomas went, in large parts, to the same ecclesiastical institutions, there are numerous cases where the former fails to mention the latter, most notably in the five diplomas issued for Saint-Denis. 127 Confirming the acts of past rulers—among them Charles the Bald—and installing prayer services for them was certainly important for Charles. However, his practice was not substantially different from his predecessors.

While the observations we have made so far apply to Charles' entire reign, there were some significant changes in his diplomas following the acquisition of Lotharingia. Most prominent is the change of his intitulation to *rex Francorum, vir illustris*, 128 a phrasing derived from diplomas issued by Pippin, Charlemagne and Carloman. 129 When trying to interpret this change in regard to Charles' use of Carolingian tradition to strengthen his rule, however, we should take into consideration at which audience the diplomas were directed. From the little that is known about how diplomas were issued and used, it seems that they were meant


to be presented and probably read aloud to a larger community.130 Yet, who, apart from some specialists familiar with early Carolingian diplomas and, what is more, with diplomas from such a short period of time, would have recognised the allusion made to his illustrious ancestors? Charles dropped the *vir illustris* after only a few months, shortening his new intitulatio to *rex Francorum*, 131 which seems to indicate that if he had indeed tried to use the memory of these earlier Carolingians to strengthen his legitimacy, he had soon recognised the futility of this means of doing so.

The new seal and monogram used by the king in 913 pose another problem in this regard. Both imitate those of Charlemagne, 132 and as visual signs may have been recognised by a wider audience, which may have been able to compare Charles' diplomas with his ancestor's. Hence they may have served the purpose of demonstrating the link between Charles himself and Charlemagne. 133 However, both are isolated cases taken from one single diploma,134 a diploma that appears to have been written not in Charles' chancellery, but in that of Archbishop Ratbod of Trier by a scribe called Theodulf. 135 This would also explain why this diploma resembles the Lotharingian tradition so much.136 Considering that Ratbod had kept his distance to the king up until that point,137 this diploma was probably meant as a symbol of the archbishop's entrance into the king's favour. As it was drawn up in Ratbod's chancellery, we may even question whether the seal and monogram were developed at Charles' own court. Likely, they had been presented to the king by the archbishop, to honour him by letting him appear as his ancestor.

While Charles' diplomatic practice does not indicate that he set out to strengthen his rule by putting special emphasis on his Carolingian heritage, he certainly possessed a highly developed sense of legitimacy. The most revealing source, in this regard, is a part of Witger's *Genealogia Arnulfi comitis.*138 The first part, entirely separate to the second part which is dedicated to the family of Arnulf I of Flanders, covers the descendants of the Frankish emperors and kings and


was "dictated by King Charles, restorer of Compiègne after two fires."139 As Geoffrey Koziol convincingly argues, Witger used an already existing genealogy when he produced his own draft and only made some further additions for Charles' own successors, meaning that the core part was indeed dictated by the king to the canons of Saint-Corneille. 140 The genealogy traces the Carolingian family from its origins, giving the same content as others of its kind. It is only when it reaches Charles Martel that the record changes, becoming more detailed than the mere *genuit*-enumerations of the first five generations. Charlemagne's entry, for example, reads "Emperor Charles fathered Charles, Louis and Pippin, Rotrude and Berta on queen Hildegard, Drogo and Hugh and Rothaid on a concubine."141 The entry betrays three features of the genealogy: a deep knowledge about the family, the inclusion of daughters in the account and, most importantly, the distinction between children born by queens and those born by concubines.142 This detail is found in the entries for Louis the Pious, 143 Charles the Bald144 and Louis the Stammerer145 hence, for the lineage holding the West Frankish realm. The other lineages are traced to their respective ends (as marked by the death of the last crowned ruler146) but are significantly less comprehensive—daughters and illegitimate offspring are omitted here.147 Two of these entries are nevertheless remarkable: "King Carloman fathered King Arnulf "148 and "King Arnulf fathered Louis on Queen Oda, Zwentibold, however, on a concubine."149 The first entry fails to mention the illegitimate birth of Arnulf which is hinted at only by the fact that his mother is left out. The second entry, on the other hand, is the only one outside the West Frankish line that mentions offspring by a concubine—undoubtedly because Zwentibold was, next to his father, the only illegitimate son to be crowned. And while Arnulf 's entry is also the only one to attribute a title to an offspring (i.e. Carloman begat "King" Arnulf), both only mention Arnulf as king, not as emperor.


The genealogy includes a striking entry, dedicated to Charles' own father: "King Louis fathered Louis and Carloman and Hildegard on the so-called queen Ansgarde, and also, posthumously, Charles and Ermentrude on Queen Adelaide."150 Both cases, the treatment of Arnulf and the slight against his half-brothers, reveals an even more finely tuned sense of legitimacy than the distinction between the offspring of queens and concubines already indicated. Arnulf, despite being born of a concubine, is depicted as a legitimate ruler by simply omitting the circumstances of his birth from the genealogy, yet, at the same time, degraded by denying him the title of emperor. The legitimacy of Charles' brothers, on the other hand, is questioned by the remark on Louis the Stammerer's marriage problem151 while at the same time it portrays Charles as their father's true heir because he was born from the legitimate queen.

As argued above, Charles did not emphasise his Carolingian heritage to underline his claim to the throne and his practice of confirming his ancestors' diplomas did not differ from his predecessors. His complicated relations with his brothers, however, seems to be reflected in his diplomas. Carloman II is only mentioned four times,152 Louis III only appears once. While in the latter's case this is not surprising—Louis appears to have issued only a very small number of diplomas—the former's case is different. There is a distinctive overlap in institutions that were granted charters from both kings153 yet Charles only confirmed three of Carloman's diplomas.154 On far more occasions, he clearly excluded him from confirmations of previous charters, such as those issued by his father and grandfather.155 Selective citation was certainly not a practice unique to Charles and certainly not one he limited to his brothers' diplomas.156 Yet in these cases, there seems to be a particular underlying animosity that even surfaces in the dedication Charles made in another diploma, denying them their rank: "for the salvation of our grandfather Emperor Charles and King Louis, our most pious father, and our brothers, namely Louis and Carloman."157 Charles acknowledged his brothers, yet

150 See chapter III.1.3, n. 145. Interestingly, the genealogy omits Charles' third sister, Gisela, who had been married to Count Robert of Troyes and died before 12th December 884. She is mentionned in DCmII 80 as *Gislae sororis nostrae*. See Settipani, Préhistoire, 317 with n. 861 and 862.



<sup>151</sup> See chapter I.1.1.

<sup>154</sup> DDChS 23, 105 and 119.

their having become kings while he had been cast aside left its traces. They were family; their royalty, however, was another matter.

Other recent rulers were neglected in Charles' diplomas.158 Charles the Fat, for example, is only mentioned three times,159 despite Charles issuing a large number of diplomas for the same institutions as the emperor.160 However, once again, Charles' practice does not differ much from his predecessors'. Odo mentions Charles the Fat three, possibly four times161 in 50 diplomas, Arnulf and Zwentibold in their charters for Lotharingian recipients not at all, Louis the Child only once.162 Like the emperor, diplomas from these last three rulers were hardly ever confirmed by Charles: twice in the cases of Arnulf and Zwentibold, once for Louis the Child, 163 although again there are a large number of institutions that received diplomas from several of them.164 However, given that only 21 of Charles' diplomas for Lotharingian recipients are preserved, these numbers may be rather expected than exceptional and hardly differ from the practice under Zwentibold and Louis the Child. 165 Odo's appearances in Charles' diplomas—three confirmations166 and one memorial service167—are different to


the cases above, even if not at first glance. Geoffrey Koziol argued that Charles did not consider Odo a legitimate king.168 This conclusion is based on Charles' practice of marking his predecessors as his relatives,169 hence naturally excluding Odo. However, these expressions can also be found in earlier diplomas170 and tie in well with other expressions common in Louis the Stammerer's and Carloman II's diplomas concerning their royal descent, as we have shown above. The language of Charles' diplomas certainly does not reflect the new situation with a non-Carolingian predecessor, yet to expect that Charles would drop this phrasing only to include Odo into the line seems to be expecting too much. In this context, it is also worth pointing out that the dating clauses of a number of Charles' early diplomas add "in succession to Odo" (*in successione Odonis*) to the normal dating,171 demonstrating that he did indeed acknowledge Odo as his legitimate predecessor.172 The low number of Odo's appearances certainly shows that he was not one of Charles' favourite kings, yet he certainly did not fall victim to a *damnatio memoriae*. 173 Nevertheless, mentions of Odo may indeed have carried a specific weight, given that his brother was one of the most powerful men of the realm and one of the key players at the royal court. We must therefore take mentions of Odo into consideration when moving on to discuss relations between Charles and the *marchio*.

#### **III.2 Breaking it down: Networks of royal power**

The beginning of Charles' reign as sole king is marked as a "renewal" *redintegratio*—of his rule, attributed to the divine will by the official language of his diplomas.174 For the author of the *Annales Vedastini*, his installation on the throne was no less consequential. Since the realm had been his father's, it was


<sup>168</sup> Koziol, Politics, 500−501.

just that it should now be his.175 However, the annals also note that his accession was acclaimed by an assembly of Franks; hence, despite his claim, Charles still needed the general consent of the nobles.176 Odo, already on his deathbed, had tried to ensure this consent by asking the nobles to serve Charles loyally,177 probably in accordance with the treaty concluded between him and Charles earlier in 897. However, as a run-through of who was present at Charles' coronation shows, Odo's death-bed appeal to the nobles did not have the desired effect. Archbishop Fulk was present at the coronation, since the ceremony took place at Reims. Count Heribert I's presence is indicated by the *Annales Vedastini*—he was the reason why Baldwin of Flanders proclaimed his loyalty to the new regime through envoys, rather than coming in person.178 Others, however, were missing: Odo's own brother Robert, Richard the Jusiticiar and William the Pious. Robert's absence was probably due to the Northmen, who, after a settlement with Odo, had spent the winter along the Loire179 and were now devastating the bordering regions before returning to their ships. After they were gone, Robert met the king and was received honourably. Richard and William, on the other hand, appear to have waited for Robert to act. Only when he had acknowledged Charles as king, did they come to court,180 indicating that they did not take it for granted that Robert would agree to Charles' being the new king. This is certainly surprising since Robert, given his importance for Odo's rule, undoubtedly had played a major role in the negotiations of 897. However, as would appear from Richard's and William's hesitance, they seem to have harboured certain doubts towards the actual implementation of the treaty. Hence, the episode points us the initial problem of Charles' reign: to find a way to integrate his old opponents who had been key figures of Odo's rule and the other powerful magnates of the realm who had abstained from supporting him during the previous years and to mediate between their interests and those of his old supporters who still formed the core of his network.


#### **III.2.1 Close associates and allies**

#### **III.2.1.1 The first years**

The first years of Charles' reign were marked by a peculiar situation. As indicated by the *Annales Vedastini*, Robert and Richard were present at the royal court and took part in discussing and planning of affairs of the realm; they were present, for instance, in 900 to organise a campaign against the Northmen.181 During these first years, however, neither of them appeared in royal diplomas; although afterwards, first Richard and then Robert become dominant. Instead, diplomas issued during these years tended to be issued to those who had supported Charles during the fight against Odo. Archbishop Fulk and Heribert I occupied prominent positions within the inner circle around the king. The archbishop resumed his service as archchancellor182 and intervened before Charles on behalf of the monks of Saint-Amand. 183 The count, on the other hand, received a diploma for his abbey Saint-Crépin of Soissons, which emphasises his position as one of the king's most important advisors by naming him "our beloved count and abbot",184 and he took part in the council preparing the Viking campaign of 900.185 His influence, as well as Fulk's, was also visible one year earlier, when he and the archbishop successfully spoke against Baldwin of Flanders, who, soon after he had been forced to give up Saint-Vaast, approached the king to recover his lost lands.186

But also others of Charles' old supporters remained present at court. Bishop Anskeric of Paris appears to have played an important role, becoming Fulk's successor as archchancellor after the latter's murder in 900187 and acting as Charles' ambassador on two occasions. In 899, he represented the king at a meeting at St Goar with Zwentibold and a legation sent by Arnulf. 188 A year later, when the conflict between Charles and Robert had broken out, he negotiated with the latter at Tours.189 Furthermore, the bishop was granted two diplomas for his church.190 The first of these is worth a closer look. It was issued in 907, shortly after Charles' marriage to Frederuna, 191 who herself intervened on behalf of Anskeric. Frederuna was not the only petitioner, though: besides her, the diploma also makes note of Abbess Gisela of Nivelles, daughter of Lothar II and Charles' cousin, Robert of Neustria, the countess Adelaide—either Robert's daughter and spouse of Herib-


ert II of Vermandois192 or the wife of Richard the Justiciar193—the counts Altmar and Erkanger as well as another Robert to who we will turn later on. The presence of Charles' relative and some of the most important nobles of the realm—Robert, of course, but also Adelaide, representative of either Heribert II or Richard as well as Altmar and Erkanger, both of whom were among Charles' closer supporters places a special emphasis on the occasion. The small assembly portrayed by this diploma is nothing less than a portrait of the elite of the realm: the most important people at court, enriched with the presence of another member of the royal family and with Anskeric as the diploma's recipient in an especially honourable position. But the charter also provides us with further insight into the continuing influence of Charles' earliest supporters, to whom not only the bishop, but also Count Erkanger had belonged. This was Erkanger's second appearance in Charles' diplomas. In 901, he had intervened on behalf of a certain royal *fidelis* named Tedricus, 194 in a diploma that demonstrates well how the old associates formed links with those desiring access to the king.

Another example of the continuing influence of this circle and its involvement in important political decisions and the construction of links to other nobles is another diploma issued at the request of Count Hugh of Maine and Count Ecfrid, another of Charles' early supporters.195 It was issued at a liturgically propitious time: delivered on the eve of All Saints Day, it indicates that Charles, Hugh and Ecfrid celebrated one of the most important Christian holidays together at the abbey of Saint-Benoît-sur-Loire—as already noted, this was far outside the region the king usually visited—and at a liminal point between Francia and Maine, on the border of Robert's counties. Indeed, the diploma was issued after Robert had left the royal court in anger196 and marks Charles' efforts to find allies against the powerful *marchio.*197 Hugh of Maine was the son of Rothild, sister of Louis the Stammerer and abbess of Chelles, and Roger of Maine, who had taken Le Mans and thwarted Odo's and Robert's efforts to install their own man, Gauzlin, in the


city.198 Thus, Hugh was Charles' cousin, and in fact the diploma describes him as the king's "beloved kinsman,"199 but, more importantly, he was Robert's enemy.

Others belonging to the old circle around Charles also continued to be present at court. Count Aledramnus intervened together with Bishop Mancio for the latter's church of Châlons200 and Bishop Heitilo of Noyon requested two diplomas for his church.201 Besides Charles' old friends, however, also other nobles were heard at the royal court. Count Altmar, who we have already seen among the nobles figuring in the diploma of 907, was probably also behind the king's charter for his abbey of Saint-Médard of Soissons.202 Furthermore, some of the nobles who had originally supported Odo were now integrated into Charles' rule. The bishops Honoratus of Beauvais and Rudolf of Laon203 were among the first to intervene in a royal diploma, in their case on behalf of the monks of Saint-Denis, for a grant of immunity within their new fortification.204 This was another diploma of distinctive political importance: the abbey, held by Odo after the death of his chancellor Ebolus, 205 appears not to have been passed on to his brother Robert. 206 Odo had made Saint Dionysius his patron207 and was even buried in the abbey. Already in 896, Charles had tried to link himself to the saint when he issued a diploma for the abbey of Salonnes, where relics of Dionysos were held,208 an abbey that only half a year earlier had been restored by Zwentibold to Saint-Denis. 209 In his new diploma, Charles now renewed the connection by claiming the saint as his own special patron210 and indirectly declaring his own overlordship over the abbey where not only Odo was entombed, but more importantly also his brothers Louis and Carloman and, most importantly of all, Charles the Bald. 211 Issued at the abbey itself, this diploma was meant as a political symbol, demonstrating that the king



<sup>208</sup> DChS 7.

was seizing control of one of the central places associated with kingship and that he now had also gained the support of former opponents. While this is Honoratus' only appearance in Charles' diplomas, Rudolf intervened again in 905, together with a Count Odilard, on behalf of the royal notary, Ernustus, 212 thus marking his continuing influence.

#### **III.2.1.2 Shifting balances**

The end of this first period, during which the group of Charles' old supporters dominated the circle around the king, ended with the murder of Archbishop Fulk on 17th June 900. Almost immediately afterwards, Charles' first diploma for Richard the Justiciar was issued,213 followed by Robert leaving the court.214 The close sequence of these events seems to indicate that they were connected to each other. The best way to address this question is to remember Robert's position at court during Odo's last years, as analysed in the previous chapter. In 888, Robert had taken over most of his brother's *honores*, most importantly the counties and abbeys along the Loire as well as Paris and the appertaining military command. During the fight against Charles, he had become the most important noble at Odo's court, acting as leader of his armies and negotiating on his behalf. This certainly influenced the way he interpreted his own position within the hierarchy of the nobles and at the royal court. He developed a strong confidence in his abilities and wisdom, a certain "sageness" that had to be acknowledged and used by the ruler.215

Charles' succession, however, meant that the political balance in the realm shifted. Integrating powerful nobles into his rule did not necessarily conflict with Charles' own perception of kingship. The language of a number of his diplomas underlines his willingness to take counsel into consideration and take his decisions in accordance with those surrounding him.216 However, the position that Robert desired was already taken by the nobles who had supported Charles' claim to the throne from the beginning, the group around Archbishop Fulk and Heribert I of Vermandois. Their presence appears to have made an integration of Robert on the basis of the latter's aspirations impossible. Fulk and Heribert would undoubtedly not have accepted what, in their eyes, must have been a presumption.

<sup>212</sup> DChS 51.

<sup>213</sup> DChS 32, 26th June 900.

<sup>214</sup> The account of the *Annales Vedastini* (900, 81−82) indicates the temporal proximity of both events, placing Robert leaving the court immediately after Fulk's murder and Heriveus' election.

<sup>215</sup> Koziol, Robert, 250−251, from Robert's only royal diploma, issued in 923 (DRoI 1). "Sageness" 251.

<sup>216</sup> Up until the end of 911: DDChS 10, 16 (*cum fidelium nostrorum consensu, venerabilium videlicet episcoporum*), 41, 90 (*suggerentibus quoque una cum dilectissima genitrice nostra Adeleide et regni nostri primoribus, tam episcopis quam comitbus, ceterisque fidelibus*), 49, 108 (*cum consilio regni nostri utriusque [ordinis] principum*), 53, 115 (*a nostris, tam episcopis quam laicis, fidelibus responsa accepimus*), 56, 121−122 (*fideliumque nostrorum benigne consulta suscipimus*) and 57, 124 (*Prefatorum igitur principum cognoscentes consilia esse salubria, eorum acquievimus benigne postulationibus atque communi consensu fidelium*).

Their revolt against Odo had, in part, been motivated by their desire to enter the innermost part of the circle around the king and they had been willing to fight for four devastating years to achieve their aim. At this point, giving up this position and, of all people, for their old opponent, would simply not have been possible. Robert had access to the court and certainly played a major role in the decisions made there, as the *Annales Vedastini* indicate,217 but he was denied the position he had occupied under Odo.

The murder of Fulk, however, meant that a power vacuum was created in the inner circle and that a new balance of power needed to be found. Charles' choice, however, did not fall on Robert but on Richard, for whom now a so-far unparalleled sequence of diplomas was issued at the palaces of Verberie218 and Compiègne.219 In these diplomas Richard was described as "illustrious count and our beloved *marchio*"220 and "illustrious count and *marchio,*"221 culminating in the third description as "our beloved and dearest Richard, venerable count."222 At this point, Robert was probably still present at court. As the *Annales Vedastini* report, it was Richard's most important confidant, Manasses, who made some negative comments about Robert to the king. Robert himself was not present at that moment and only later learnt about these words, taking them as reason to leave.223 We can read this scene as a sign that Richard's influence had already eclipsed that of Robert at court. That Manasses dared to insult him was clearly a sign for Robert that, for the time being, his was a lost cause. Instead of him, Richard now dominated at court, and he and the king celebrated Christmas together at Reims.224

We should, however, not solely reduce this second period to one of rivalry between Robert and Richard. Particularly in 900 and 901, Charles did his best to increase the circle around him by concluding new alliances to keep Robert in check. We have already mentioned Charles' visit to Saint-Benoît-sur-Loire, where he met with Hugh of Maine. His cousin was not the only new royal ally, however. Just before Charles issued his first diploma for Richard, he confirmed a grant for


the church of Saint-Christophe at Paris made by Viscount Grimoard, 225 indicating that he had won one further supporter in Paris, possibly from one of Robert's own men. In November 901, he confirmed the rights and possessions of Corbie at the request of Count Ermenfrid of Amiens,226 brother of its abbot Franco, who had emphasised his independence from Odo by not only displaying his own monogram on the coins minted at the abbey, but also by omitting the name of the king.227 This diploma leads us to a private charter Robert issued in September 900 for Saint-Martin of Tours228 after he had already left the royal court. Among a large number of other bishops and nobles, this charter was witnessed by Bishop Otgar of Amiens, who "confirmed" (*roboravit*) it as *episcophilax*229—two expressions which set him apart in the list of the witnesses. Did Otgar belong to Robert's supporters? While this conclusion comes to mind, we should also point out the presence of Charles' archchancellor Anskeric among those signing the charter, indicating that he tried to mediate between the king and the *marchio*. 230 Otgar might thus also have belonged to the royal embassy sent to Robert. If Otgar was Robert's ally, Charles' diploma for Ermenfrid was not only a symbol of his alliance with the count, but also a sign that he had now taken control of the Amienois and was ready to confront Robert's ally in this area. But even if this was not the case, the diploma, nevertheless, remains exceptional for the early years. It was the first time that Charles referred to a larger group of nobles in a diploma, the "first men of our realm, bishops as well as counts and other *fideles*"231, who (alongside his mother Adelaide) had advised him. This points to a larger assembly at that moment, an assembly that allowed Charles to display his own power and to act as a king by confirming the diplomas of his predecessors: he mentions his grandfather Charles the Bald, Clothar III and his mother, St. Bathilde, 232 founder of Corbie, as well as bishops and popes.233 The assembly at Fresnoy was a demonstration of Charles' kingship, going back to his grandfather and the Merovingians, displaying him as a king who protected the church and listened to his councillors. It was a moment from which he could draw confidence.

In addition to Grimoard and Ermenfrid, we now also find Baldwin of Flanders at Charles' side. Soon after Odo's death, Baldwin had made a new attempt to increase his influence in the north-east of the realm by attacking Péronne. Charles

225 DChS 30, 23rd June 900. Grimoard is otherwise not known. His grant to Saint-Christophe seems to indicate that he was viscount of Paris.

228 DRoI 42, 13th September 900, Tours.

232 On her, see Hartmann, Königin, 82−85.

<sup>226</sup> DChS 41.

<sup>227</sup> Kaiser, Bischofsherrschaft, 601. On the coins, see Doubliez, Monnayage, 283−310, esp. 293.

<sup>229</sup> DRoI 42, 164: *Otgerius episcophilax roboravi.* On the use of Greek words in charters from the region see Jarousseau, Évêques, 374. See also McNair, Development, 57.

<sup>230</sup> Anskeric had already previously served as Charles ambassador at the meeting of St Goar with Zwentibold and a legation of Arnulf in early 899 (Regino, Chronicon 899, 149−147).

<sup>231</sup> DChS 41, 90: *...regni nostri primoribus, tam episcopis quam comitibus, ceterisque fidelibus…*

<sup>233</sup> On the bishops and popes, see Lauer, Recueil Charles III, 87 with n. 2−7.

managed to subdue the count once again and seized Saint-Vaast from him, presenting it to Archbishop Fulk, who exchanged it with Count Altmar. 234 When Baldwin's subsequent efforts to regain his lost *honores* were thwarted by the archbishop and Count Heribert, his men murdered Fulk. 235 Unfortunately, the account of the *Annales Vedastini* ends shortly after, so we only have Flodoard's report of a synod under Fulk's successor, Archbishop Heriveus, during which Baldwin's men were excommunicated for their deed.236 Baldwin, however, was not only spared this punishment, but received a diploma shortly after confirming his new foundation at Bergues-Saint-Winnoc; 237 he was also given control over Saint-Bertin, 238 which had been one of his targets in 892 but then had been passed to Archbishop Fulk by Odo. 239 The disparity between the actions taken against the murderers and those taken against their overlord can be explained by Charles' political situation at that very moment: the outbreak of the conflict with Robert. The king could not risk an open conflict with Baldwin like the year before—on the contrary, he needed all the allies he could possibly muster to keep the *marchio* of Neustria in check.

#### **III.2.1.3 A new balance**

The second period of Charles' rule, marked by Richard's dominance and Robert's absence from court, lasted until 902/903, when the latter returned while the former disappeared from the royal diplomas. Due to the lack of narrative sources, the reasons for this change are almost impossible to assess however a hypothesis can be brought forward based on Charles' diplomas. As we have argued, over the past decade Richard had obtained a position in Burgundy that had allowed him to establish himself as an intermediary between the local nobles and the king.240 The diplomas Charles issued for Burgundian recipients during the first years of his rule seem to indicate that the king acknowledged this state of affairs. While Burgundians did receive diplomas from him, they only did so at Richard's plea. There is one exception, however. In March 902, Charles confirmed the restitution of estranged property at Mailly to the church of Auxerre at the request of its bishop Herfrid. 241 Is this diploma, which does not mention Richard in any way, a sign that Charles intended to revoke the *status quo* and re-establish direct contact between himself and the Burgundian nobility, thus threatening Richard's control over the region?


The diploma for Herfrid was followed by a last one for Richard, delivered four months later in July and granting property to his right-hand man Manasses. 242 In contrast to the earlier diplomas issued at the *marchio*'s request, it does not use the exceptional honorifics such as "beloved" and "dearest"243 but simply addresses him as "illustrious count."244 While this change may indicate that relations between Charles and Richard had come under strain, there is also a hint that this diploma may have been intended to signal quite the opposite. It was delivered at Rueil, a *villa* that had been granted to the abbey of Saint-Denis by Charles the Bald Charles in 875.245 This grant had been made in exchange for a prayer service that included the emperor, his second wife Richilde246 and, for the first time, Richilde's and Richard's brother Boso, underlining the proximity between the latter and Charles the Simple's grandfather.247 Had Rueil been chosen to tie Charles the Simple's relation with Richard to the memory the earlier diploma represented? If we follow these interpretations, in early 902 Charles would have made an attempt to strengthen his own contacts with the Burgundian nobility and thus to weaken Richard's control over the region. This would have led to a deterioration of the relations between him and the *marchio* and, in turn, to efforts to mend these.

Be it as it may, with the diploma issued at Rueil Charles' rule started to transition into its third phase. From April 903 onwards, not Richard, but Robert became "our highly beloved count,"248 the most important noble of the realm, who celebrated important Christian holidays with the king—probably Easter and definitely Pentecost, according to the dates on which the respective diplomas were issued and the one who was received by Charles at the city of Charles the Bald, Compiègne.249 Just like Richard before, Robert was now a dominating presence in royal diplomas, receiving four more250 between 904 and 911 while also making a prominent appearance in the diploma of 907.251 How important Robert had become for Charles is best demonstrated by the first of these diplomas. Issued for Saint-Martin of Tours, it instituted a memorial service not only for Charles himself, but also for


Robert and his brother, the lord and king, Odo. 252 Another diploma again reflects the strength of Robert's new position within the hierarchy of the nobility. Issued for Saint-Denis, it shows William the Pious and Robert acting together.253 Robert, despite being named second in the diploma, is clearly the dominating of the two. While William is described as "reverend count" (*comes venerandus*), Robert is addressed as "count and *marchio*" (*comes et marchisus*), thus underlining his superiority in rank—an impression that is further strengthened by William restoring property to Robert's abbey of Saint-Denis. Hence the diploma marks the acknowledgement of Robert's position by one of the most powerful men of the realm.

Robert's dominance, however, does not mean that the old circle around the king had lost all of its influence. While those we labelled as Charles' allies against Robert before 903 only made single appearances in the royal diplomas, some of his older associates remained notable also in the period after 903. The most obvious example is of course the diploma from 907, in which Anskeric, Erkanger and Altmar appear alongside Robert.254 Furthermore, Rudolf of Laon continued to be present at court, intervening together with a Count Odilard for Charles' notary Ernustus. 255 Two more diplomas for Saint-Médard of Soissons256 were undoubtedly issued at the request of Heribert II of Vermandois, who succeeded his father between 900 and 906.257 Finally, contact with the church of Paris survived the death of Anskeric. His successor Theodulf was given a diploma in which he himself was addressed as "our beloved and venerable bishop,"258 emphasising his importance to the king. This was not his only appearance in Charles' diplomas. In 918, the king confirmed, along with other property, a donation made by the bishop to Saint-Marcel of Paris.259

Even more important was Fulk's successor at Reims, the new Archbishop Heriveus. Heriveus was the nephew of one of Odo's closest supporters, Count Hucbald of Senlis. 260 It was probably due to this relationship that Heriveus became Odo's notary in 894.261 Once Odo died, like Honoratus of Beauvais and Rudolf of Laon, Heriveus joined Charles' side and became the new king's notary under archchancellor Fulk. 262 Not even three weeks after the murder of Fulk, Heriveus was elected as his successor by a number of bishops from the archdiocese of Reims:


Riculf of Soissons, Dodilo of Cambrai, Otgar of Amiens, Mancio of Châlons, Rudolf of Laon and Otfrid of Senlis. 263 The only source indicating that Charles had a hand in Heriveus' election as archbishop is Richer, who reports that he succeeded Fulk by "royal grant."264 Richer's report, however, is based on Flodoard, who does not mention any involvement of the king in the affair while Fulk had done his best to secure the free episcopal election.265 Nevertheless, Heriveus appears to have been Charles' choice. That he had belonged to the royal chancellery even after Odo's death certainly indicates that the king trusted him. But more importantly, Heriveus' first official act as bishop was clearly geared towards Charles' political interests. The same synod that elected him archbishop also decided on the fate of Fulk's murderers, excommunicating them while leaving their master, Baldwin of Flanders, unharmed266—a turn of events that perfectly fit in with Charles' own efforts to gain Baldwin as a supporter in the ensuing conflict with Robert.

Heriveus' election also reveals how well he was connected within the ecclesiastic elite of Francia. Six out of nine suffragans voted for him in his election267 while the anathem against Baldwin's supporters, issued on the same occasion, was also signed by the others, as well as by Archbishop Wido of Rouen and Bishop Angelrannus of Meaux. 268 The same bishops, or in some cases their successors, were also present at the council of Trosly in 909.269 Apart from Trosly, Heriveus also appears to have organised a number of provincial synods, during which they discussed not only spiritual matters, but also the state of the realm.270 While these latter were probably only attended by the suffragan bishops of the church of Reims, Heriveus also kept in contact with important men outside the borders of his province. When Wido of Rouen needed advice in the conversion of the Northmen, he turned to Heriveus.271 In 902, the archbishop of Reims met with Archbishop Hatto of Mainz, one of the leading nobles in the regency for Louis the Child, 272 to confer not only over ecclesiastical affairs but also over those of use for the realm.273 Moreover, the archbishop also appears to have been very well connected in the West Frankish elite outside the church. Given his past in Odo's chancellery, it may well have been


his connections to the Robertians that helped bringing Robert back to Charles' court. His influence within and without the Church is also visible in 920, when, as Flodoard reports, the Frankish counts abandoned Charles. 274 The circumstance that only the counts protested against the king's relation to his intimate Hagano seems to point to Heriveus' control over the bishops of his diocese. Some of them, like Abbo of Soissons and Bovo of Châlons, were close to Charles and certainly did not need to be convinced to remain loyal,275 in other cases however it was probably due to the archbishop that the bishops stood with the king. On the other hand, his connections to the secular nobles become apparent when Flodoard ascribes the settlement found between Charles and the counts to his mediation.276

Thus, Heriveus appears to have been a key figure of Charles' reign, well connected within and beyond his province and deeply involved in the politics of the realm. It is therefore hardly surprising that Charles' choice fell on him to succeed Anskeric as archchancellor after his death in 911.277 Heriveus held the office up until 919 when he was replaced by Archbishop Roger of Trier,278 a development that seems to have left their relation unaffected, as can be seen from the events in 920 as well as from Heriveus' support for Charles in the succession of Bishop Stephen at Liège.279 While there may have been tensions between him and the king in 922,280 the proposed evidence for this is ambiguous. Immediately preceding the breakout of the hostilities, there were negotiations between Heriveus' *fideles* and the rebels.281 Whether this indicates cooperation between Heriveus and the rebels is not clear, however, and it seems at least possible that the intention of the meeting was to win the archbishop as mediator in the ensuing conflict. Soon after, however, Reims did become a holdfast for Charles' enemies. When hostilities broke out, Charles devastated the possessions of the church of Reims along the Meuse. Just before Pentecost, inhabitants of the city stole horses from Charles' army, leading to a futile assault on Reims. Soon after, Robert was crowned king at Saint-Remy. 282 Flodoard does not directly connect Heriveus to these events. He only mentions the archbishop's death immediately after the coronation, making it appear as a divine sign against the justness of Robert's cause and thus indicating that Heriveus at least


tolerated the proceedings.283 However, Heriveus appears to have been sick for quite some time before his death284 and may have lost control over affairs at Reims.285 In this context, we should also note the fates of Heriveus' brother and nephew: shortly after the election of the archbishop's successor Seulf, they were accused of having broken their oaths, imprisoned by Heribert II and Robert and deprived of their property.286 While the nature of their betrayal remains obscure, it seems possible that they had opted for Charles instead of Robert, which may in turn point towards Heriveus' own convictions. In any case, it seems that there was a strong opposition against Charles within the city and the cathedral,287 an opposition over which Heriveus, already on his deathbed, would have had only limited influence.

Robert's integration into Charles' rule in 903 did not mean that the king from then on solely depended on him, but that a balance had been found between him, the circle of Charles' old supporters and the other nobles who had joined this circle after 898. This new-found stability even allowed for the reintegration of Richard into the royal court without threatening Robert's position. In 907, Richard requested a diploma for Otbert, the provost of Langres, appearing at the royal court together with his right-hand man Manasses and Bishop Argrim of Langres.288 Given the date of the diploma, 4th April 907, Charles and Richard seem to have celebrated Easter together at Compiègne, indicating the importance Charles still attributed to Richard and that their relationship had probably not suffered. A year later, Richard received another diploma, this time for Saint-Martin of Autun as "count and illustrious *marchio.*"289 By 907/908, therefore, Charles had successfully created a finely balanced network of support that included all of the factions that had previously struggled for influence at the royal court.

#### **III.2.1.4 Lotharingia**

In late 911, Charles acquired a new *regnum*, Lotharingia. Before focussing on his network of support there and the impact this event had on his relations with the Western nobles, we first need to address some questions concerning the nature of this *regnum*, its borders and how they affected the composition of the Lotharingian nobility. From south to north, Lotharingia's290 western border ran along the Saone, making an eastward turn around Chalon-sur-Saone rejoining the river


eastwards of Gray and then leaving it again to follow the borders of the diocese. North of Langres the frontier crossed the Marne to the west and back, running parallel to the Maas and touching the river north of Mouzon from where it turned to the west, now following the borders of the dioceses of Liège and Cambrai in a westward turn, joining the Scheldt north of Cambrai and following it to the North Sea. Thus, parts of the dioceses of Reims and Langres were part of Lotharingia.291 In the east, the Rhine seems have been the border, with some exceptions: Mainz, Worms and Speyer belonged to the Eastern kingdom while Friesland—the modern Holland including the Kennemerland and Utrecht but not Middle and East Friesland—belonged to the *regnum Hlotharii*. 292 At Zwentibold's time, Verdun seems to have been the most southern city of Lotharingia, with Besançon being contested by Arnulf and Rudolf of Upper Burgundy, the latter being able to assert his control of the episcopal see by 903.293 Northern Alsace seems to have been held by Arnulf, the south by Rudolf. 294 Thus the Rhine frontier remained stable, Alsace and Friesland constituting what Schneider calls "Bruchlinien."295

Six out of ten Lotharingian dioceses, Strasbourg, Basel, Trier, Cologne, Utrecht and Cambrai, owned territory in other kingdoms, with some having over half of their territory outside Lotharingia. Moreover, for the bishops of Strasbourg, Basel and Cambrai, their respective metropolitans were located outside of the *regnum*, at Mainz, Besançon and Reims. The Lotharingian churches were thus strongly oriented towards the East and the West.296 Like the churches, the nobility had close relations with families outside the *regnum*. As Régine Le Jan argues, families originating from the Trier area were also implanted into the Verdunois, along the Middle Rhine, in the Vermandois and the Cambrésis, their connections not only surviving the partition of Verdun in 843 but persisting at least until Ottonian times. Other examples of continuing relations would be the Adalberos, descendants of Count Wigeric of the Bidgau and Count of the Palace of Charles the Simple, who provided an archbishop of Reims and bishop of Laon during the 10th century.297 To this one could also add King Odo's brother Robert, who had held a county in Lotharingia in the early 880s;298 his nephew, Count Megingaud, who was murdered by Count Alberich in 892,299 the Lotharingians who held their

they had either become part of the Western kingdom or of Boso's realm. Much shorter but less precise is Parisse, Lotharingie, 34.


*amicitia* with Odo, although they were supposed to support Charles the Simple in 894300 and, of course, Reginar Longneck.301 For similar phenomena in the East, we could name Count Gerhard, who married Oda, daughter of Count Otto of Saxony, 302 or the Konradiner family, who had a strong influence on Lotharingian politics during the first decade of the 10th century.303

Did these nobles consider themselves to be Lotharingians and were they attached to the idea of the existence of an entity called Lotharingia?304 A middle kingdom was certainly perceived as a separate *regnum*, although, in reality, this entity proved to be an extremely fragile one.305 In fact, it seems that the idea of Lotharingia was created by those foreign to the *regnum*, 306 while the Lotharingians themselves had not forgotten the old origin of the region as part of Francia.307 Hence, doubts concerning the existence of a specific Lotharingian identity are understandable,308 even if some scholars argue in its favour.309 It may best serve us if we follow Régine Le Jan, who acknowledges the existence of a Lotharingian identity, but denies it being consistent and describes it as overall too weak to serve as a framework for the development of an independent kingdom.310 Thus, Lotharingia presents itself not as a monolithic block resting between two kingdoms in the east and west, but as a region with a fragmented aristocracy,311 which latter pursued individual interests by using their relations reaching beyond the borders of the *regnum* into the neighbouring kingdoms.

These individual interests become apparent under the reigns of Zwentibold and Louis the Child, providing us with the necessary background to understand the implementation of Charles' rule over Lotharingia. During these years, four factions, centred on individuals and families, struggled for control over the *regnum*: Archbishop Ratbod of Trier, Reginar Longneck, the Matfrid family and the Konradiner. The first to emerge as the leading figure was Ratbod, who became


Zwentibold's archchancellor312 and appears in four of the king's first diplomas, intervening as "most holy archbishop" and "our beloved" on behalf of Prüm, Saint-Mihiel, Echternach and his own church.313 The following year, however, his influence seems to have suffered considerably. He did not intervene anymore and by the end of the year even lost his position as archchancellor to Hermann of Cologne. 314 This was undoubtedly caused by the rise of Reginar at the royal court, who, at the same time, received the abbey of Saint-Servais of Maastricht, which had been given to the church of Trier by Arnulf in 889.315 Ratbod still remained in contact with the court, however, receiving further diplomas316 and slowly regaining his influence and offices,317 until May 898, when Zwentibold turned away from Reginar and restored Saint-Servais to the church of Trier.318 Under Louis the Child, Ratbod's importance remained unbroken, probably due to an alliance with the Konradiner family.319 Louis again confirmed his possession of Saint-Servais, 320 along with a number of other diplomas for the church of Trier.321 His influence at court, however, is probably best reflected by his continued service as archchancellor for Lotharingian affairs,322 his intervention on behalf of Saint-Evre of Toul323 and his personal receipt of a royal gift.324

Reginar's first appearance at Zwentibold's side is connected to the king's expedition to the Western realm in support of Charles. When tensions arose between the two kings, Reginar left Charles' side for Zwentibold. 325 From this point onwards, his star was on the rise. He intervened in front of the king326 and received Saint-Servais and Echternach;327 the latter, like the former, previously held by


Archbishop Ratbod. 328 The high point of their relations was undoubtedly reached in late 897, when Reginar spent Christmas with Zwentibold and intervened as "our beloved" for Saint-Evre of Toul.329 Reginar's own possessions and *honores* are hard to track330 and widely scattered, but the centre of his influence seems to have been in the Hesbaye and the Hainaut region, as whose count he is most commonly identified but never actually named in the sources. His intervention for an abbey so far from his own possessions underlines his influence at court just like Regino of Prüm's description of him as Zwentibold's "most loyal and only advisor."331 It was probably also due to his influence that, after Megingaud's murderer Alberich had been killed by Count Stephen, Zwentibold intervened and withdrew not only Stephen's own *honores*, but also those of his supporters, to distribute them amongst his own followers.332 Yet this moment also proved to be the zenith of Reginar's influence at Zwentibold's court. Shortly after, the king restored his favour towards Ratbod and, in turn, withdrew all the *honores* and possessions Reginar held within his realm and ordered him to leave the *regnum* within a fortnight.333 Instead of leaving the realm, the count withdrew to Durfos, sought out allies and called Charles into the realm.334 Zwentibold's rule now started to deteriorate. After the negotiations that took place at St Goar early in 899 to settle the conflicts in Lotharingia, the king renewed his efforts to make the count submit, yet not only failed but was also resisted by his bishops, who refused to excommunicate the rebels.335 Rejected by the vast majority of the Lotharingian nobles and without the help of his father after Arnulf 's death on 8th December 899, Zwentibold sought refuge in devastating the country. The Lotharingian aristocracy chose Arnulf 's legitimate son Louis as a replacement king, to whom they paid their homage at Thionville and Aachen. Zwentibold found his end in battle against the Counts Stephen, Gerhard and Matfrid on 13th August 900.336


These latter two belonged to the powerful Matfrid family,337 which had its power centre around Metz with more possessions in the Bliesgau and the Speyergau, and influence over the Jülichgau, the Zülpichgau and the Eifelgau, where the third brother Richer became abbot of Prüm in 899.338 Early during Zwentibold's reign, the family was represented at the royal court by Abbot Stephen of Saint-Mihiel. 339 When Zwentibold went against Count Stephen, the brothers Gerhard and Matfrid supported the latter, in turn losing their *honores*. 340 While it seems possible that they soon reentered the king's favour and supported Zwentibold against Reginar, 341 they turned away from him again after the conference of St Goar—as noted, they killed him in combat. Under Louis the Child, the Matfrids renewed their efforts to extend their influence. Not long after Zwentibold's death, Gerhard married his widow, Oda, daughter of Otto of Saxony and sister of Henry the Fowler, who was to become king of the Eastern realm in 919.342 This brought them into competition with the Konradiner family, who had dominated Arnulf 's last years343 and whose own relative Louis had now become king.344 This family, highly involved in the regency for Louis together with Archbishop Hatto of Mainz and Bishop Adalbero of Augsburg, now also held some of the *honores* the Matfrids had lost in 897, including at least the abbeys of Saint-Maximin and Oeren of Trier.345 In addition to these, Gebhard, the leading member of the Konradiner in Lotharingia, whose dominance at court is not only reflected by his numerous interventions in Lotharingian matters346 but also by him being described *dux* of Lotharingia347 in one of Louis' diplomas, held properties in the Wormsgau and was count of Wetterau.348 Other Konradiner were given *honores* in Lotharingia: Eberhard and Otto were given comital rights along the Lower Rhine while their


brother Conrad the Younger, the later king, received the abbey of Kaiserwerth and became count of the Keldachgau.349 Overall, they seem to have dominated the Middle Rhine between Worms and Cologne.350 The Konradiners' dominance only became contestable when they became involved in the famous feud against the Babenberger351 in 906 and this was indeed the moment the Matfrids chose to strike. Gerhard and Matfrid captured Saint-Maximin and Oeren, but soon enough Conrad the Younger mustered an army in Lotharingia and forced them to retreat to the Bliesgau, where an armistice was concluded until the end of the Easter week. In October 906, during the assembly at Metz, the Matfrids were judged and lost their *honores*. 352 Nothing further is known of the affair, but it can be assumed that they soon submitted and were reinstated into their holdings. At court, however, their influence remained low: not a single one of Louis' diplomas shows them intervening.

Also present at Metz was Reginar Longneck,353 who probably had recovered his *honores* upon the fall of Zwentibold. His renewed political importance is not only reflected by his own connection with the Konradiner,354 but also by two diplomas that confirmed him holding the abbeys of Stavelot and Echternach.355 His big moment seems to have come when Konradiner power in Lotharingia was shaken by the death of Gebhard during the battle of Augsburg against the Hungarians in 910.356 In a private charter he issued after Gebhard's death to exchange goods with a certain Harduin, he was depicted as "count and *missus* dominicus as well as abbot."357 Sending *missi* around the realm to serve as royal legates and thereby as means to relate the royal centre with the local power holders was a practice that, by the early 10th century, had long ceased to exist.358 Of course Reginar's charter was not a royal diploma and therefore does not necessarily reflect his actual standing at court. But he at least claimed to be a *missus*, the link between the nobles of the area and the king. That he could act as such is demonstrated by a number of other charters issued by him, showing him in connection with counts like Wigeric359,


Otbert360, Isaac and Meginhard. 361 We should therefore consider this charter as a sign of Reginar's importance both at court and within the Lotharingian nobility, which placed him in a position similar to the one Gebhard had held until his death.362

#### **III.2.1.5 The early Lotharingian network**

By 911, Reginar appears to have occupied a central position within the Lotharingian nobility, rivalled by Archbishop Ratbod of Trier; while the Matfrid family had lost much of its former influence. It is therefore hardly surprising that Reginar became the centre of Charles' new network within Lotharingia, especially when taking into consideration their earlier contacts before 895 and in 899. As Charles entered his new *regnum*, Reginar moved southwards from his own powerbase to meet the king on his way to Alsace. For the next months he probably accompanied Charles, thus spending not only Christmas but also Easter with the king.363 Reginar was now given control over several abbeys: Saint-Maximin of Trier,364 formerly in the hands of the Konradiner family, probably Chièvremont365 and, once again, Saint-Servais366 of Maastricht, taken from the church of Trier. Ratbod, on the other hand, did not appear in Charles' diplomas until August 913,367 indicating that the most important cleric of Lotharingia kept his distance and waited to see how the events would unfold. Consequently, Charles gave no heed to the archbishop's claims and returned the abbey to Reginar, making it clear that it was the count who was in the dominant position. Ratbod's reluctance to acknowledge Charles was probably also why Saint-Maximin was granted to Reginar, who thus gained a foothold within the heart of Ratbod's territory and ensured that the king could extend his influence over the city.368 Reginar remained Charles' most important supporter in Lotharingia until his death in 915, as signalled by his appearances in five royal diplomas,369 setting him apart from the other nobles not only by the number of his petitions, but also by the use of the *marchio* title,370 otherwise only reserved for Robert and Richard in the West.371

360 Recueil Stavelot-Malmédy I, N° 50, 120−121. 6th April 907.

361 Recueil Stavelot-Malmédy I, N° 51, 122. 1st June 911. Isaac, count of Cambrai, also appears in Louis' last diploma for Lotharingia, DLCh 76, together with Count Conrad the Younger and Count Warner.

362 See Brunner, Fürstentitel 289−290, Schieffer, Kanzlei, 114−115 and Hlawitschka, Lotharingien 193−194, and last Schneider, Suche, 144. Goetz, Dux, 334 argues for an actual royal office.

363 DDChS 69 (1st January 912) and 72 (12th April 912, Easter Sunday).


<sup>364</sup> DChS 69.

The use of the title of *marchio* raises the question of Reginar's actual position within Lotharingia, since it was otherwise only applied to Robert and Richard in the West—two men whose control over Neustria and Burgundy was rather tight. Comparing the applications of this title, we can make note of remarkable differences concerning the epithets accompanying it. In the West, these were substantially more elaborate: "Richard, our illustrious count and beloved *marchio*"372 or "Robert, our most beloved *marchio* and abbot."373 In contrast, the epithets used for Reginar remain on a level no different from those used for other nobles,374 the exception being the otherwise unused "strenuous"375 and the more common "illustrious."376 Interestingly, Reginar is never described as "beloved." We should also note that, in fact, Reginar is never depicted as a *marchio* in diplomas in which he intervenes on his own, but only when he appears with the most important man of the West, Robert. This happened on only two occasions, once at an unknown date after April 912377 and then in August 915.378 Both diplomas were issued for the church of Liège to which Robert appears to have had earlier ties.379 In the first diploma, Reginar and Robert are depicted in the same way as "count and *demarchus*"; in the second, the title of "strenuous *marchio*" is reserved for Reginar alone, while Robert's importance is underlined by the singular description as "our Highness' most loyal executor."380 The fact that Reginar is only named *marchio* when appearing with Robert indicates that the use of the title probably relies on Robert and is then extended to Reginar to emphasise his importance at Charles' court by honouring him on par with the Robertian. As Andrea Stieldorf has remarked, for Reginar "*marchio*" is only used as an honorary title to express a particularly high degree of proximity to the king, not to describe a closely defined official function.381



<sup>372</sup> DChS 32, 68: …*Richaldi*, *illustris comitis et dilecti markionis nostri…*

<sup>378</sup> DChS 81.

Reginar was undoubtedly a very powerful man within Lotharingia, royally endowed with several rich abbeys and therefore having enormous resources at his disposal. Yet, the sources do not indicate that he controlled a large number of counties, enjoyed the functions of a wider military command or had any influence on the episcopal sees in his vicinity382—all of which are applicable to different degrees to Robert and Richard in the Western realm. Thus, Reginar's importance to Charles did not so much stem from his control over significant abbatial resources, but from his personal connections to other important Lotharingian nobles like his sons-in-law, Count Berengar and Count Wigeric, who he introduced to Charles' court,383 or the counts Otbert, 384 Isaac and Meginhard. 385 Charles used Reginar to gain access to his personal network, but Reginar was not an all-dominating figure like Robert in Neustria or Richard in Burgundy.386

Indeed, while Reginar was the most prevalent figure in Charles' diplomas within Lotharingia and was honoured by the *marchio* title, the king's connections to the Lotharingian nobility were by no means channelled through him as in the case of Burgundy and Richard. Charles' initial journey to his new *regnum* shows a large variety of contacts covering most of Lotharingia. His first contact was with Bishop Stephen of Cambrai, 387 who not only received two diplomas but also came with a group of counts, Guarner, Theoderic388 and Letard, the latter probably count of the Moselgau.389 In the south, Charles first passed through the territories of the


Matfrids around Metz.390 While none of the diplomas preserved indicates their presence at Charles' court at that time, it, nevertheless, seems highly probable that were in contact. Next, he met with Bishop Drogo of Toul and Count Ricuin of Verdun391 as well as Abbess Rotrude of Andlau, 392 indicating the extension of his network to the south of Lotharingia. The composition of this group reveals a distinctive shift in the circle of nobles in close contact with Charles in comparison with Louis the Child. The church of Cambrai had been favoured by Zwentibold393 and Ricuin had belonged to his last supporters.394 Both, however, fell out of favour under Louis and do not once appear in his diplomas. Stephen was then granted the right to erect a castle, install a market and the right to issue coins at a place close to the holdings of his family.395 Ricuin probably gained decisive influence over Saint-Mihiel, on whose behalf he intervened twice and was described as "most noble man."396 Drogo's predecessor at Toul, Ludelm, had been close to both Zwentibold and Louis, 397 yet Drogo himself did not appear at Louis' court398 while he received two diplomas from Charles. 399 These examples indicate that Charles sought out allies amongst those whose ambitions had been curbed by Louis. By allowing these nobles access to his court, Charles satisfied their ambitions and therefore tied them to his own rule. This method, however, did not work on all occasions, notably so in the regions described by Jens Schneider as "Bruchlinien," Alsace and Friesland.400 Strasbourg and its Bishop Otbert stood with Conrad401 and so did Bishop Radbod of Utrecht402—the former's church last having been in contact with Arnulf, the latter's with Zwentibold. 403

Different from these cases is that of Bishop Stephen of Liège, the former abbot of Saint-Mihiel. Stephen had been close to both Zwentibold and Louis the Child,


requesting diplomas for both his abbey and his new bishopric.404 Now his church was granted some smaller abbeys on the Meuse and the Demer at the request of Reginar, Robert of Neustria and another Robert. 405 The presence of both of these men indicates how well-connected Stephen was within the highest ranks of the nobility. But what is more, Stephen was also related both to the Matfrid family and Charles himself406 and had long had contacts with the Western realm. Before he had become bishop of Liège, he appears to have aimed for another (unknown) bishopric, using a connection to Archbishop Fulk to gain the see, albeit without success.407 But his relation to Charles himself becomes visible long before 912. At the beginning of his reign, Charles issued a diploma for Saint-Mihiel. 408 While this diploma has to be read in the context of the relations between Charles and Zwentibold and Charles and Robert, 409 the choice to issue a diploma for this abbey was most probably also due to the relation between the king and its abbot—Stephen. 410 Now Charles not only issued a diploma for the church of Liège, but also confirmed a donation made to a monk at Saint-Mihiel, which had initially been granted by Louis the Child on behalf of Stephen. 411 The bishop of Liège is, like Reginar, one example of a noble who preserved his influence at court and was perhaps able to extend it; undoubtedly, very much like Reginar, due to his earlier connections to Charles.

While Reginar and Stephen were easily integrated within Charles' rule, Archbishop Ratbod of Trier, Reginar's old rival, appears to have been reluctant to acknowledge Charles as the new king. Like other Lotharingian bishops, his origins lay in the east of the *regnum*, in his case in Alemannia.412 Maybe it was due to these connections that he remained absent from Charles' court until August 913, when he met with the king at Thionville, on Charles' way back to the Western realm.413 Ratbod probably waited to see whether the new king in the East, Conrad, would be able to successfully challenge Charles. 414 Only when it became clear that the


410 Hlawitscka, Lotharingien, 171−172.


<sup>404</sup> Stephen appears in DZ 8 and DLCh 36 and 50. Saint-Mihiel and Liège received, without him being mentioned, the diplomas DDZ 3 and 24 as well as DLCh 55, 57 and 62.

<sup>405</sup> DChS 65.

<sup>406</sup> DChS 81, 181: …*nostre consanguinitati affinis dilectissimi*…

<sup>409</sup> See below.

<sup>411</sup> DChS 73 and DLCh 36.

Carolingian would prevail did Ratbod seek him out—long after Charles had given a number of abbeys originally belonging to the church of Trier to Reginar. The diploma celebrating Ratbod's entry into Charles' service415 was a carefully produced textual object, written and designed not in the royal but the archiepiscopal chancellery.416 Its preamble referred to Gelasian doctrine, calling upon the king as a partner of the church of God,417 thus laying down the basis of the alliance: as long as Charles protected the Church, the archbishop would stand at his side.418 The diploma also displayed a new monogram and seal, both modelled after examples from Charlemagne and Louis the Pious. 419 These references may have been inspired by Charles' interests in the early Carolingians, as can be seen from the use of the intitulatio *rex Francorum, vir illustris*. At that point, however, Charles had already given up the use of *vir illustris*. 420 Therefore, it may indeed have been the case that monogram and seal had also been made as a gift for Charles in the archiepiscopal chancellery. In any case, they were only used in this one diploma. This newly found concord between Charles and the archbishop was celebrated by another singular event. Ratbod, who was made the king's new archchaplain,421 also acted as his archchancellor, despite this position being already occupied by Heriveus. If he had hoped to regain the position he had held under Louis the Child, that is to say, to act as archchancellor in all cases involving Lotharingian recipients, Ratbod was to be disappointed. The royal Lotharingian chancellery had


<sup>415</sup> DChS 74.

come to an end,422 and there was to be only one royal chancellery—the one under the archbishop of Reims.

#### **III.2.1.6 A king of two** *regna***?**

As we have seen, Charles' initial Lotharingian network contains no surprises but ties in with the one of Louis the Child, extending it to include those nobles who had risen under Zwentibold but lost their position under his successor. Charles maintained close contact with its nobles after his initial journey to the *regnum Hlotharii* from late 911 until 913.423 He returned there for a second visit from the second half of 915 to the first half of 916424 and again in the second half of 917.425 Two years later, in 919, Charles again spent at least several months in Lotharingia in the context of the approaching conflict with Reginar's son Count Gislebert. 426 This conflict and its implications also caused his expeditions in the following years, up until the revolt of the nobles in the West in 922. Charles' itinerary is of course very sparse since the number of preserved diplomas is quite small. Nevertheless, trying to estimate the durations of his visits in Lotharingia and comparing them to his visits in the West, the king appears to have spent less than one third of his time between 914 and 919, a period covering the outbreak of the conflict with Gislebert but not the beginning of Charles' struggle with the West Frankish nobles, in Lotharingia.427 Within this same period, after the initial journey and the following concentration of diplomas, but before the deep ruptures of his reign beginning in 920, eight out of thirty diplomas Charles issued went to Lotharingian recipients, less than one third.428 We have already remarked that Charles reacted to the acquisition of his new realm with an increase of his diplomatic activity while the number of recipients from the Western realm remained more or less stable.429 Both Charles' visits during these years and his increase in contacts with nobles point towards the last phase of his reign, which was marked by his efforts to find a


<sup>422</sup> Schieffer, Kanzlei, 142.

<sup>429</sup> See chapter III.1.1.

balance between Robert, the other nobles from the West and those from his newly acquired *regnum*.

While the numbers of Charles' diplomas and his itinerary certainly help us to understand how the balance shifted after 911, they also limit our view to a certain degree since a distinction between the two *regna* may be too sharp to illustrate the actual realities. As we have already argued, nobles from both realms had interests in and connections to each other. The church of Reims possessed various properties within Lotharingia which became the reason for a conflict with Count Erlebald. 430 Reginar possessed property in the Western realm, Robert was related to nobles within Lotharingia431 and had his own contacts to the Liège area, contacts that became traceable again in 919 when Count Berengar sent a legation to him.432 Richard's family also had interests in the region. Richard's son Boso acquired the estates of his father's sister Richilde, the second wife of Charles the Bald, land held in precaria from the abbey of Gorze,433 furthermore pursued claims on the property of her daughter Rothild434 and later took control of the abbeys Remiremont and Moyenmoutier,435 thus becoming one of the most powerful figures in southern Lotharingia. Finally, nobles from both regions acted together, as in the case of the rebellion against Charles in 922, when Robert's son Hugh concluded an alliance with Reginar's son Gislebert. 436 Hence, noble relations did not stop at the borders, interests and connections were highly intertwined between the two *regna*.

The same image is also provided by the royal charters. Counts and bishops from the frontier area crossed the borders on various occasions,437 but also those living further away from the border followed the king. For example, a diploma issued at Gondreville shows Richard, Robert and Abbot Wicheramnus intervening for the latter's abbey of Saint-Philibert of Tournus.438 Two others were issued at Herstal: one for Abbot Acfred of San-Esteban of Bagnoles, confirming his abbey's rights,439 and another for Robert for Saint-Martin of Tours.440 Reginar also travelled to the West, visiting the king at Attigny where Charles issued a diploma


for Echternach.441 Furthermore, nobles from both *regna* can be seen cooperating in royal charters. The counts Raoul de Gouy442 and Sigard of the Liugau requested a diploma for the church of Cambrai443 and there are of course also the already mentioned diplomas for the Liège, featuring both Reginar and Robert. 444

In this context, it is interesting to analyse the lists of nobles in two of Charles' diplomas noting judgements in Lotharingia, as well as in the treaty of Bonn made with Henry the Fowler. The first diploma, dating to 916, marks the restoration of the abbey of Süsteren to Prüm445 and notes the presence of four bishops, fourteen counts and a number of other witnesses.446 However, among them only one, Raoul de Gouy, came from the Western realm. At the assembly of Herstal in 919 where Saint-Servais of Maastricht was withdrawn from Gislebert and restored to the church of Trier,447 five bishops and eleven counts are accounted for,448 at least four of them from the West.449 The numbers at Bonn are similar, with five bishops and ten counts,450 among whom three were from Francia.451 Within these groups, nobles from the Western realm are underrepresented in comparison with those from Lotharingia. However, the Treaty of Bonn followed a ceasefire452 and did not require the presence of all of Charles' nobles, but rather those directly involved with matters along the border—like the archbishops of Cologne and Trier or the bishop of Utrecht. The judgements were most of all regional affairs involving mainly those nobles concerned with the matter.453 They allowed the king to display the power


<sup>441</sup> DChS 76.

necessary to enforce the decisions taken there, again making the presence of nobles from the West not specifically necessary. In this context we should also take into consideration that Western nobles may have been present but unrecorded on these occasions. For example, Robert is attested to have been at Herstal only a fortnight after the judgement of 919, which makes it likely that he had also attended the assembly dealing with the case of Saint-Servais. 454 Indeed, if Charles had need for the presence of nobles from both *regna*, he was certainly able to call upon them. In a letter to the bishops of his realm concerning the succession issue of the see at Liège,455 he referred to an assembly that had taken place earlier, involving no less than sixteen archbishops and bishops as well as *marchiones*, counts and other nobles from his realm.456 Charles' Lotharingia only consisted of nine bishoprics,457 indicating that there must have also been bishops from the Western *regnum* present, most likely from the archdiocese of Reims. Charles' reference to *marchiones* points towards a large assembly with nobles from both realms since, after the death of Reginar in 915, the only men we know of carrying this title were Robert and Richard.458

As the evidence suggests, there was no clear distinction between the nobles from the two *regna*. Connections at all levels were manifold and it seems that Charles made no distinction between them. The strongest indication of such a policy is the royal chancellery. Under Louis the Child, a specific chancellery for Lotharingian affairs under Archbishop Ratbod of Trier had existed. Charles, as we have seen, put an end to this. Heriveus appears as chancellor in all diplomas, whether they were issued for Western or Lotharingian recipients, and so did his successor Roger of Trier. There are exceptions when the archbishop of Trier appears in the chancellor position already before 919, yet these cases are very few.459 Charles did not intend to make a distinction between the two *regna*: he saw himself as ruling over one single realm.

#### **III.2.1.7 The later years**

With these observations in mind, we can now turn to Charles' network after 913. Robert remained central, with thirteen appearances in royal diplomas,460 as did Archbishop Heriveus of Reims. They were not the only nobles from the West to


<sup>454</sup> The *judicium* dates to Pentecost Sunday, 13th June 919 (DChS 100). Robert received a diploma for Tours on 27th June 919 (DChS 101).

have kept their positions at court however. Count Heribert II intervened together with Bishop Abbo of Soissons and Robert on behalf of the latter's abbey Saint-Germain-des-Prés in 918.461 By that point, Heribert had married a daughter of Robert,462 indicating that relations between the two had become closer.463 However, Heribert was also count of the Mézerais, where the abbey of Croix-Saint-Ouen (that was now given to Saint-Germain) was situated and his presence, as well Abbo's, 464 may have been simply the result of his own involvement in the affair. In any case, there is a second occasion to take into consideration. As Flodoard tells us, when the West Frankish nobles started their rebellion in 922, Heribert was with Charles, accompanying him out of Laon to Lotharingia.465 Being at the king's side at this crucial moment certainly indicates that Heribert belonged to Charles' inner circle. Next to these nobles, Richard's family was still present at court: not only the *marchio* himself466 but also two of his sons, Hugh the Black467 and Boso; 468 although Richard's overall presence at court can hardly be compared to the period just after 900. Furthermore, a number of Charles' earliest supporters seem to have maintained their contact with the king: Erkanger, Ecfrid and Ermenfrid, all of whom appear in the lists of witnesses.469 However, caution is necessary concerning this list of men, since quite a long time had passed between their earlier and their later appearances and it is rather likely that the earlier bearers of the names had died and hence are not identical with the later ones.470 In any case, at least Erkanger and Ermenfrid seem to have been in close contact with the king. Erkanger also accompanied Charles to Bonn, and Ermenfrid intervened at Aachen on behalf of the abbey of Brogne.471 Last among those present both before and after 912 was Anskeric's successor as bishop of Paris, Theodulf. Charles now confirmed his

462 On the marriage, see Schwager, Graf, 68 and Exkurs III, 401−402, against the hypothesis that Heribert's wife was the same Adelaide who appears DChS 57. On this hypothesis, see also Settipani, Préhistoire, 225, n. 242.


<sup>461</sup> DChS 92.

<sup>463</sup> Schwager, Graf, 68.

restoration of property to the church of Paris, as well as a donation made by the bishop to Saint-Cloud. 472

Within Lotharingia, three factions can be made out: those centred on the old rivals Reginar, the Matfrid family and the church of Trier, from 916 onwards represented by its new archbishop Roger. The most influential of these groups was at first the group around Reginar. The count himself received one more diploma before his death in 915 for his abbey of Echternach473 and intervened in a second one together with Robert for the church of Liège.474 His death marked the beginning of the end for his faction's influence at court, with the Matfrid family subsequently on the rise. The first sign of this shift was the first assembly at Herstal in 916, where the abbey of Süsteren was restored to Prüm under Abbot Richer, one of the Matfrid brothers. It has been proposed that this moment should be seen as the beginning of the conflict between Charles and Reginar's son Gislebert since Süsteren was situated in the latter's heartland and might have even been taken from Prüm by him or someone close to him.475 However, the case is less clear than it appears at first glance. Erich Wisplinghoff has remarked on the presence of Gislebert and his brother Reginar at the assembly, yet assumed that they had been forced to give their consent.476 However, the two brothers were by no means the only ones belonging to Reginar's old network present at Herstal. Among the witnesses we also find his two sons-in-law, Berengar, who he had introduced to Charles' court in 912 when they intervened together on behalf of a priest from Berengar's county on Easter Sunday,477 and Wigeric, who had become the king's new count of the palace. Of the first seven counts appearing in the diploma, four can thus be directly linked to Reginar's old faction and two of them, Wigeric and Gislebert, were even named before Matfrid. It is also likely that Count Ricuin of Verdun was related to this group, whose initial appearances in Charles' diplomas were followed by another one for Saint-Mihiel in late 915.478 Ricuin and Wigeric had been in contact already during Zwentibold's reign, when both of them intervened on behalf of the church of Trier.479 After Wigeric's death, Ricuin married his widow Cunigunde, 480 tying him closer to Reginar's family. When the conflict between the king and Gislebert broke out, Ricuin appears to have remained neutral


at first.481 In 921, he entered into conflict with Charles, 482 although he appears to have been forced to give in shortly afterwards.483 His son Otto, on the other hand, seems to have been one of Gislebert's closest supporters, fighting with him against Charles484 and witnessing the count's charters.485 Their earlier connection to Wigeric and Cunigunde, as well as their later support for Gislebert, indicate that in 916 they most probably were also among those supporting the count. Hence, the assembly at Herstal marks a slight rise of influence for the Matfrid family, yet the presence of no less than six of the members of Reginar's circle certainly underlines the political weight they still possessed at the royal court.

Gislebert's position at court, however, seems to have differed significantly from Reginar's. No diplomas mentioning him in central positions are preserved and after 916 he appears only as the usurper of Saint-Servais in the diplomas restoring the abbey to the church of Trier.486 Why did Gislebert not inherit his father's position at Charles' side? The main reason for this may have been the particular nature of Reginar's importance for the king. As we have argued, this importance was based not on the count's possessions, but on the personal network that he offered Charles access to. Gislebert, on the other hand, does not seem to have inherited his father's position in this network. In contrast to his father, only one of his private charters dating to Charles' reign was signed by a witness, by the aforementioned Count Otto, Ricuin's son.487 Richer's description of Gislebert's revolt matches this impression. When the count prepared for conflict with Charles, he distributed landed property among the higher nobles and gifts of gold and silver among the lower ones.488 Richer's account, of course, has to be treated with utmost care, yet, nevertheless, he seems to confirm that Gislebert, in contrast to his father, did not possess a strong network of personal contacts upon which he could rely on. He first had to create one to further his interests. Once this network was created and a *modus vivendi* with the king had been found, Gislebert indeed appears to have gained the same acknowledgement by the king as his father. In the already


mentioned private charter Gislebert signs as "*dux*, count and abbot."489 That this, in fact, reflected the king's own position and was not simply a sign of Gislebert's aspirations can be deduced from the charter's dating line, reading "in the 9th year of the reign of the most pious lord, King Charles"490 and thus not only acknowledging Charles as the ruling king, but also, and in contrast to Gislebert's previous charters,491 attributing him with an epithet in the superlative. The presence of Bishop Baldric of Utrecht, whose brother's wife descended from Gislebert's family,492 at Charles' court during these years493 may also be a sign of the count's growing influence.

From 916 onwards, however, Reginar's rivals were on the rise. Count Matfrid was present at Herstal in 916 and 919 as well as at Bonn in 921. His growing importance is best reflected in his position within the lists drawn up on these occasions. In 916 he was still fourth among the counts; in the other two he was first.494 His brother Richer also rose at Charles' court. He received the diploma restoring Süsteren to Prüm in 916 and three years later, upon his request, Charles renewed the privileges given to the abbey by his predecessors.495 This diploma not only demonstrates Richer's access to the king, but describes the relationship between Charles and the abbey in a very significant manner. In contrast to the diploma he issued in 916, Charles now makes mention of the abbey's founder Pippin and his own relation to him,496 thus emphasising his own connection to Prüm and its abbot. Hence, it is not surprising that, not long after, Richer became Charles' candidate for the episcopal see of Liège in succession to Bishop Stephen, after the previous candidate, Hilduin, declared for Gislebert, 497 thus ensuring that the Matfrids continued to act as a counterweight to the Reginarids in the region. Besides them, Stephen of Liège498 continued to be close to the king. After Charles' initial grant for his church, issued at the intervention of Robert and Reginar, the king restored a forest in 915, again at the request of these two men.499 In this second case, however, there was also another member of the Matfrid family present: Bishop


Dado of Verdun, whose sister had married Count Matfrid. 500 Dado's appearance with Robert and Reginar shows the importance he enjoyed at court, as does his participation in both assemblies at Herstal, where he was named third and second among the bishops.501 Like Dado, Stephen may have also been present on the latter occasion. Unfortunately, however, neither diploma names the bishoprics of those attending and the Bishop Stephen appearing in them may have been his namesake at Cambrai.

Finally, the two assemblies at Herstal also mark the ascent of Ratbod's successor as archbishop of Trier, Roger. The first one in 916 shows him as the first of the participants in the judgements in favour of Prüm.502 More significantly, however, this assembly is also the first of three exceptions prior to 919, in which the archbishop replaced Heriveus as archchancellor.503 While these appearances mark a distinctive influence, his position appears significantly strengthened again in 919, when Charles and the second assembly of Herstal withdrew Saint-Servais from Gislebert and restored the abbey to the church of Trier. Not even a month later, when Charles confirmed the judgement of Herstal in another diploma, Roger also became Charles' sole archchancellor, a position he seems to have kept up until the battle of Soissons.504 The timing of his ascent to this central position was certainly no coincidence. The conflict between Charles and Gislebert did not begin with the assembly of Herstal. On the contrary, the judgement must have been prepared some time in advance, gathering supporters and concluding new alliances against the count. Roger, due to his position at the head of the church of Trier and to the rivalry between him and Gislebert, presented himself as the ideal ally and was therefore bound to the king in two ways: through the handing over of Saint-Servais and through his appointment to one of the most central positions at court, that of archchancellor.

Charles' rule thus saw the continuation of the rivalries between the three most powerful factions, the Reginarids, the Matfrids and the church of Trier. Reginar's death marks the beginning of a shift away from his faction in favour of the


Matfrids; and in 919, when the conflict with Gislebert broke out, increasingly in favour of Roger, although without casting the Matfrids aside. On the contrary, Charles appears to have successfully integrated both factions since nothing indicates any conflicts between the two. However, we should not focus solely on these factions. The diplomas from the two assemblies of Herstal and the Treaty of Bonn contain long lists of participants—bishops, counts and men without title indicating the broad support Charles enjoyed. While, undoubtedly, not all of them had significant influence at the royal court, some certainly did since they reappear several times or can be traced in the royal diplomas. For example, Archbishop Hermann of Cologne features in 916 and 921.505 His involvement in the succession at Liège506 may be explained by the see belonging to his archdiocese, although his own position in the case seems rather dubious. Hermann ordained Hilduin, yet later claimed that he had been forced to do so.507 Hermann's actions were probably due to the influence of Gislebert and Henry the Fowler, who controlled large parts of his archdiocese,508 before he changed his course for Charles' side. Present at Bonn, he also organised a synod on behalf of Charles and Henry in 922,509 which had originally been planned to be a grand meeting involving both kings, yet fell rather short of the high expectations due to the rebellion of the West Frankish nobles which kept Charles away.510 Hermann's actual position at Charles' court is hard to establish. He did cooperate with the king and his position in between the realms probably made him the ideal mediator between the interests of the two kings—hence also his presence at Bonn. Yet, this very same position probably also made him an unreliable ally for Charles, leading to the ordination of Hilduin at a time when the king himself was preoccupied with the nobles in the West.511 While Hermann undoubtedly wielded some influence at court, we should also not overemphasise his position. Others were more important.

Besides the archbishop, we should also note a number of counts. Theoderic, count of Kimmen, may have already been in contact with Charles in 912;512 later he accompanied the king at Herstal in 916 and Bonn while his brother Walker was also present in 919.513 Theoderic, however, also received a royal grant in 922, just after Charles had been forced to abandon his position at Reims against Robert's


<sup>505</sup> DChS 84 and MGH Const. I, N° 1, 1−2.

<sup>506</sup> Hermann asked Hilduin to present himself in front of a synod (MGH Conc. VI, N° 2, c. 7, 47). Pope John X asked Hermann to come to Rome on the same matter. MGH Conc. VI, N° 3, A, 53−55 and B, 55−56.

<sup>507</sup> Flodoard, Annales 920, 4; MGH Conc. VI, N° 2, c. 2, 45.

<sup>508</sup> Zimmermann, Streit, 23−24. On the position of the archbishop of Cologne between Lotharingia and the Eastern realm, see chapter III.2.1.4.

<sup>509</sup> MGH Conc. VI, N° 4, 68.

<sup>510</sup> See the editor's comment, MGH Conc. VI, N° 4, 57−58.

army and his own forces started to disintegrate.514 These frequent appearances indicate that the brothers were close to the king and the 922 diploma is a safe sign that they remained loyal to Charles even in times of crisis. Huntger of Vimeu, who was present in 919,515 was probably less influential, yet at least also had access to king, as is demonstrated by a diploma issued at his request.516 Probably closer to Charles was Count Erlebald of the *pagus Castriciensis*. Like Huntger, he was present in 916,517 shortly after he had intervened together with Ricuin on behalf of Saint-Mihiel. 518 In 920, however, Erlebald entered into conflict with the church of Reims. When he was excommunicated and his situation became desperate, he fled to Charles, yet was killed before reaching him.519 That he sought help from the king points towards close relations—he had access to him and could hope to profit from such a move, even against one of Charles strongest supporters, Archbishop Heriveus. This deduction is further confirmed by Charles' actions after the count's death. The year after, at the synod of Trosly, the king intervened to demand, against common practice, his posthumous reconciliation.520

The connection between Erlebald and Charles already points us towards a last group of nobles that emerged during the second half of Charles reign. The central documents for this group are two diplomas dating to 920 and 921.521 The first one shows the counts Hagano, Raoul de Gouy and Sigard intervening on behalf of Bishop Stephen of Cambrai and his church; the second the bishops Abbo of Soissons and Stephen of Cambrai together with the counts Hagano, Helgaud of Ponthieu and Raoul, supporting a judgement in favour of the king's brother-inlaw, 522 Bishop Bovo of Châlons, and his church. Apart from Count Helgaud, all of these men appear to have held elevated positions within the circle around the king, marked by their frequent appearances in Charles' diplomas—even setting aside their appearances in the judgements and at Bonn. Stephen of Cambrai received no fewer than four royal charters,523 Abbo made two interventions, both of them for important abbeys (Saint-Germain-des-Prés and Saint-Maur of Fossés524), Raoul and Sigard each made another two,525 while Hagano's position as Charles' intimate companion is well known and will be addressed later. A number


of these cases show some of these nobles working together, thus strengthening the image of a group cooperating with each other. For example, one of the diplomas for the church of Cambrai shows Sigard intervening together with Count Isaac of Cambrai; 526 while the diploma for Saint-Maur features Hagano alongside Abbo. 527 Comital intervention *per se* is nothing special in Charles' diplomas. Yet, the number of the appearances of these men is singular.

In contrast to counts, bishops very rarely intervene,528 and Abbo appears to have been in an especially influential position at court. The abbey Saint-Maur, for which he intervened together with Hagano, had been founded by Adelaide's great-grandfather, as Charles' diploma does not fail to mention.529 Even in the 920s, it was closely linked to her family, its abbot Rumald being her kinsman and the abbacy reserved for family members.530 In the diploma, Abbo, Hagano and Rumald are mentioned to have restored and reformed the abbey,531 thereby linking the former two to Charles' family even closer than by their joint intervention and emphasising their proximity to the king. Abbo's intervention for Saint-Germain-des-Prés, on the other hand, shows him acting together with both Robert, its abbot, and Heribert II, in whose county were located the *honores* that Charles handed over to the abbey,532 and thus close to the most powerful man of the realm and another, who would become one of its leading nobles once Charles had been deposed.533 In fact, Abbo, the "excellent" and "most reverent bishop",534 appears to have been one of the central men of the realm. When the Western counts assembled to force Charles to dismiss Hagano as his councillor, they did so at Abbo's see, Soissons.535 This choice certainly was not made at random, but may point to the bishop's central role in these events. Either he had a part in organising the assembly or Soissons was chosen because the nobles expected Abbo to be able to mediate between them and the king—or both. Indeed, Abbo's importance did not end with Charles' rule. The bishop became Robert's archchancellor and, after Robert's death, Raoul's. 536 Both seem to have highly valued his connections to the nobility of Francia and placed some trust in him. But his position did not depend on his connections to the royal court. When Heribert II, after the death of Archbishop Seulf of Reims, installed his infant son in the see, Abbo aligned himself with the


<sup>526</sup> DChS 128.

<sup>527</sup> DChS 108.

<sup>528</sup> Clear cases are only DChS 10, with Honoratus of Beauvais and Rudolf of Laon intervening for Saint-Denis, and DChS 81, with Dado of Verdun supporting a request for the church of Liège.

<sup>529</sup> DChS 108, 260: *…qua continebatur qualiter ipsum monasterium Bego, genitricis nostrae proavus…*

<sup>530</sup> DChS 108.

<sup>531</sup> DChS 108.

<sup>532</sup> DChS 92. See also chapter III.2.1.7, n. 464.

<sup>533</sup> On Heribert II's later career, see Schwager, Graf.

count, agreed to the appointment and obtained the vicariate for the archdiocese for himself.537

Like Abbo, Stephen appears to have been well connected within the nobility. In the first diploma he received from Charles for his church, three counts intervened on his behalf, Guarner, Theoderic and Letard; 538 while in later ones we find, besides Hagano, Raoul and Sigard, Count Isaac of Cambrai, who had already been in contact with Louis the Child and Reginar. 539 Stephen thus provided Charles with access to his own network, with Letard and Isaac also appearing at Herstal in 916 and at Bonn respectively.540 Moreover, one of the diplomas issued for him describes him as "the entire realm's strenuous man,"541 a choice of words that probably points towards his heavy involvement in the daily affairs of the royal court, probably similar to Abbo's. After Charles' deposition, however, Stephen's influence, unlike Abbo's, seems to have declined. A conflict with Isaac ensued that only ended about 25 years later, when Otto I withdrew all comital rights from Isaac and handed them to Stephen's successor, Fulbert. 542 For the time being, Stephen probably was able to keep the count at bay. A synod at Trosly judged Isaac liable to pay a fine for burning down one of the bishop's castles.543 However, Stephen's influence appears to have been on the decline. The judgement of the synod may also have been due to the presence of Heribert II, who also had been attacked by Isaac. More importantly, however, Stephen appears neither in the diplomas of Henry the Fowler nor of Otto I. Only his successor Fulbert seems to have reentered royal favour.544

#### **III.2.1.8 Liberty of choice?**

As we have seen, the network of Charles' later years differs considerably from the one Charles possessed before the acquisition of Lotharingia. Most of Charles' early supporters had died. Archbishop Fulk, Bishop Anskeric and Count Heribert I of Vermandois had all passed away during the first decade of the 9th century and were succeeded by Heriveus, Theodulf and Heribert II. Relations between the king and these successors, however, appear to have differed fundamentally from those with their predecessors. Theodulf was still in contact with the king, yet only played a minor role at court. Heribert II's number of appearances is rather limited and they do not, apart from his appearance with Robert and Abbo, reveal any


specific political weight. Yet he was present at Charles' side in 922, a circumstance indicating that he was close to the king. Heriveus finally did play a central role at court; however, whether his influence was comparable to that of Fulk seems doubtful, especially since the latter's position was occupied by Robert. Within the Lotharingian nobility, Charles appears to have been caught up in rivalries predating his reign, balancing between the factions of Reginar and his son Gislebert, the Matfrid family and the church of Trier.

From the interplay between these different groups, the extent to which Charles was at liberty to choose those around him becomes visible. The old group around Fulk had surrounded Charles since his first coronation in 893 and could build their claim to be close to him on their support during these years. Dismissing them would have meant, especially in the beginning, the loss of Charles' entire support base. This situation changed with Fulk's death, creating at the same time a vacuum of power and an opportunity for the king to decide on his future allies. His choices, however, were limited, the only two options being the powerful Richard or the even more powerful Robert. Here, Charles fell victim to a development that had occurred since the times of Charles the Bald: the rise of the *marchiones*. In Lotharingia, a similar situation had evolved. Here as well, Charles had to find a balance between the different claims to have access to the circle surrounding him. The circumstances were already in place and the king had to work within them by balancing between the different factions after the initial alliance with Reginar had ceased to exist with the latter's death.

This, however, is not to say that Charles did not have any agency when it came to choosing those around him. After the acquisition of Lotharingia, a new group of nobles arose, composed of men from both *regna*, most of whom appear to have had a very close relationship with Charles. The nature of these relationships may indicate that their position at the king's side was less due to their actual power and more the result of the king inviting them to his court. Within this group, Hagano is of course the prime example of a noble being elevated by the king. But likewise, Abbo may also have owed his position to the king. Nothing is known about him or how he became bishop of Soissons, and so the proposition that he was Charles' choice for the see must remain speculation based on his later influence at court.545 Yet, he was only one example of how the king could use episcopal succession to install men of his choice in key positions around the realm. Three, possibly four, men belonging to the royal chancellery or chapel are known to have become bishops, and the sees they occupied were spread throughout the realm. Heriveus, of course, became archbishop of Reims. Herluin, who succeeded Heriveus as notary in the

<sup>545</sup> Lauer, Recueil Charles III, 258, identifies Abbo as a relative of Charles' mother Adelaide, which would make it quite possible that the king was indeed involved in his appointment. This, however, appears to be based on a misreading of DChS 108. See Werner, Nachkommen, 434, n. 19 and Dufour, Recueil Robert et Raoul, XXV, n. 6. Against this reading Koziol, Politics, 524, although without giving any argument.

royal chancellery,546 might be identical with the namesake who became bishop of Beauvais during the 900s. At Girona, Guigo, a man who had been educated at the royal palace, became bishop in 908.547 And finally, fourteen years later, Gauzlin, up until this moment Charles' notary, was appointed bishop at Toul548 after the see had been vacant for some time. As at Reims and at Girona, where Archbishop Arnust undoubtedly played a central role in the choice of Guigo,549 here as well we can note local cooperation with the king. After Bishop Drogo's death,550 the canons of Toul sought out Charles at Duisburg. Soon after, the king issued a diploma in their favour and about a fortnight later Gauzlin was anointed.551 In two other cases, relatives of Charles became bishops. Frederuna's brother Bovo became bishop at Châlons552 and in 913 at Strasbourg, Gauzfrid, one of Charles' nephews, succeeded Bishop Otbert, yet died soon after.553 In both cases, the sources do not provide details of the circumstances of their appointments. However, their connection to Charles makes it highly likely that he was involved.

Episcopal successions did not occur very frequently and, given the different groups that struggled to raise their own candidate to a see, the few occasions in which we can actually trace royal influence may not be surprising. Two cases, however, do betray the particular interests of the king in having men of his choice appointed. The first was at Strasbourg, where, after the death of Gauzfrid, a noble Lotharingian called Ricuin554 became bishop. As the records of the synod of Hohenaltheim in 916 tell us, this had happened without the consent of the metropolitan of the archdiocese, the archbishop of Mainz. Ricuin was requested to present himself at the synod yet failed to do so555—did he, like his predecessor, belong to Charles' supporters; and was he therefore opposed by the elites of the Eastern realm? Such an interpretation appears, at least, not to be unfounded, making it likely that he was one of Charles' candidates.556 More obvious is the case of Liège after Bishop Stephen's death. As Harald Zimmermann has argued,557 it seems that at first the clergy and people of Liège elected Hilduin, assuming that he was an acceptable candidate for the king. Charles, however, appears never to have confirmed Hilduin, himself at that moment being occupied by the problems with the West Frankish nobles. Instead of waiting for the confirmation, Hilduin appears


to have allied himself with Gislebert and Henry the Fowler, both attempting to exploit Charles' momentary weakness to extend their influence in Lotharingia. When the king had managed to stabilise his position however, Charles did his best to remove Hilduin and replace him with Abbot Richer of Prüm, calling upon a general assembly, initiating a synod and finally also involving Pope John X in the matter.558 The king's effort speaks volumes about how seriously he took such cases and how far he would go to protect his interests if he felt that they were threatened.

Hence, episcopal elections were a field used by Charles to increase his influence and a field where his agency in installing men of his choice becomes apparent. The choices he made on these occasions reveal the considerations he had to take into account. Richer at Liège belonged to the Matfrid family and the conflict between the king and Gislebert ensuing around the bishopric certainly overlapped with the rivalry between the two factions. By choosing Richer, Charles made certain that he had the support of the Matfrids against Reginar's son and his supporters. In the other cases, the king probably did not install his candidates against the resistance of the local elites. This, undoubtedly, also applies to the circle surrounding Charles and the choices the king made in who he invited there. Some, like Robert, were able to claim access to this circle due to the position they held within society; others, like Fulk and Heribert, due to old connections or, like Bovo, because they were relatives; while, finally, another group had access to this circle due to their personal connections to the king. These different categories did not exclude each other and there would certainly have been cases when a noble could claim a position within the circle based on his own power as well as his personal relation to the king. However, as the problems surrounding the integration of Robert of Neustria again demonstrate, the question of access to the circle was not limited to the direct relations between the king and the respective nobles. When making his choice, Charles needed to take into consideration the interests of those already within the circle. In the case of Robert, Archbishop Fulk's interests were opposed to giving the *marchio* the position he desired and therefore his integration only succeeded after Fulk's death. When the Lotharingian nobles needed to be integrated under Charles' rule, it was, in turn, Robert's interests that were at stake. The king's reaction to this becomes clear in the two diplomas that show Robert and Reginar intervening together on behalf of the church of Liège, with Charles carefully balancing between the two men. Hence, the network surrounding Charles was a fragile construction dependant on his ability to mediate between the different interests of the various nobles. Integrating new men into the network was therefore always a difficult and dangerous endeavour, threatening to offset the balance between the individuals and factions already present within the network. We will have to keep this in mind when we now turn to the role his family and friends played in royal politics.

<sup>558</sup> MGH Conc. VI, N° 2, c. 2, 45 and c. 7, 47 as well as the letters A and B, 53−56 with Zimmermann, Streit, 26−27.

#### **III.2.2 Family and friends**

#### **The queens**

When Charles became king, the opportunities offered by his family connections were rather limited. Of his five siblings, his half-brothers Louis and Carloman, as well as his sisters Hildegard, Gisela and Ermentrude, three were already dead,559 leaving only one niece, Ermentrude's daughter Cunigunde and more distant relatives. Hence, the main role at court fell to Charles' mother Adelaide and, after her, to his wives Frederuna and Eadgifu and their relatives. At court, the queen occupied a central role.560 According to Hincmar, she was in charge of the royal household and had access to the royal treasury, a position that connected her to the dealings at court.561 Given her position at the king's side, the queen was considered to be one of the most influential members of the royal court, expected to act as a mediator and to intercede in front of the king on behalf of others, most importantly her own family and friends.562 Depending on her character and abilities, a queen could become one of the key figures of the realm, strongly influencing its affairs.563 Interventions by queens, however, were rather rare compared to the overall number of royal diplomas—with the notable exception of Charles' mother Adelaide. 564 From 898 until her death in 901565 she intervened no fewer than eleven times, featuring in only slightly less than a third of the diplomas issued during these years.566

Should we take the high number of her appearances as a sign of her special importance to Charles and of her playing a key role at court? The epithets used to describe her are the common ones used for queens, "dearest," (*carissima*) "most beloved" (*dilectissima*) and "sweetest," *(per)dulcissima*567 with the exception of the "venerable" (*venerabilis*),568 which might indicate that Adelaide had entered a convent.569 These diplomas went to a large group of recipients, containing likewise


abbeys, churches and laymen. On all but two570 of these occasions, Adelaide appears together with other nobles, either petitioning or intervening alongside her. It is remarkable that none of these interventions were for relatives; of friends, only Charles' early supporters Count Erkanger571 and Count Ecfrid appear, the latter together with Charles' cousin Count Hugh of Maine. 572 This last diploma, however, is a highly political one, marking the alliance between the king and Hugh against Robert of Neustria.573 This may point us towards a reason for Adelaide's frequent appearances, as there seems to be a pattern behind them. Her first intervention after 898 saw her together with bishops Honoratus of Beauvais and Rudolf of Laon, two former allies of Odo. 574 In 900, she appears in the first diploma issued for Richard the Justiciar575 and in 901 she supported Count Ermenfrid of Amiens.576 In all of these cases, the diplomas issued mark newly forged alliances. In this context, it is interesting to note that Adelaide only intervened on Richard's behalf once, although four more diplomas were issued on behalf of the *marchio*  shortly afterwards.577 Were her appearances limited to these occasions? At first glance, the diplomas for the recipients in Septimania seem to speak against such a conclusion. Adelaide intervened twice for the church of Narbonne, in both 898 and in 899,578 and supported another petition of Archbishop Arnust of Narbonne for a *fidelis* called Stephen and his wife Anne. 579 However, the latter two of these cases took place at the assembly of Tours-sur-Marne, an event of special political importance for Charles that gave him the opportunity to present himself as king in front of a greater audience.580

All in all, it seems that Adelaide only intervened at moments of great political significance, when a certain message needed to be conveyed, giving her appearances a strong symbolic weight. The question remains, however, whether this symbolism was a reflection of her political importance. None of the diplomas names Adelaide as queen. Instead, she is presented as "our mother" (*genitrix nostra*),581 placing the emphasis on her personal connection to Charles. 582 However, her ap-


<sup>570</sup> DDChS 11 (for Saint-Mihiel) and 22 (for San-Juan-de-las-Abadesas).

pearance would only have had significant value, if she occupied a central position at her son's side and if this was generally known and acknowledged. Therefore, it seems plausible that her frequent interventions were indeed a sign of her importance within Charles' inner circle. This allows us to cast some further light on the changes that the royal network underwent during the first years of his reign. The death of Archbishop Fulk in 900 had not only deprived Charles of his most central advisor, but also provided him with the opportunity to fully integrate Robert into his rule by granting him the position formerly occupied by the archbishop. However, the king's choice did not fall on Robert but on Richard, with Adelaide intervening in the very first diploma celebrating this new alliance. It seems reasonable to assume that she, although for different reasons than Fulk, was likewise opposed to Robert: shortly after her death, relations between Charles and Richard appear to have started to deteriorate rapidly, leading to Richard leaving the court and finally to the integration of Robert. 583 With Adelaide's death, opposition to Robert at court seems to have come to an end. Although his mother's death certainly meant a loss for Charles, it did increase his liberty to pursue his own policies towards Robert.

Several years after Adelaide's death, Charles married Frederuna. 584 Not much is known about her family. The argument that she was a sister of Mathilde of Saxony, Henry the Fowler's wife,585 has been convincingly refuted by Eduard Hlawitschka.586 Frederuna has also been seen as a Lotharingian noblewoman, based on the assumption of a connection through kinship between her and Charles' later intimate, the Lotharingian Count Hagano. 587 This hypothesis is based on two of Charles' diplomas for the canons of Compiègne, both issued after Frederuna's death, which see Hagano intervening on her behalf588 and included in prayer services with the king, Frederuna and her brother Bovo. 589 However, as Philippe Depreux argues, this is circumstantial evidence and falls short of proving anything, especially when considering the case of Bishop Stephen of Liège.590 The latter

Otto I's mother Mathilde, who is often addressed as such in her son's diplomas (e. g. DO I 186, 228 and 328).


was addressed in another diploma as "our most beloved kinsman by marriage,"591 indicating that he was considered as belonging to Charles' family—something that is never done in the case of Hagano, despite his frequent appearances.592 The connections of Frederuna's brother Bovo may provide more clues. As Flodoard tells us, in 956, Bovo's nephew Berengar became Bishop of Cambrai. 593 This same Berengar is described by the *Gesta Episcoporum Cameracenisum* as a relative of Otto I. 594 Bovo's East Frankish roots may have also been the reason why his death was noted by the Merseburg necrology.595 In this context, another hypothesis, first proposed by Reinhard Wenskus, is worth making note of. Allegedly, Bishop Bovo of Châlons was identical to a monk named Bovo at Corvey.596 And indeed the *Catalogus abbatum Corbeiensium* does mention a *Bovo episcopus* for the abbacy of yet another Bovo, Bovo II (900−916).597 These years coincide with the beginning of Bovo's episcopate at Châlons. According to the episcopal catalogue of the church of Châlons, after Mancio's death in 910, a certain Letoldus occupied the see for four years,598 meaning that Frederuna's brother Bovo became bishop around 913. However, whether this Bishop Bovo really belonged to the same family as the abbot, the Ekbertiner, as Wenskus proposes, seems rather uncertain, with the shared name Bovo apparently being the only reason for this hypothesis.599 What seems clear is that Bovo and, therefore, also Frederuna, descended from an important Saxon family, somehow related to the Ottonians—this may have been the Ekbertiner or the Popponids, in whose family the name Bovo was also common;600 but this question must remain open.

Two diplomas from 907 mark the marriage and the ascent of the new queen. The first, dating to 19th April, tells us that, according to the law and with the consent of his advisors, Charles had taken a young noblewoman called Frederuna for his wife and made her his *consors regni* ("partner in rule").601 With this diploma,


Charles granted his bride a dowry,602 consisting of two palaces603 belonging to the royal fisc: Corbeny, including his own foundation of Saint-Pierre, which he had constructed for the relics of Saint Marcoul only a year before;604 and Ponthion. The second diploma, issued a month later, saw Frederuna intervening together with Abbess Gisela of Nivelles, the daughter of Lothar II, Robert of Neustria, a Countess Adelaide, the counts Altmar and Erkanger as well as the *fidelis* Robert on behalf of bishop Anskeric of Paris and his Church, granting the church the abbey of Saint-Pierre of Rebais, previously held by the bishop.605 As we have already discussed, this diploma seems to be a kind of who-is-who of the realm, displaying Charles' family and his most important advisors. The use of the *consors regni*, a very uncommon expression for West Frankish diplomas,606 as well as the presentation of Frederuna within the circle of the most influential nobles of the realm, both indicate that Charles intended to let Frederuna partake in the political affairs of the realm.

Early in the 10th century, two new *ordines* were created dealing with the coronation of the queen.607 One of them derives the position of the queen from her marriage to the king by pointing out the positive effects for her people of the marriage between the Jewish queen Esther and the Persian king, Ahasuerus/Xerxes. 608 This is basically the same reasoning also to be found in Charles' first diploma for Frederuna, in which her new importance is derived from the marriage to him: he made her the *consors regni* because she had become his wife.609 If we find the same thought pattern in Charles' diploma as well as the *ordo*, can we assume a direct connection between the two? It seems at least likely that Frederuna was crowned and anointed queen and that the *ordine* was created or used for this occasion.

These first diplomas show the importance the marriage had for Charles and that he appears to have at least intended to make his wife part of his rule. However, Régine Le Jan argues that the queen's dowry was too limited to one region to allow her to completely fulfil the role as a *consors regni*. 610 While we cannot argue against the limited distribution of the dowry, we should not forget that Charles

*secundum leges atque statuta priorum nobis nuptiali conubio sociavimus regnique consortem statuimus.* For the meaning of *consors regni*, see for example Fößel, Consors regni, 83.


himself mostly appears to have remained within Francia, the very same region where he chose to endow his wife. Hence, the size and distribution of the dowry do not help us when establishing Frederuna's influence at court. More promising is the evidence of the royal diplomas. Up until her death in 917, only one diploma is preserved mentioning her. In 915, Count Raoul de Gouy and the *fidelis* Etbert intervened on her behalf in front of the king, asking him to donate some land at Compiègne to her to construct a chapel for Saint Clement.611 While a low number of interventions does not necessarily indicate that the queen was without influence at court,612 her absence from the diplomas, especially compared to the presence of Charles' mother Adelaide, is remarkable. Her connection to Raoul de Gouy indicates that she was part of the link between Charles and this special circle of close associates around him. Except for this, only the appointment of her brother Bovo as bishop of Châlons may give us an indication of her importance, assuming that he owed his position to her influence with the king.

Despite the lack of evidence from the diplomas issued during her lifetime, Frederuna appears to have been very close to Charles. Before her death, Charles' diplomas twice connect her spiritual welfare to measures taken by the king.613 Immediately after her death on 10th February 917,614 for about sixteen months, each of Charles' diplomas either dealt with Frederuna's last wishes or established prayer services on her behalf.615 Frederuna's donation of Corbeny to Saint-Remy was confirmed,616 Ponthion given to Saint-Corneille, 617 some property to her chaplain Mauger618 and other estates to her foundation of Saint-Clement, 619 with two of the latter cases emphasising her wish620 or even letting her appear as petitioner.621


Prayer services for Charles, her and their children were installed at Saint-Remy, 622 at Saint-Denis for himself, her and Robert, 623 at Saint-Germain-des-Prés for Charles and her,624 at Saint-Corneille for her625 and at Saint-Clement for Charles, her, Bovo and Hagano. 626 This latter was the last diploma mentioning her or her death,627 letting her appear as if she was still among the living and marks the high point and end of this period.

Frederuna's death appears to have left a deep impression on Charles, making us believe that, indeed, a deep connection existed between him and his wife. If this was indeed the case, we can assume that her influence on him was rather large. Yet, in contrast to Adelaide, Frederuna only appears twice connected to important nobles: once just after her marriage in a diploma that probably was meant to show her at the centre of the court and a second time eight years later, with two men who seem to have belonged to the circle of men close to the king. Her appearances are rare and do not signify any new alliances or moments of special political importance—apart from her own wedding. Within the wider network of Charles' political relations, Frederuna, therefore, does not seem to have played a major role. As the first diplomas for her indicate, Charles had certainly intended to make her part of his rule, yet, based on the lack of interventions from her side, we must assume that she did not fulfil that role, possibly because of a lack of interest on her side.628

Unlike Frederuna's origins, those of Charles' second wife Eadgifu are well known. She was the daughter of Edward the Elder, king of the Anglo-Saxons. 629 Due to this connection it has been suggested that, by marrying her, Charles had tried to strengthen his connections to Baldwin of Flanders, who had married Ælfthryd, a daughter of Alfred the Great and aunt to Eadgifu; and their son Arnulf. 630 While such a connection may indeed have played a role, other concerns probably were of greater importance. Viking incursions had by no means ended with the Treaty of Saint-Clair-sur-Epte and still posed a major threat. In this context, a report from the Anglo-Saxon Chronicle is most interesting: "And that same year [916] Earl Thurcetel went across the sea to France, along with the men


who were willing to serve him, with King Edward's peace and support."631 While this is the only mention of such an event, it, nevertheless, appears that not only Northmen used England unhindered as a basis for ventures into the West Frankish realm, but also that at least some of them enjoyed the support of the king. Bearing this in mind, Charles' marriage to the daughter of the same king only a few years later632 may be seen in a different light. A marriage alliance with Edward tied the king to Charles and would, undoubtedly, put an end to Edward's support for the Northmen heading to the coasts of the West Frankish realm. Therefore, it seems safe enough to assume that the choice of Charles' bride can also be read in the context of his policies to protect the realm from Viking incursions—not the first time was such an alliance concluded, as Charles the Bald had married his daughter Judith to King Æthelwulf of Wessex to strengthen the alliance between the two against the Northmen.633

Of Eadgifu's role at court, however, the sources make no mention at all. She appears neither in the narrative accounts, setting back in around 919, nor in Charles' diplomas. In regard to the effort Charles had made to introduce Frederuna to the court and the elites of the realm, the latter is rather surprising. It has been suggested that the king transferred Frederuna's original dowry to her,634 yet this seems rather unlikely, given that he had confirmed Frederuna's donations of the same dowry to Saint-Remy and Saint-Corneille only in 917.635 That Eadgifu was certainly capable and interested in influencing royal politics has been shown by Simon MacLean,636 yet all the evidence we have only accounts for her later years. Any judgement on her importance at Charles' court is a mere assumption and we must therefore refrain from drawing any conclusions.

#### **Kinsmen and marriage alliances**

As we have already seen, we can find a number of Charles' relatives at court. Count Aledramnus, a relative of Charles' father Louis, 637 was an early supporter; Count Hugh of Maine, Charles' cousin, an important ally against Robert. Others such as Charles' nephew Gauzfrid and his brother-in-law Bovo were installed as bishops. There are also a number of other examples of how Charles furthered the interests of his kinsmen and how he constructed new relations with nobles important to him. The most obvious example is certainly Count Wigeric, who


<sup>631</sup> Anglo-Saxon Chronicle, MS A, 916, 66: *7 þy ilcan geare for Þurcytel eorl ofer sæ on Froncland mid þam mannum þe him gelæstan woldon mid Eadweardes cynges friþe 7 fultume.* Translation by Whitelock, Anglo-Saxon Chronicle, 64.

was married to Cunigunde, Reginar Longneck's daughter with Charles' half-sister Ermentrude638—Wigeric, therefore, was Charles' nephew. Wigeric's first appearance was in a diploma for the church of Liège, issued at the request of Reginar and Robert as well as another Robert. 639 In this diploma, the abbey of Hastières, held by Wigeric, was given to the church of Liège. Charles and Reginar, however, made sure that Hastières, as well as another abbey already belonging to the church, were to be held by Wigeric, his wife and their son until the last of them had died.640 More importantly, however, Wigeric also became Charles' new count of the palace.641 His proximity to the king is also clear from the first assembly held at Herstal in 916, when he was named first of the counts listed as participants in the judgement.642 Of Wigeric's further actions, nothing is known. He died before 922, leaving his widow Cunigunde to marry Count Ricuin of Verdun.643 His position as count of the palace, however, placed him in direct vicinity to the king which granted him access to the inner circle and established a second connection between Charles and the Reginars.

Wigeric's marriage to Charles' niece points us to the king's own female offspring. As the *Genealogica dictata a Carolo Rege* tells us, Charles had a number of daughters from his first wife Frederuna as well as from one or more of his concubines,644 allowing him to construct marriage alliances with men of his choice. A number of these connections can be traced in the sources, although some of them are rather problematic to assess. In Alberic of Trois-Fontaines' 13th century chronicle, for example, Raoul de Gouy is mentioned to have been married to Louis IV's sister—hence Charles' daughter.645 The famous poem *Raoul de Cambrai*, probably composed a little earlier, during the reign of Philippe-Auguste (1180−1223),646 names Alais, sister of King Louis, as his wife.647 While the *Genealogica* does not name an Alais, it does list an Alpais as a natural daughter of Charles648—close enough to make an identification possible. However, since both sources only date roughly 300 years after Charles' death, doubts are allowed. And it seems indeed, as Philippe Grierson has argued, that Raoul was not married to Alpais, but to Eldegarde, the daughter of Count Ermenfrid. 649


Yet, tracing Alpais is not futile. Indeed, it seems that she married another noble close to Charles, Count Erlebald of *Castriciensis*, 650 who, in 921, sought the king's help against the church of Reims and who Charles posthumously tried to free from excommunication. Therefore, Erlebald's flight to Charles and the king's efforts on his behalf appear in a different light: Erlebald was his son-in-law, someone particularly close to the king. Also a second marriage of Charles' daughters can be traced. Otto the Great's count of the palace, Gottfrid, was, as Eduard Hlawitschka has shown, married to a certain Ermentrude who, in turn, appears to be identical with Charles' daughter of the same name.651 Gottfrid was the son of none other than Count Gerhard, 652 one of the Matfrid brothers and therefore a member of a leading Lotharingian noble family belonging to the king's most important supporters. Nothing indicates when this marriage took place; however, it seems very likely that it had been arranged by Charles—after his deposition, the Matfrid family would have had nothing to gain from such an alliance. Of Charles' other daughters, only one more can traced in this context: Gisela, of whom Dudo of Saint-Quentin claims that Charles married her to the Viking leader Rollo. However, since this supposed marriage relates more to Charles' policy towards the Northmen, we will return to this problem later on. In any case, we can note that Charles made use of his daughters to construct or reinforce links to a number of nobles important to him. In this respect, Charles' practice differed considerably from that of his great-great-grandfather Charlemagne and Louis the German, who intended their daughters to join convents, yet fell in line with Louis the Pious and most of his descendants.653

#### **Close friends**

As we have seen, to some extent Charles' circle was made up of relatives, such as Bovo of Châlons; but more importantly, by nobles of his choice, such as Raoul de Gouy, Stephen of Cambrai and Abbo of Soissons. Within the latter group, we also find, as mentioned, a certain Hagano, who Flodoard introduces to us as "his councillor, whom he had elevated from mediocrity."654 Who was this man that the king was not willing to dismiss, even when threatened with the defection of a significant group of nobles? Most often, scholars have taken Flodoard's words as the basis of their interpretations and, in addition, assumed that he was of Lotharingian

652 Hlawitschka, Ahnen I,2, 260.

<sup>650</sup> Werner, Nachkommen, 461. The marriage between Erlebald and Alpais was originally proposed by Vanderkindere, Formation II, 206−207, based on a charter issued by the count naming his wife Alpais and the proximity between the Charles and Erlebald.

<sup>651</sup> Hlawitschka, Ahnen I,2, 261 with Hlawitschka, Anfänge, 57−58 and 68−69. For Ermentrude, see chapter III.2.2, n. 644.

<sup>653</sup> On the various daughters of Carolingian kings and their roles, see Schieffer, Töchter and Hartmann, Königin, 189−195.

<sup>654</sup> Flodoard, Annales 920, 2: *Pene omnes Franciae comites regem suum, Karolum, apud urbem Suessonicam, quia Haganonem consiliarium suum, quem de mediocribus potentem fecerat, dimittere nolebat, reliquerunt.*

origin.655 This hypothesis seems to be based on his first appearance in the sources in 916, when he figures as third among the participants without title, the *proceres*, in the judgement of Herstal.656 Since all the bishops and counts participating in the *judicium* appear to have been Lotharingian nobles, it seems not too speculative to assume that the same applies for the *proceres*. This conclusion however, given that almost nothing is known about these men, must be treated with utmost care, even more so as the assembly of 919, also at Herstal, was visited by West Frankish nobles.657 Regarding Hagano's "mediocre" origins, the account of Folcuin of Saint-Bertin may be of help. His *Gesta*, composed forty years after Charles' deposition, differ slightly from Flodoard in that they describe Hagano as someone "whose lineage and nobility were unknown to the Franks."658 Two conclusions can be drawn from this account: a) Hagano's origins were not necessarily mediocre but b) they were either unknown to the nobles of the Western realm or they chose to ignore them. In both cases, Hagano appears not to have been a member of the highest echelons of the aristocracy, but someone who had joined the king from the outside. While this does not necessarily mean that Hagano was a Lotharingian noble, it makes it much more likely that, at the very least, he was not from Francia. Concerning Hagano's family, two hypotheses have been developed by scholars. Hagano has long been considered a relative of Charles' wife Frederuna, yet, as we have argued, this appears to be unfounded.659 The second hypothesis, however, sees in him a member of the Haganonide family, based in the Middle Rhine in the vicinity of Worms and Mainz660 and with connections to the Matfrid family.661 It is probably here, in the border region between Lotharingia and the East Frankish realm, where we can find Hagano's origins—by no means mediocre, since his ancestors appear to have served Charlemagne662 and Louis the Pious.663


660 Le Jan, Famille, 59−60 and Depreux, Comte, 390−391. On the Haganonide family in the area, attested since the 7th century, see Gockel, Königshöfe, 296−297.

661 Depreux, Comte, 388−389 discusses this theory based on the *Vita Gerardi abbatis Broniensis* (c. 1, 656: *Pater eius Stantius ex prosapia Haganonis Austrasiorum ducis prosatus, mater autem Plictruidis nuncupata domni Stepheni Tungrorum episcopi soror traditur extitisse germana.*). Stephen's relation to the Matfrid Gerhard is confirmed by DLCh 57. See also above. A connection between Hagano and the Matfrid family may also be indicated by an entry in the *Liber Memorialis* of Remiremont, showing Hagano next to the count of the palace Gottfrid, Bishop Richer of Liège, Matfrid and Ermentrude. Hlawitschka, Anfänge, 72. Liber Memorialis Remiremont, 85 fol. 39r: *IIII kal. april. obiit Godefridus comes; XII kal. … obiit Otlint; X kal. iuli obiit Richerus episcopus; XIIII kal. sept. obiit Matfridus; VIIII kal. … Haganon anniversarius; Adelardus. Ermentrut, Gerart, Godefrit…*

<sup>662</sup> Codex Laureshamensis N° 3, 273.

<sup>663</sup> Urkundenbuch Coblenz N° 62 70. Metz, Reichsgut, 152−154.

Hagano's ascent was, as mentioned, marked by his appearance at Herstal in 916, when he featured inconspicuously among the *proceres*. Eighteen months after the assembly, he makes his second appearance, intervening in a diploma for Saint-Corneille664 and one year later his third, this time making requests for Frederuna's chaplain Mauger and her foundation of Saint-Clément. 665 Both diplomas reveal a very special relationship with Charles. The first shows Hagano as the one who is requesting donations and privileges for the souls of the deceased Frederuna as well as Charles' own,666 while in the second Hagano is included in the prayer service installed for Charles and Bovo. 667 Hence, Hagano's ascent takes place in direct connection with some of the most intimate diplomas Charles issued for his deceased queen, which seems to indicate a very private connection between the two, probably developed over the death of his beloved wife.668

Flodoard tells us that early in 920—undoubtedly in late January—the Frankish nobles left Charles because he was unwilling to dismiss Hagano. 669 Given the evidence of the two diplomas, it seems safe to assume that Charles denied their request because of the special bond that had developed between the two—Hagano had become a friend, his intimate, someone he deeply cared for. Richer gives us a very interesting description of the relation between the two: "the same [Hagano] alone served the king while the magnates remained far away, and often in public he took the king's hat from his head and put it on his own."670 Richer is, of course, not a reliable source and should not, by any means, be taken literally. Nevertheless, his words seem to capture the relation between the king and his friend, at least as it was seen from the outside. Richer's description of Hagano as the only one having access to Charles while the nobles remained distant is definitely wrong—in 919 alone, Charles issued six diplomas, among them one for Robert of Neustria and another relating to the second assembly at Herstal.671 Nevertheless, if Charles and Hagano spent a lot of time together, this might have been perceived negatively, giving rise to rumours among the nobles that someone who they thought of as being an outsider, or at least not belonging to their own networks, had gained an unreasonable amount of influence over the king and the decisions made at court. This image might have been further fuelled by Hagano's promotion. In 916 he was


among the *proceres*, in 917 Charles' diploma depicts him as "our *fidelis*" and in 918 as "our *fidelis* and venerable count."672 Hagano might, of course, have inherited the title from someone within his family, yet, given his close relation with Charles, it seems likely that the king granted it to his companion as a sign of his favour.

As Flodoard informs us, seven months after the assembly of Soissons, Archbishop Heriveus was able to mediate a settlement between Charles and the Frankish nobles.673 One would expect that the conditions of the settlement involved Hagano in some way and, indeed, the position of Charles' intimate appears significantly changed from that moment onwards. In fact, at the end of Flodoard's putative seven-month period, Charles issued a diploma at Herstal showing the counts Raoul de Gouy and Sigard intervening together with Hagano for the church of Cambrai.674 This diploma is only the first in a series, followed by three more in 921, in which Hagano features. The first one, issued for the abbey of Fossés in April, shows him together with Bishop Abbo of Soissons and Abbot Rumald; the second one, in August, with Count Ermenfrid for the abbey of Brogne; the third one as one of the participants in a judgement in favour of the church of Châlons, alongside the counts Helgaud and Raoul de Gouy as well as the bishops Abbo of Soissons and Stephen of Cambrai. 675 Indeed, the first diploma from 920 is also the first one to show the group of nobles drawing their influence from their special relation to Charles acting together, although all of them had previously been in the king's vicinity. Given that this was the first diploma issued after the reconciliation with the nobles from the Western realm and given Hagano's own presence in it, it seems to carry special meaning. Undoubtedly this was a reaction to the reproaches the nobles articulated at Soissons and reflected the settlement negotiated thereafter.

Nothing is actually known about the terms agreed upon between Charles and the nobles, yet they must have included something about the relation between the king and his intimate. Since Flodoard fails to provide us with any details, we have to rely on the evidence of the royal diplomas in which, from September 920 on, Hagano becomes much more visible. Charles, as it would appear, had changed his tactics concerning Hagano. The two of them certainly continued to be extremely close: in 921 Hagano was included in a memorial service for the king, Frederuna and her relatives.676 Yet at the same time Charles now let Hagano openly participate in the political decisions taken at court, using those nobles most attached to himself to introduce his intimate into the networks of the Frankish nobility. The outsider was meant to become an accepted member of the circle at court, his relation with

674 DChS 106 (8th September 920).

<sup>672</sup> DChS 90 (26th July 917), 205 (*…fideli nostro Haganone…*) and DChS 95 (26th May 918), 219 (*… fidele nostro ac venerabili comite Aganone…*).

<sup>673</sup> Flodoard, Annales 920, 2.

<sup>675</sup> DDChS 108 (22nd April 921, for Saint-Maur of Fossés), 127 (27th August 921, for the abbey of Brogne) and 112 (20th September 921, *judicium* for the church of Châlons).

<sup>676</sup> DChS 108.

the king publicly displayed to end any possible rumours. The next step would have been to widen this circle, and indeed Hagano features among Charles' supporters in the treaty of Bonn.677 This was not an unreasonable strategy: Hagano was not the first friend Charles had introduced to the highest circles.678 In 898, Charles issued a diploma for a certain Theodosius, granting him property in Septimania.679 This diploma was requested by a *fidelis* called Robert. This same Robert also appears in a number of other diplomas: in 904, he intervened in royal charter in favour of Bishop Mancio and the church of Châlons, this time depicted as "our *fidelis* and highly beloved";680 in 907, he is mentioned in the diploma introducing Frederuna to the court, together with members of Charles' family and some of the highest ranking nobles of the realm, among them Robert of Neustria, Count Erkanger and Bishop Anskeric of Paris, again described as "our beloved";681 a diploma issued around 912 shows him as "our beloved" intervening together with two of the most important nobles, Robert of Neustria and Reginar Longneck, on behalf of the church of Liège;682 in 915, finally, he appears in a diploma together with Richard the Justiciar, alongside whom he intervened on behalf of the abbey of Tournus.683 Nothing more is known about him,684 but the diplomas, nonetheless, give us the impression of a remarkable man. Carrying no title, yet nevertheless being depicted as "our beloved" (*dilectus nobis*), he seems to have been close to Charles, a conclusion that is further strengthened by his interventions on behalf of others. At the same time, his position seems to have been accepted by the leading nobles of the realm, most notably of course Robert of Neustria. This Robert was not alone however: in 915, we find another "beloved" *fidelis* called Etbert in Charles' vicinity, intervening together with Raoul de Gouy on behalf of Frederuna, granting her property to found a chapel in honour of Saint-Clément at Compiègne.685 Like Robert, Etbert seems to have been close to Charles, and like the former, he was accepted by leading members of the court.

Thus, the West Frankish nobility was not generally opposed to Charles' having intimates and to them participating in political affairs at court. What they seem to have opposed, however, was the special relation Charles and Hagano had developed during the years after Frederuna's death. Charles' solution to this problem seems to


683 DChS 82 (10th October 915). Lauer identified this Robert as Robert of Neustria. However, while the magnate is otherwise always described with one or several of the titles *comes*, *marchio* or *abbas* as well as epithets like *dilectus* or *venerabilis*, this Robert is only depicted as *fidelis nostri*, concording with the aforementioned diplomas.


have been the introduction of Hagano into the higher circles of the nobility and the public display of his new role at court. If Hagano was to be accepted by the other nobles, he had to have a position similar to their own. This, however, also meant that Hagano needed to be provided with *honores* according to his new position within the hierarchy of the nobility. In fact, Flodoard's account contains a number of references that indicate that by 922 Hagano had large resources at his disposal; he could muster a personal retinue strong enough to undertake military operations and had been able to assemble a treasury.686 Finally, Charles wanted to give Hagano his own lay abbacy. The king's choice fell on the abbey of Chelles, held by his aunt Rothild. If, however, he had expected to encounter less resistance because of their kinship, he was soon proven wrong. Rothild's daughter was married to Robert of Neustria's son, Hugh the Great, who now intervened.687 The conflict soon developed into the rebellion of the West Frankish nobles, leading to the coronation of Robert of Neustria.

#### **III.2.3 The role of the** *marchiones*

Three men within the Western realm are generally qualified as *marchiones*: Robert of Neustria, Richard the Justiciar and William the Pious.688 The three of them distinguished themselves from the other nobles by concentrating a large number of *honores*, including important abbeys, in their hands; by installing viscounts within the regions they controlled; and by close cooperation with bishops and abbots within their vicinity.689 There certainly were distinctive differences in how these men had come to power, Robert with royal consent taking control of a conglomerate of *honores* already long in existence, Richard extending his influence by means of violence at a time when royal attention was focused elsewhere; yet the result was rather similar. Karl Ferdinand Werner probably went too far when describing the *marchiones* as direct deputies of the king, wielding viceroyal power within their regions690 and therefore attributing the system a degree of institutionalisation that negates the differences between the different regional hegemonies.691 Nevertheless, his approach seems to point in the right direction.

As we have seen above, in Burgundy, Richard had established himself as intermediary, controlling the direct contacts between the local nobility and the king, a position that seems comparable with those of the other *marchiones*. Such an accumulation of power was nothing new.692 As mentioned, the conglomeration

687 Flodoard, Annales 922, 8.


<sup>686</sup> See chapter VI.4.

<sup>688</sup> On the meaning of the *marchio*-title, see Stieldorf, Marken, esp. 213−215 for the Western realm under Charles the Simple.

<sup>689</sup> For William, see Lauranson-Rosaz, Roi, 426−428.

of *honores* in Robert's hand, for example, had developed over quite a long time and passed from one magnate to the other. Charles the Fat had built his rule on the cooperation of these magnates, using them as conduits to transmit the royal will into the regions controlled by, for example, Odo and Bernard of Auvergne. 693 Charles, as we have seen, built his early reign mainly on his old allies, Archbishop Fulk of Reims, Bishop Anskeric of Paris and Count Heribert I of Vermandois. Richard the Justiciar and William the Pious, who had followed their own course during the struggle between Charles and Odo, remained distant at first, waiting to see how Robert of Neustria would act. Only after the latter had paid his allegiance to Charles, did the former two follow.

#### **William the Pious**

A private charter issued by William just before this moment in May 898 has been preserved that is most enlightening about his attitude towards the king.694 The charter dates "the year in which Odo, king of the Francs and Aquitanians, died"695—meaning that five months after Odo's death and Charles' coronation at Reims, William was not yet willing to accept the latter as his new overlord. His attitude towards the new king and his own self-perception becomes even clearer when analysing the charter's legal content. William donated property to the abbey of Brioude, although when it came to the clauses protecting his grant, he made an interesting addition: neither man, nor *king*, nor count, nor abbot, nor canon or any other person was allowed to estrange the property.696 While such clauses were common, the addition of the king certainly was not697 and betrays the *marchio*'s exceptional self-confidence at that moment. Soon after, William paid his allegiance to Charles and, in 899, he was the first magnate to appear in one of the king's diplomas. Present at the important assembly of Tours-sur-Marne, he served as *ambasciator* in a diploma Charles issued at the request of William's friend Gerald of Aurillac. 698 William's support for the king on such an occasion appears to have come at a price. An *ambasciator* was someone who investigated the justness of a plea to the king, serving at the same time as intermediary on behalf of the petitioner in front of the king.699 Hence, like Richard in Burgundy, William acted as

<sup>693</sup> MacLean, Kingship.

<sup>694</sup> Cartulaire Brioude, N° 309, 313−315.

<sup>695</sup> Cartulaire Brioude, N° 309, 315: …*anno quo mortuus est Odo rex Francorum vel Aquitanorum.*

<sup>696</sup> Cartulaire Brioude, N° 309, 314−315: *Nullusque hominum, neque rex, neque comes, neque abbas, neque canonici aut aliqua persona vendendi, concambiandi aut quolibet modo ab ipsa casa alienandi potestatem habeant.*

<sup>697</sup> A similar phrasing was applied in the foundation charter for Cluny. Recueil Cluny 1, N° 112, 124−128 (11th September 910): *Placuit etiam huic testamento inseri ut ab hac die nec nostro, nec parentum nostrorum, nec fastibus regie magnitudinis, nec cujuslibet terrenȩ potestatis jugo subiciantur idem monachi ibi congregati; neque aliquis principum secularium, non comes quisquam, nec episcopus quilibet, non pontifex supradicte sedis Romanae…*

<sup>698</sup> DChS 21. See above, 10−11.

<sup>699</sup> Depreux, Bitte, 84−85.

a layer between the local nobility and Charles, who seems to have acknowledged William's position south of the Loire. Furthermore, a private charter issued by his sister Ava in 901 was signed by William as "*princeps* and count of the palace,"700 a title that is repeated in 905 in one of his own charters.701 Charles had made William his new count of the palace, succeeding or replacing Count Elduin, who had held this office previously. William appears too few times in royal diplomas to conclude that he was in fact an important figure in the palace itself.702 The title therefore has another meaning, indicating his proximity to the king and his importance within the hierarchy of the nobles.

The relations between William and the king appear to have faltered around 908. As with Richard, Charles seems to have intervened within William's sphere of influence south of the Loire. The marking point is a diploma the king issued on 5th June 909 for Abbot Regembald of Psalmody. 703 Next to the abbot, Count Raymond of Nîmes and Albi, an important noble from the South, can also be found in the diploma, he is depicted as "illustrious count and our beloved *fidelis*"704, a very rare combination of epithets and titles that emphasises his importance for Charles. The *Vita Geraldi* tells us that Raymond took the nephew of William's friend Gerald prisoner705 while his son took the Gothic March from William's successor, two incidents that indicate the rivalry between the two.706 From the moment Charles' diploma was issued, there is a shift in the tone of William's private charters. The foundation charter of Cluny, dating to 11th September 910, repeats the exemption of the king from intervening in the abbey's affairs.707 Equally important, William now started referring to Odo in his charters, indicating that his grants were made for the wellbeing of his *senior*, King Odo, 708 while Charles never received such attention. These mentions of Odo may further indicate that William now was leaning more towards Odo's brother, Robert of Neustria, with whom he had already appeared in a royal diploma in 905, restoring property in the Limousin to Robert's abbey of Saint-Denis. 709 Similarly, a certain shift also seems to have taken place in favour of Richard the Justiciar. William's private charter from November 916


<sup>700</sup> Recueil Cluny I, N° 74, 83−84 (29thNovember 901), 84: *princeps et comes palatio.*

<sup>709</sup> DChS 50.

seems to have been issued during a large regional assembly, including his nephew William the Younger, two bishops, three abbots, two counts and signing even before the bishops, a certain Boso, who may well be identical with Richard's son.710 However, after William's death on 6th July 918,711 Richard' son Raoul and Robert of Neustria seem to have joined forces to take Berry from William's successor, his nephew William. This enmity seems to have lasted until 924, when William the Younger payed his homage to Raoul, by now the king, in return for having Berry restored;712 which at the same time indicates that he did not support Robert and Raoul in the rebellion against Charles.

#### **Richard the Justiciar**

After Richard was replaced by Robert at court in 903, he disappears from the sources for several years. Early in 907, a solution seems to have been found that allowed Richard to return to court. In a diploma issued for the provost of the church of Langres, Otbert, Richard intervened alongside the bishop of the church, Argrim, and his own ally, Manasses.713 The importance Charles attributed to Richard's return to court is marked not only by the place the diploma was issued— Compiègne—but also by the date, the 4th April 907 being an Easter Saturday. Hence, the king and the magnate celebrated Easter together in Charles' most important palace. About a year later, Richard again requested a diploma, this time marking the restoration of property to the abbey of Saint-Martin of Autun.714 While Richard himself only appears in Charles' diplomas once more—in 915 he visited the king at Gondreville in Lotharingia, supporting the abbot of Saint-Philibert of Tournus alongside the *fidelis* Robert715—contact with Charles was also had by his sons. In 914, Hugh received property from the king716 while Boso accompanied Charles in 921 at Bonn.717 The fact that his sons did indeed represent him becomes clear from Richard's private charters. A judgement held in 916 shows Richard in the presence of his three sons, Raoul, Hugh and Boso. The charter making note of the judgement is signed by Raoul, who did so at his father's order and as his deputy,718 a procedure that was repeated two years later,719 while another charter issued by Richard shows


all of his sons among the witnesses.720 Thus, Richard did his best to integrate his sons into his rule and establish Raoul as his successor in Burgundy.

Richard's importance at Charles' court after his return in 907 is hard to estimate. The effort Charles made with his first diploma in 907 certainly indicates that Richard was still considered to be a key figure within the realm. The newly won cooperation must have involved Charles' acknowledgement of Richard's control over Burgundy and its nobles. The Burgundian elites still held contact with the king: as we have seen, Bishop Argrim of Langres and Abbot Wicheramnus of Saint-Philibert sought out Charles, and the *Gesta Pontificum Autissiodorensium* mentions that Bishop Geran was on his way to the king in December of 914, when he died at Soissons.721 However, in all of these cases, Richard appears to have played a key role. While the Argrim and Wicheramnus were accompanied to Charles by the *marchio*, Geran seems to have acted with Richard's consent. Thus, all three cases demonstrate Richard's influence over contacts between the local elites and the king. This was probably also his role at court. As with William south of the Loire, Richard acted as mediator between Charles and the Burgundian nobility. While certainly not as influential as Robert, he remained a key figure in the affairs of the realm and was, at times, represented at court by his sons. Whether the, extremely pro-Ricardian, account of the *Chronicon S. Benigni Divionensis* is accurate in its judgement that Richard had always been loyal to Charles722 must remain unclear, yet it may be worth considering that the final rebellion against Charles only occurred after Richard's death and with the aid of his son Raoul, while in 920, during his lifetime, it had still been possible to find an agreement.

Richard's return to court, however, is unlikely to have happened against Robert's will. In fact, during the following years, an increasing level of cooperation between the two can be traced. The first sign of this is their joint campaign against the Northmen leading to the battle of Chartres in 911. From this campaign, a letter is preserved which was copied during the 11th century into a 9th century manuscript: "Count Robert and *dux* Manasses greet Count Richard. Know that we set off against the Northmen, but not having found them we are returning to Paris sending this to you and enquire whether you will come to us or not."723 We will deal with the campaign itself in our next chapter, for now however, it is important to note that Manasses, Richard's ally whose offensive words had been the final reason for Robert leaving the court in 900, was now working together with that same Robert, undoubtedly on behalf of Richard. This cooperation was seemingly strengthened over the course of the following years. Richard's aforementioned

<sup>720</sup> Roserot, Chartes, N° 13 (18th May 918), 185.

<sup>721</sup> Gesta Pontificum Autissiodorensium I, 199−201.

<sup>722</sup> Chronicon S. Benigni Divionensis, 123: *Et hoc post mortem Richardi Ducis, qui ab exequutione iustitie cognomen accepit. Ipse namque quamdiu vixit, Carolo regi semper fidelis exstitit.*

<sup>723</sup> Bibl. commun. Chartres, ms. 92, fol. 38v (printed in Catalogue Chartres, 48): *Rotbertus comes et dux Manassae Richardo comiti salutem. Scitote quoniam fuimus perrecti contra Normannos, sed non invenientes regressi sumus Parisius mittentes ad vos et requirimus utrum vos necne venieris ad nos*.

private charter dating to 918 was issued "for the absolution of our lord, the glorious king Odo, and his most noble brother Robert, the illustrious *marchio*" and "for the peace of our lord Richard, most pious duke, and his most noble wife and their most famous sons."724 While the list of people for whose souls the charter had been issued also contains Charles the Bald and his wife Ermentrude, no mention is made of Charles the Simple, not even in the dating. The charter becomes even more interesting when taking into consideration the fact that the restoration of the *villa* of Tillenay to the church of Autun confirmed by the charter actually also confirmed the terms of a preceding royal diploma, issued by Odo for Bishop Adalgarius in 893.725 The 918 charter was the sign of an alliance concluded at this moment, constructing a link via Odo's diploma and indicating their friendship by commemorating their families.726 The charter was not the only sign of the new alliance however. Probably around the same time, Richard's son Raoul married Robert's daughter Emma, 727 strengthening their bond through matrimony. This new alliance was not necessarily aimed against the king: already before 922, Raoul and Robert joined their forces against William the Younger, taking Bourges and the Berry from William the Pious' successor.728

#### **Robert of Neustria**

While William and Richard were both very important men for Charles' rule, Robert remained the leading man, whose proximity was apparently sought by both. His central role within the realm is undisputed, but his relations with Charles have recently been reassessed by Geoffrey Koziol. Most commonly, scholars have assumed that, after the initial problems had been solved, Robert and the king cooperated up until around 919, when there was a sudden breakdown in their relations leading to the assembly of Soissons.729 However, as Koziol rightfully points out, the absence of news about tensions and conflicts between the king and the *marchio* between 900 and 919 is not necessarily a sign of Charles' and Robert's good relations, but simply due to the lack of narrative sources for this period. In opposition to the older research, Koziol argues instead in favour of constant tensions because of differing political views concerning kingship and the relations with the Northmen, meaning that there was a direct path leading from the early


<sup>724</sup> Recueil Robert et Raoul, N° 51 (after 1st September 918), 205: *…sed et pro absolutione domni Odonis gloriosi regis et nobilissimi fratris illius domni Rothberti illustris marchionis… et quiete domni Richardi piissimi ducis et nobilissime conjugis illius ac clarissimorum filiorum illorum et omnium fidelium suorum…*

<sup>725</sup> DOdo 35. On this diploma and the forgeries produced by Bishop Adalgar of Autun, see Koziol, Politics, 358−361.

<sup>726</sup> On the role of prayer requirements in the formation of alliances between nobles, see Althoff, Amicitiae and Ludwig, Krise.

conflict between Charles and Robert to the later rebellion.730 While we will deal with Charles' politics towards the Northmen in the next chapter, at this point it seems necessary to discuss Koziol's arguments concerning Charles' vision of kingship.

Koziol's first argument revolves around Charles' policies towards lay abbacies. Based on Charles' diplomas, he reconstructs a political programme of the king aimed at containing the abuse of power by the lay abbots against their convents and at the protection of the *mensae* of the abbeys by means of royal control over transactions, with royal diplomas serving as guarantees; hence royal control over the dealings between lay abbots and the monks.731 Indeed, as Koziol convincingly shows, it was not uncommon for Charles' diplomas to contain such clauses. The question is, however, whether these measures were directed against the magnates holding these abbeys and whether Charles was indeed as opposed to lay abbacies as Koziol claims. The best way to address this issue is, undoubtedly, to analyse Robert's politics towards his own abbacies. A number of his private charters for Saint-Martin of Tours can help us to shed some light on this issue. In 892, Robert, faced with the resistance of the canons, who threatened to bring the case before the king, returned property to them that his vassal Patericus has usurped.732 A similar case occurred in 899, when Robert restored a *cella* to the abbey's hospice, which his brother Odo had given to one of his *fideles* without having consulted the chapter733—a measure he confirmed again one year later.734 Robert, as we can see, was willing to correct former misdealings even if they had been done by himself or his brother and even if this meant giving up some power. Similarly, Robert appears not to have opposed royal interventions in the dealings between abbot and monks. The first example are of course the diplomas and charters concerning Saint-Amand in 906,735 in which Charles confirmed an exchange between Robert and the monks of the same abbey, an exchange that is laid down more precisely in two private charters issued shortly afterwards, both of them mentioning Charles' approval.736 There is also another document showing that Robert was not opposed to royal intervention and was even willing to use the king as guarantor. In 912, he issued a charter for Marmoutier confirming that the abbey was independent from the archbishop of Tours who had tried to bring some of its property under his control and furthermore promising to obtain a royal diploma addressing this problem.737 Thus, Robert was neither opposed to restoring property he or his supporters held nor to the king acting as a guarantor if necessary. In this light, it


seems rather questionable whether Charles' and Robert's opinions concerning the management of abbacies really differed fundamentally from each other. Similarly, if Robert really had resented Charles' politics towards the abbeys he held, why would he have consented to the king issuing diplomas expressing these same politics for these very same abbeys?

This brings us to the second issue Koziol identifies as a cause for tensions between the king and the magnate: the right of nobles to participate in ruling the realm. According to him, Robert derived his right to counsel the king from his rank within the hierarchy of the society while for Charles "counsel and consensus were elements in the display of royal majesty," also meaning that the king was at complete liberty to choose his councillors.738 Koziol is right to point out that Charles' diplomas convey a very distinctive image of royal majesty, describing even the highest magnates petitioning the king on their knees.739 However, at the same time, these very same diplomas also make numerous references to Charles, the king, listening to his advisors740 and even though there are others emphasising Charles' majesty,741 they, nevertheless, seem to reflect the political reality. As we have seen, those surrounding the king were most often the most powerful men of the realm, old allies and new key figures. Despite all the emphasis on his royal majesty, Charles was very well aware of the political necessities and did his best to include the magnates into his rule. Thus, in 900, we find Robert, Richard and Heribert I at his side, planning a campaign against the Northmen;742 thus we also find Charles referring to the judgement of the nobles on various occasions743 and surrounded by them when treating with Henry the Fowler at Bonn in 921.744 The only time Charles can be seen to contravene this policy was with Hagano, with whom he enjoyed a very close and personal relation. Koziol is certainly right to point out the strain Hagano's position placed on the relations with the nobles. However, as we have argued, Charles reacted to his in a way that acknowledged the nobles' concerns, trying to remove its cause by establishing Hagano within the West Frankish hierarchy. In any case, Hagano was a very special—and isolated case that should not overshadow the other evidence provided by the sources.

739 Koziol, Politics, 513, refering to DChS 65, 147−148 (*comes Reynerus et demarcus et Rotbertus comes et demarcus … pronis flagitaverunt genibus*).

740 For example DDChS 10, 16 (*Quorum congrue petitioni…, aurem clementiae nostrae accomodantes, cum fidelium nostrorum consensu…*), 49, 108 (*Cujus humillimam peitionem cum consilio regni nostri utriusque ordinis principum mentaliter considerantes…*), 53, 115 (…*a nostris, tam episcopus quam et laicis, fidelibus responsa accepimus…*), 56, 121–122 (*Si regum consuetudines antiquorum exequimur necnon patrum mores praecedentium imitamur fideliumque nostrorum benigne consulta suscipimus…*), 88, 199 (…*consultu fidelium procerum nostrorum, decrevimus…*).


<sup>738</sup> Koziol, Politics, 513.

<sup>741</sup> Koziol, Politics, 514.

Hence, Charles' and Robert's convictions concerning kingship appear not to have differed as much as Koziol claims. As we have argued, the early tensions between the king and the *marchio* seem to have stemmed from Robert who was aspiring a position at court equal to his standing in the hierarchy of the nobles of the realm. Charles, at first, blocked these aspirations, but not because he insisted on his right to choose those surrounding him, but because the position was already occupied by Archbishop Fulk of Reims. The arising conflict when the vacancy following the latter's murder was not filled by Robert but by Richard, did create a rift throughout the reign. As we have seen, Charles created a network of alliances to contain the *marchio*. Robert's own reaction to the rising tensions is documented by a private charter issued at Tours in September 900.745 Within the witness list, we find a large number of nobles from the region, Archbishop Erbernus of Tours, the bishops Raino of Angers, Fulcher of Nantes and Berno of Orléans, a certain abbot Aimo and, finally, the four viscounts Hatto, Guarnegaud, Fulk and Rainald. The charter was meant as a demonstration of Robert's power, to be conveyed by the king's new archchancellor, Bishop Anskeric of Paris, who signed the charter at the *marchio*'s invitation.746 Anskeric's presence, however, also indicates that from the beginning both sides remained in contact with each other, probably trying to negotiate a settlement, and hence that both sides were interested in finding a solution. For the time being, such an accord appears to have been out of reach. Richard's position at court, undoubtedly, was one of the reasons why Robert could not yet be integrated into Charles' rule and, as we have argued, it is also likely that Charles' mother, Adelaide, was opposed to making concessions. By 903, however, after Adelaide's death and the breakdown in relations between Charles and Richard, Robert not only returned to court but also, from now on, was seemingly acknowledged as the leading noble of the realm.747

From this moment on, up until Flodoard's account begins, and possibly even after, the sources at our disposal do indicate a general cooperation between the

746 Recueil Robert et Raoul, N° 42, 164: *Askericus, Parisiorum episcopus, hanc auctoritatem, rogante ipso Rotberto comite, firmavi.* Next in the list is Bishop Otger of Amiens, who signed the list as *episcophilax*. Since it is not clear whether he belonged to Robert or was part of Anskeric's embassy, I have refrained from mentioning him here. See also chapter III.2.1.2.

747 The process of his final integration into Charles' rule was probably furthered by Robert's marriage to Beatrix, daughter of Heribert I of Vermandois (on the marriage see Werner, Nachkommen, 458). Unfortunately, the date of the marriage is unknown. Settipani, Préhistoire, 407−408. According to him, Werner argued for 897 as the date since during this year a) Heribert I reconciliated with Robert's brother Odo, b) Robert received the abbey of Morienval in the Vermandois, c) Robert and Beatrix' son, Hugh the Great, was already married in 914 since during this year his brother-in-law Hugh of Maine figured in one of Robert's private charters. However, a) the reconciliation between Odo and Heribert does not necessitate a marriage, b) nothing indicates that Morienval passed into Robert's hands in 897 (see below) and, c) Hugh of Maine's presence in Robert's charter does not necessitate that the marriage between his sister and Robert's son had already taken place. It seems to me that any date other date would equally be possible—for example before 893, when Odo furthered Heribert's interests, and likewise around 898 or even only in 903, when Robert and Charles (and hence Heribert), were on good terms.

<sup>745</sup> Recueil Robert et Raoul, N° 42, 157−165 (13th September 900, Tours).

king and the *marchio*. The most interesting evidence for this is, without a doubt, one of Robert's private charters, informing us that the *marchio* had been absent from Tours for two years because he had been charged by the king with the management of various affairs in the *regna* of Francia and Neustria.748 This corresponds well with an expression used in one of Charles' own diplomas, issued in 915 and depicting Robert as "our Highness' most loyal executor."749 Hence, Robert's private charter reflects not only the *marchio*'s view of how he participated in the management of the realm, but also the king's, thus indicating that they indeed mirrored the actual state of affairs.

If we follow the charter dating to 11th November 912, Robert's absence from Tours covered most of the years 911 and 912, that is to say the time of the campaign against the Northmen leading to the battle of Chartres and Charles' initial voyage to Lotharingia. Concerning the campaign against the Northmen, hardly anything is known at all apart from the battle and the aforementioned letter sent by Robert and Manasses to Richard. 750 So far, nothing in the sources allowed us to determine the role the king played in this campaign. Robert's charter, however, when read in relation with these events, indicates that Robert and Richard had acted on the king's behalf, instructed by him to protect the realm from the Viking incursion. It is interesting to note that this command falls into a period which is marked by a long absence of diplomas issued for Robert. After the initial period with a high density of diplomas for the *marchio*, from September 906751 up until August 911, Robert is only mentioned once in the royal diplomas, namely in 907, when he appears together with other important nobles for the realm in the charter introducing Frederuna to the highest nobility.752 The new diploma, issued 3rd August 911 at Compiègne for Saint-Denis, 753 only two weeks after the battle of Chartres,754 undoubtedly marked the reception of the victorious *marchio* by the king. Read in the context of Robert's charter, the relations between the two appear to have been strong when no charters were issued.

The high degree of trust between the two is also revealed by the second period of Robert's absence from Neustria marked by his charter, Charles' voyage to Lotharingia. Two royal diplomas mention Robert shortly after the acquisition of

754 20th July 911. Eckel, Charles, 70.

<sup>748</sup> Recueil Robert et Raoul, N° 47, 181−186 at 184: *Domnus Rotbertus […] gloriosus abbas necnon et comes, propter diversa regnorum Francię atque Neustrię negotia, quibus a rege prepositus erat, ab urbe Turonica fere per biennium defuisset…* On this charter, see also Guyotjeannin, Notice and Guillot, Formes, 83, n. 101.

<sup>749</sup> DChS 81, 181: *… Rotbertus, nostre serenitatis exequtor fidelissimus…*

<sup>750</sup> See chapter III.2.3, n. 723.

<sup>751</sup> DChS 54 (7th September 906, for Saint-Amand)

<sup>752</sup> DChS 57 (21st May 907).

<sup>753</sup> DChS 66.

the new *regnum*. 755 Both were probably issued after April 912756 which, in connection with Robert's charter making no mention of Lotharingia, seems to indicate that Robert only joined the king later on, possibly as late as 913. Hence, Charles, while he himself was absent, charged Robert with the management of affairs in the region that had so far been his most important basis of support. This choice is much less surprising when taking in consideration that Robert had already built up connections within Francia. Charles himself may have installed the *marchio* as abbot of Saint-Amand in 906, when he issued a diploma for the abbey at Robert's request.757 On this occasion, Robert also issued a private charter for the monks,758 containing a rather large witness list featuring the new bishop of Noyon, Robert, and five counts, Almanus, Odilard, Hilmeradus, Richer and Erlebald. At least two of them, Odilard and Erlebald, appear to have had close connections to Charles, 759 the others being unknown. Their relations to the king points towards them acting as witnesses on Charles' behalf, meant to guarantee Robert's concessions to the monks. However, at the same time, the charter also demonstrates that Robert now entered into relations with nobles outside Neustria, allowing him to build up a network of connections within Francia, a network he could use on Charles' behalf to mediate the royal will and carry out the tasks he had been charged with by the king.

The absence of diplomas issued for Robert from 907 to 910 contrasts with the situation after Charles' acquisition of Lotharingia. From 912 until 919, the *marchio* received seven diplomas for his abbeys and intervened in another three for the


<sup>755</sup> DDChS 65 and 9.

<sup>758</sup> Recueil Robert et Raoul, N° 46, 178−181.

<sup>759</sup> For Erlebald, see above. Odilard intervened in 905 together with Bishop Rudolf of Laon on behalf of Charles' notary Ernustus (DChS 51). Almanus may be identical with Count Altmar of Arras, which would make him the third.

churches of Tours and Liège.760 The king appears to have reacted to the structural changes in the realm and in the circle surrounding him by publicly demonstrating the accord between him and the most powerful noble of the realm. The two diplomas issued for the church of Liège, showing Robert intervening together with Reginar, can, in this context, be read as a demonstration of Robert's importance and the care Charles attributed to manage the relations between him and the *marchio*. They are the only examples of Robert intervening on behalf of a church, not only outside his own zone of influence, but also the only examples of Robert intervening together with other nobles.761 By these diplomas, Charles demonstrated that, after the acquisition of the new *regnum*, Robert still occupied the central position at his side, more important than the leading noble from Lotharingia, Reginar Longneck.

Relations between Charles and Robert appear to have peaked (at least for the first time) in 914, when the king issued two diplomas for Saint-Aignan at the *marchio*'s request.762 The accumulation of titles and epithets used to describe Robert reached thus-far unprecedented levels: "Robert, our most beloved *marchio* and abbot."763 More important, however, are the circumstances under which the diplomas appear to have been issued. Only two days after these two, another one was delivered, this time for Richard's son Hugh the Black, according to Philippe Lauer issued at Collège (Sarthe).764 If Lauer's identification is indeed correct, this would mean that in June 914 Charles, together with Robert and Richard's son, was on a voyage to the very west of his realm, close to the Breton frontier. Since the death of Alan the Great in 907, Brittany had lived through troubled times.765 Alan appears to have been succeeded by Count Uurmaelon, 766 whose death in 913, along with the destruction of the abbey of Saint-Gwénolé of Landévennec in the centre of the Breton royal lands, marks the breakdown of the internal order of the *regnum*. In the wake of these events, the Breton clergy appears to have fled to Western Francia, evacuating their relics, while Brittany lay open to the incursions


of the Northmen. These events can hardly have been left unanswered by Charles and Robert and the deliverance of the three diplomas for Robert and Hugh may be the sign of their reaction: a campaign with troops, composed not only of Robert's own men, but also of contingents from Burgundy and Francia to stabilise the frontier and possibly also make territorial gains. This fits well into the picture: before the campaign, Robert had already shown an interest in the Breton frontier. In November 912, Bishop Isaias of Nantes witnessed a charter issued by the *marchio*767 and, in the following year, one of Robert's viscounts, Fulk of Anjou, became count of Nantes, probably after the city had been taken from the Northmen.768 Finally, in 921, Robert concluded a treaty with the Northmen operating on the Loire, conceding them Brittany and the *pagus* of Nantes.769 Flodoard's account of these events seems to indicate that Robert at least claimed a certain kind of authority not only over Nantes, which can be explained by his viscount Fulk being count there, but also over Brittany itself. If, in 914, a campaign to the Breton frontier had indeed taken place, this would explain how and when such a claim could have been created. Apart from this (possible) campaign and the two diplomas for Robert, there is also another sign that the cooperation between the king and the *marchio* had reached a very high level. A private charter issued by Robert in March 914 was witnessed by his son, Hugh the Great, as "Hugh, son of the abbot and count, who, after himself, was already given his *honores,*" that is to say, he had been confirmed as his father's heir.770 Hugh's succession in Robert's *honores* is repeated by a charter supposedly issued two months later, adding that he would hold all of the *honores* with Charles' consent.771 This second charter, however, appears to be a forgery produced in Saint-Martin of Tours.772 While this devaluates the charter as a source, we may ask ourselves why the monks would introduce such a phrasing if there had not been a tradition at the monastery recording Charles' role in the succession. Following this reading, it would seem that, early in 914, Charles had indeed confirmed Hugh as Robert's successor, undoubtedly a sign of his favour.

The highest density of royal diplomas for Robert—after 903—is recorded even later, between 917 and 919, when five diplomas were issued by the king. The first two of these diplomas continue the accumulation of epithets and titles attributed to Robert. He is the "abbot and dearest *demarcus* and our most loyal [*fidelis*],"773

767 Recueil Robert et Raoul, N° 47, 181−186 (11th November 912).

768 Werner, Untersuchungen I, 286−287 (31st March 914). The charter is signed by *Fulconis Namnetensis comitis et Andegavensis vicecomitis*. In Robert's charter, dating to 11th November 912, Fulk signed without any titles, in contrast to the counts featuring in it before him.

769 Flodoard, Annales 921, 6.

770 Werner, Untersuchungen I, 286−287 (31st March 914), 287: *Signum domni Hugonis, filii sui abbatis et comitis cui post ipsum iam sui honores dati erant.*

771 Recueil Robert et Raoul N° 48B, 187−199, 192: … *filius noster Hugo, cui post nos cum seniore nostro rege Karolo omnes honores nostros impetratos habemus…*

772 See comment of Jean Dufour, Recueil Robert et Raoul N° 48B, 188−191.

773 DChS 89 (28th May 917), 201: *…abbas […] atque demarcus carissimus ac fidelissimus noster Rotbertus..*.

the "venerable *marchio*, namely our realm's counsellor and aid, as well as abbot"774 and even, strangely enough, "our kinsman," who is united with Charles in a prayer service.775 The two diplomas again mark Robert's special position at Charles' court, underlined by the third one, issued on Ascension Day at Compiègne.776 Interestingly enough, these diplomas fall into the same period as the diplomas indicating the special relation between Charles and Hagano. 777 It is hard to tell whether these two developments were related to each other. Did Charles try to reinforce relations with Robert to prevent him from being irritated about Charles' intimate? This would imply that the king was aware of the dangers of his favour towards Hagano. Indeed, despite diplomas for Robert still being issued, from May 918 onwards their relations seem to have cooled down. The exuberant epithets used in the first diplomas disappear from May 918 onwards, leaving Robert to be depicted simply as "venerable abbot"778 and "our venerable count and *marchio.*"779

This, however, does not mean that irreconcilable differences had now developed between the two. Flodoard's account of the events at Soissons speaks of "almost all of the counts of Francia" who decided to abandon the king because of his councillor Hagano. 780 Robert is not mentioned, although he was undoubtedly present at the assembly. This is revealed by a royal diploma issued on that occasion at Robert's request for Notre-Dame of Morienval.781 The diploma, naming the *marchio* as its abbot, is the first sign that Robert had taken control of the abbey. Karl Ferdinand Werner assumed that Morienval had passed from its last known abbot Theoderic, the count of Vermandois, directly to Robert. 782 This assumption is based on the record of the *Gallia Christiana*, which names Theoderic and after him Robert as lay abbots of Morienval.783 This statement, however, is solely based


on Charles' diploma, in which Theoderic is named as a former abbot.784 Hence, it is possible that Robert did not directly follow Theoderic as abbot, but was given the abbey at a much later point, maybe even only at Soissons in 920. In any case, the diploma, not only being issued at Robert's request but also confirming one of his brother's diplomas,785 was a symbol of the continuing cooperation between Charles and Robert and was directed towards the assembly of the counts. This also allows us to cast some doubts on Flodoard's accounts of the events: since the monk does not mention Robert at all in the context of Soissons while the diploma issued at that moment suggests that Charles still enjoyed Robert's support, should we really believe him when he tells us that the reconciliation between the king and the nobles was solely Archbishop Heriveus' doing? It would rather seem that Robert had also been involved in these negotiations, taking the king's side.

For the outcome of the negotiations, Robert certainly was the decisive factor. Mostly isolated at the beginning of Charles' reign, by now he had built up a network covering most the West Frankish realm. We have already seen how he had become close with William the Pious, although this link died with William. 786 By 920, other connections were more important. Robert's daughter Emma had married Richard's son Raoul, tying the two families close together.787 This marriage was probably the result of a development that had started around 907, when Richard returned to court, as shown by a royal diploma issued at his and Manasses' request for Bishop Argrim of Langres,788 a development that led to the cooperation of the two *marchiones* in the campaigns of 911 and 914. The marriage between Raoul, and Emma as well as Richard's private charter of 918789 mark the end of this convergence, leading to an open alliance that found its first target in William the Younger, from whom Robert and Raoul took the Berry790. Probably around the same time, Robert and Richard's sons also appear to have visited Remiremont, where they made an entry in the *Liber Memorialis*. 791 And there were other important ties Robert had forged over the past decades. Heribert II of Vermandois mar-

*Carlomanni, quae coenobio igne cremato exusta fuerant, XIII cal. Febr. indict. VIII, anno XXVIII, regnante Carolo, redintegrante XXIII, largiore hereditate adeptae VIIII; hic est annus 920.*

784 DChS 105, 251: *Praetera, Theoderico comite venerabili et abbate jam dicti monasterii depraecante postea, dignus memoria rex Karlomannus, qui noster fuit frater, saepedictae congregationi in pago Silvanectensi dedit de jure suae proprietatis fiscum qui vocatur Fraxinedus, et est situs super flumen Altona, in quo habentur mansi septuaginta in his locus conjacentes...* Carloman II's diploma issued on that occasion (DCmII 90) is not preserved. The phrasing of the *Gallia Christiana* recording Theoderic corresponds exactly with Charles' diploma, as does its account for Robert (apart from Robert's genealogy which is based on a later tradition).


ried another of Robert's daughters, possibly the Adelaide792 appearing in Charles' diploma of 907.793 Heribert appears to have witnessed one of Robert's charters for Marmoutiers794 and, like Bishop Abbo of Soissons, was also involved in the grant of the abbey of Croix-Saint-Ouen to the monks of Saint-Germain-des-Prés, where Robert was lay-abbot. 795 While not necessarily a sign of them being allied— Croix-Saint-Ouen was situated in the Mérezais, a county under Heribert's control796—the diploma issued at that occasion certainly shows them cooperating in an important case. The last of the links forged by marriage concerned Robert's son Hugh, who was married to the daughter of Count Roger of Maine and Rothild, hence the sister of Count Hugh of Maine, who we have seen among Charles' allies in 900.797 The connections to these three, Richard, Heribert II and Hugh of Maine, as well as those to the counts from Francia witnessing in Robert's private charter from 906,798 indicate that Robert's position within the West Frankish nobility had changed much in comparison to the beginning of Charles' reign. He had now managed to construct links, even forge alliances, to those men or their successors, who had originally been key members of Charles' own network, supporting the king against the ambitions of the *marchio* before his integration into the rule in 903. Robert's network was certainly not constructed with the purpose of opposing the king but, without a doubt, the consequence of his position at Charles' side that made the *marchio* a potential partner for all those who wanted to influence royal politics. Developed over decades, it was an expression of Robert's special importance within the realm and the circle around the king.

After the reconciliation with the West Frankish nobles, Charles' diplomas mainly betray his efforts to stabilise his relations with them. The central role in these efforts seems to have fallen to the circle around Abbo of Soissons, Stephen of Cambrai, Raoul de Gouy and Hagano, for whom the central diplomas were issued during this period.799 Robert's role, however, should not be underestimated either. In the case of the episcopal succession at Liège after the death of Bishop Stephen, 800 Robert seems to have participated in the assembly dealing with the case of Hilduin801 and, therefore, supported Charles' politics concerning the siege.


A royal diploma issued in 921802 further underlines their continuing cooperation. This is not to say that their relations were not subject to increasing tensions. The decline in the use of epithets and titles to describe Robert in the royal diplomas that set in around 918 is also visible in this new diploma, addressing Robert merely as "venerable abbot"803 of Saint-Amand, indicating that the rift between the two was continuing to grow.

Yet even when Charles gave Rothild's abbey Chelles to Hagano and thereby violated the interests of a family now closely tied to Robert's, the relations between the two do not seem to have been irrecoverably damaged. According to Flodoard, the events now unfolded in a number of steps: first, Charles gave Chelles to Hagano. The king having returned from Lotharingia to Laon, Robert's son Hugh met with Heriveus' *fideles* and certain counts at the *villa* of Fismes after Easter. After this meeting, he then advanced on Laon, from where Charles withdrew, accompanied by Heribert II and Hagano, *ob Haganonis amorem*, because of his love for Hagano. Hugh now met with Gislebert and both were called to a *colloquium* with Robert. Soon after, Charles returned with Lotharingian forces and started to devastate property of the church of Reims, marking the beginning of the armed conflict.804 Flodoard's account reads like a progressive escalation. The first meeting of Charles' opponents reads like the creation of a threatening posture, meant to exert pressure on the king. Charles evades this pressure by removing himself from Laon. This is the crucial moment in the development: the king refused to give in and later even started to attack his opponents. The consequence of this royal policy and the ultimate point of no return, but certainly not the original intention of the nobles, was Robert's coronation on 30th June 922,805 about two months after Hugh's initial advance after Easter.806 The decision to depose Charles was certainly

*resisteremus.* As discussed above (chapter III.2.1.6), the use of *marchiones* here refers to the magnates from the Western realm, Robert and Richard, the only nobles carrying that title in the royal diplomas after Reginar's death in 915.


<sup>802</sup> DChS 110 (11th June 921).

<sup>803</sup> DChS 110, 265: *…venerabilis abba […] Robertus…*

<sup>806</sup> See chapter III.2.3, n. 804. Easter 922 concordes with 21st April.

not a simple one. Robert's only preserved diploma807 is, as Geoffrey Koziol expressed it, "a coded defence of Robert's action,"808 depicting his coronation as a necessity for the wellbeing of the entire realm due to "indispensable reasons."809 Therefore, in Robert's view, Charles' deposition had not been inevitable but was the last resort. Up until the last moment—the king withdrawing from Laon and taking up arms—he was interested in a continuation of their cooperation. In the end, it was Charles' decision not to respond to the demands of the nobles but to pursue, according to Flodoard because of his love for Hagano, his own policy. With this decision, the basis of their cooperation was shattered.

#### **III.3 Conclusion: The development of Charles' network and its dissolution**

The beginning of Charles' rule was marked by the dominance of the circle of his old supporters from the fight against Odo, most importantly Archbishop Fulk of Reims, Heribert I of Vermandois and with his mother Adelaide probably also playing a central role. Despite the dominance of these men and women, his reign was soon acknowledged by the most powerful men of the realm who had either opposed his claim or pursued their own policies during the years preceding his final ascent to the throne. Robert of Neustria took the leading role here, with Richard the Justiciar and William the Pious following him. However, Robert, in particular, perceived this situation as unsatisfactory. He had been the key figure of Odo's later years, the right hand of the king, a position that now was held by Fulk. During these years, Charles' emphasised his right to rule, although not by underlining the legitimacy of his claim based on his Carolingian descent—in fact, his policy on this matter did not differ from that of his predecessors—but by demonstrating to the realm that he was willing to act like a true king, a protector of the Church. To do so, he used his connections to the nobles of Septimania and the Spanish March, with whom he organised the assembly of Tours-sur-Marne in 899, where he did exactly that: restored ruined churches and protected their interests. While these measures probably had no influence on the actual conditions in the distant South, the assembly of Tours-sur-Marne was, nevertheless, a powerful signal to the realm: Charles was the king, and he was willing to act as one.

The first crisis developed in 900 when Archbishop Fulk was murdered, yet the now vacant position at the king's side was not given to Robert, but to Richard. Robert now left the court and both sides mustered their forces, although most likely without ever actually taking up arms. Robert's integration into Charles' rule

<sup>807</sup> DRoI 1 (25th January 923).

<sup>808</sup> Koziol, Robert, 250. See also chapter VI.4.

<sup>809</sup> Koziol, Robert, 250−252. The diploma's address (DRoI 1, 8) reads *Notum autem manet sagacitati omnium fidelium nostrorum, procerum scilicet Francorum, qualiter per divinam clementiam, causis necessariis existentibus, omnium favore principum ad regni gubernacula moderanda regie majestatis sceptra suscepimus.* 

followed about three years later, after Adelaide's death, when the relations between the king and Richard seem to have deteriorated severely. At this point Richard disappeared from the court while his position was taken over by Robert, who now became the leading noble of the realm up until the end of Charles' rule. This, however, did not mean that he dominated the circle around the king. On the contrary, members of the group of Charles' old supporters, like Bishop Anskeric of Paris, and their successors, such as Archbishop Heriveus of Reims and Heribert II of Vermandois, can be found present at court during the entire reign, occupying key positions, like the new archbishop of Reims, who succeeded Anskeric as archchancellor. Charles' contacts with local nobles outside Francia, however, remained limited. Just like under his predecessors, petitioners and recipients from Aquitaine are very rare, again with the notable exception of Septimania and the Spanish March. In these latter cases, however, the diplomas issued for nobles from the regions were, with few exceptions, concentrated on the two assemblies at Tourssur-Marne in 899 and 922. In the other regions, most notably in Neustria and Burgundy, communications between the king and the local nobility seem to have been channelled through the local magnates, Robert and Richard respectively. The relationship between Charles and these magnates was, in general, characterised by mutual cooperation, with the king testing out his room for manoeuvre on different occasions. In the early conflict with Robert, Charles constructed a network of alliances keeping the *marchio* in check; in Richard's case, he directly intervened in his zone of influence; in Aquitaine, he later favoured William the Pious' opponents around Raymond of Toulouse. All in all, Charles' rule during these years seems to have been undisputed and, from 903, onwards stable.

While between 903 and 911 changes in the network took place only gradually old members died and were replaced by new ones, as in the case of Theodulf 's replacement of Anskeric as bishop of Paris, although not as archchancellor—the acquisition of Lotharingia was followed by major changes in its overall composition and the addition of Lotharingian nobles to the royal contacts. In his new *regnum*, Charles' rule was based on the cooperation with the already-dominating elites, Reginar Longneck, the Matfrid family and the church of Trier, who had already rivalled for influence at court under Zwentibold and Louis the Child. However, in addition, Charles also established contacts with nobles like Bishop Stephen of Cambrai and Count Ricuin of Verdun who had been staunch supporters of Zwentibold but had lost their influence at court under Louis. Up until his death in 915, Reginar and his network of personal contacts appear to have played a central role in Charles' new network. His death marked a slow turn towards the Matfrid family and the church of Trier, with his son Gislebert and his old party remaining important at least up until 916, when they participated in large numbers at the assembly of Herstal. While the acquisition of Lotharingia resulted in distinctive changes in Charles' network and his itinerary—in addition to Francia, after 911 he also travelled within Lotharingia—its overall impact on his relations to the Western nobles appears to have been rather limited. Charles

did his best to integrate Robert in particular into his rule and demonstrated publicly that Robert's position as the most important noble remained unchallenged. Diplomatic contacts with the Western realm remained on a constant level and his other supporters from earlier on remain visible and took part in Lotharingian affairs, such as the assembly of Herstal in 919 and the episcopal succession of Liège. These participations are not the only signs that Charles intended to merge his two *regna* into one. The separate Lotharingian chancellery, founded under Louis the Child, ceased to exist. Archbishop Heriveus of Reims remained sole archchancellor, later succeeded by Roger of Trier. Count Wigeric, married to Charles' niece, became the new count of the palace, replacing William the Pious in this position. Nobles from both *regna* can be found acting together in royal diplomas, which underlines the existence of strong contacts and common interests across the borders.

The last years of Charles' rule saw the formation of a new group close to the king, centred on the bishops Abbo of Soissons and Stephen of Cambrai as well as Count Raoul de Gouy. First appearing alone in the royal charters, from 920 onwards they can be found acting together. While new to the circle around the king, they appear to have had exceptionally good contacts to the rest of the nobility, especially visible in the case of Abbo, who seems to have played a key role in the relations between Charles and the nobles in early 920 and later became Robert's and Raoul's archchancellor. While this group demonstrates that Charles was able to further nobles of his choosing and to influence the composition of the circle around him, his liberty in doing so was also subject to clear restrictions. Archbishop Fulk's presence at his side had prohibited him from granting Robert the position he desired; while after the acquisition of Lotharingia it was the *marchio* whose interests the king had to take into consideration. These same limitations also become clear when considering Charles' influence over episcopal successions. While there are a number of examples of him furthering his own interests by promoting his relatives and associates, the case of the succession at Liège also demonstrates that, when encountering serious resistance, he had to ally himself with groups rivalling his opponents, in this case the Matfrid family. Here, Charles appears to have gone even further, tying the family close to him by means of marriage: his daughter Ermentrude married Gottfrid, the son of Gerhard.

His own family members certainly played an important role for Charles. His mother Adelaide's influence in royal politics becomes apparent from the number of her interventions in Charles' diplomas. On the other hand, Charles' first wife, the Saxon noblewoman Frederuna, remains mostly invisible. Intervening together with the highest nobles in 907 shortly after their marriage, in the sources she is mentioned almost exclusively only after her death, when Charles put considerable effort into establishing prayer services for her memory. This points to the close relation they had enjoyed and her importance for him; but her political influence remains indeterminable. The same also applies for his second wife Eadgifu. While being politically very active in her later life, during Charles' reign, the sources make no mention of her.

Close to the king, although not a member of his family, was Hagano, the man about whom the sources report that Charles lost his crown because of favouring him over the other nobles. Descending from a family based on the Middle Rhine in the East Frankish-Lotharingian border region, he seems to have made his first appearance in 916. His ascent appears to have taken place in the wake of Frederuna's death, when he acted as intercessor on the deceased queen's behalf in a number of diplomas and was even included in a prayer service with the king, the queen and her brother. Only after the settlement with the nobles following their protest against Hagano in early 920 is he shown acting together with other nobles. It would seem from this development that the protest of the nobles was not directed against Hagano's presence *per se*, but against the form of his presence at Charles' side. In fact, preceding Hagano, we can already trace other men who seem to have derived their position from their special relationship with the king yet who were accepted by the highest ranks of the nobility. Hence, the problem with Hagano appears to have been the way he and his relation to Charles was perceived by the nobility. The king responded to this problem by trying to introduce his intimate to the nobles and to integrate him into the highest ranks of the nobility. However, despite these efforts, in 922 some of the leading nobles rebelled against Charles. Yet, the crucial point in the breakdown of their relations appears to have been less related to Hagano and the circumstance that Charles violated the interests of a family close to Robert, but that Charles refused to negotiate over the subject at hand.

The rebellion against Charles certainly was not the result of a sudden development but points to a large crisis in the relations between the king and the nobles. Some scholars have argued that this crisis already commenced before 920,810 when, according to Flodoard's History of the church of Reims, the Hungarians devastated Lotharingia. Charles called the Frankish nobles to aid, but only Archbishop Heriveus responded by mobilising his forces.811 Flodoard's account, however, is not without problems, the first being the dating of the event. It seems to correspond with the first entry in his *Annales*, where he reports a Hungarian invasion in Lotharingia.812 However, the *Annales* do not mention Charles' call to arms and it seems possible that the *Historia* actually refers to an invasion taking place in 917.813 Furthermore, certain doubts can be issued whether Flodoard's account in

<sup>810</sup> Eckel, Charles, 107, Parisot, Royaume, 628. The connection of these events with the later rebellion has been questioned by Schmitz, Heriveus, 80 with n. 86. According to him, the nobles did not support Charles out of self-interest or a lack of interest in the defence of the realm. In this context, it is worth pointing out the campaigns against the Northmen in 911 and (possibly) in 914 which saw Robert and Richard participating.

<sup>811</sup> Flodoard, HRE IV, 14, 407.

<sup>812</sup> Flodoard, Annales 919, 1.

<sup>813</sup> For example Annales S. Medardi 917, 520 (*Hungri primum Rhenum transierunt et usque Burgundiam pervenerunt.*), Annales S. Vicentii Mettensis 917, 157 (*Ungarii primitus regnum Lotharii ingressi sunt.*)

the *Historia* is entirely accurate on this matter. While Flodoard was biased against Charles, 814 Heriveus is always depicted in a positive manner.815 For example, in the chapter describing Heriveus' appointment Flodoard provides an account of Heriveus pontificate: "While he occupied himself with enthusiasm with the spiritual affairs, the temporal goods flowed from all parts, and he dispensed with them with admirable prudence. He conferred the administration of the bishopric to able officers while he busied himself incessantly to the prayer."816 This reads much less like an objective account than the description of an ideal bishop, especially when taking into consideration that in the preceding passages Flodoard used quotes from the *Liber Pontificalis* to describe Heriveus' virtues.817 Furthermore, in this same paragraph, Flodoard remarks that Heriveus recuperated property that his predecessors had given away,818 failing to mention that, very much like these same predecessors, Heriveus also gave property of the Church to his own kinsmen.819 This same method of withholding information while depicting Heriveus as an ideal archbishop can also be found some chapters further on when Flodoard turns to the events of Soissons in 920: "The following year, almost all of the Frankish *optimates* renounced their king Charles at Soissons… This loyal and pious bishop, always firm in times of danger, courageously took care of the king, leading him from this place and bringing him with him to the city of Reims; and he accompanied him and followed him everywhere just until he had brought back these counts to him and had restored him to his *regnum*."820 Again, Flodoard's account reads like a eulogy for Heriveus, emphasising the archbishop's virtue while withholding parts of the history: the central roles Robert of Neustria and Bishop Abbo of Soissons certainly also played in mediating between the king and the nobles.

Hence, Flodoard's bias towards Heriveus and his habit of withholding certain information may also apply in the case of Charles' call to arms against the Hungarians. While it seems unlikely that Flodoard invented the entire affair, it does

and Continuatio Reginonis 917, 155 (*Ungarii per Alemanniam in Alsatiam et usque ad fines Lothariensis regni pervenerunt.*). See also Schmitz, Heriveus, 80 with n. 86. On the invasion of 917 see Lüttich, Ungarnzüge, 66–68. It is worth pointing out that neither of the sources cited above, nor the later chronicle of Herman of Reichenau (112), mention a Hungarian invasion in 919. On Hungarian invasions in general, see also Fasoli, Incursioni; Fasoli, Points and Kellner, Ungarneinfälle.


seem very likely that he exaggerated or misinterpreted the failure of the nobles to respond to the king's call. For example, Charles would probably first have called upon Lotharingian forces to respond to the invasion. Robert, on the other hand, might have been unable to respond to the call because of the threat still posed by the Northmen—in 919, Flodoard tells us that they once again devastated Brittany.821 And finally, given his account of Heriveus' role in 920, it seems rather likely that other nobles also had responded to the call. Of course, these are only assumptions based on Flodoard's way of describing Heriveus' episcopate. Especially when taking into consideration that after the reconciliation the nobles continued to cooperate with Charles, it nevertheless seems inappropriate to take Flodoard's note of the Hungarian invasion as a sign for any deeper problems between Charles and the Frankish nobles at that time.

Therefore, the first indication of a rift between Charles and the nobles seems to have been the confrontation at Soissons early in 920, leading to him being abandoned by the Frankish counts up until the mediation of an agreement seven months later. This event, while without a doubt revealing major problems in the relations between the king and parts of the nobility, should not be overemphasised either. On the one hand, the key figures of the Western realm, Robert, Richard, Heriveus and Abbo did not belong to the opposition that had formed against Charles, nor did the bishops.822 On the other hand, after an agreement had been reached, the nobles again cooperated with the king as his diplomas, featuring a rather large number of bishops and counts, as well as the assembly concerning the succession at Liège823 and Charles' entourage at the meeting with Henry the Fowler at Bonn824 demonstrate. Tensions definitely remained and the nobles most probably were still vigilant about the king, yet nothing indicates that there was an inevitable development leading from the protest of 920 to the rebellion of 922.

The keys to the dissolution of Charles' network in 922 were certainly the illness and death of Heriveus and the breakdown of relations with Robert, thus depriving Charles not only of two of his most powerful and influential supporters, but also making the latter his opponent. Robert's turn from cooperation to hostility must have placed enormous pressure on Charles' network, since many of his contacts also entertained good relations with Robert.825 Thus, Raoul, Richard's son and successor in Burgundy, was married to Robert's daughter Emma. Yet while he decided to support his father-in-law, 826 Heribert II, likewise married to one of

822 For Robert and Abbo, see above. Richard's position is not known, yet it seems to me that Flodoard's phrasing *Francorum optimates* (see chapter III.3, n. 820) and *pene omnes Franciae comites regem suum Karolum … reliquerunt* (Annales 922, 2) does not point to an involvement of the *marchio* from Burgundy. These same phrasings also exlude the bishops from the opposition against Charles.


<sup>821</sup> Flodoard, Annales 919, 1. See also Chronique de Nantes, 81−82. See also Koziol, Politics, 525.

<sup>823</sup> See chapter III.2.1.6.

Robert's daughters, remained, for the time being, on Charles' side.827 The question over whether to support Charles or Robert must have been a difficult one for the individual nobles. Private charters from the regions south of the Loire dating after the regnal years of Charles even after his deposition reveal the division of the nobles over this question. As Wojciech Falkowski has shown, there were sometimes differing opinions on the matter, even within the same monasteries since their cartularies contain charters dating after both kings.828 North of the Loire, in Neustria, Burgundy and Francia, no such datings are preserved, yet there was certainly division among the nobility: Heribert II was not the only one continuing to support Charles. Laon had remained loyal to the king and had to be conquered by Robert,829 probably indicating that its bishop, Adelelm, had joined the king's side. On the other hand, the position of Count Roger of Laon is less clear. Given the power of the bishop at Laon,830 the circumstance of the city's remaining loyal to Charles does not necessitate that Roger had done likewise. In any case, by 923, when he took part in the battle of Soissons on Robert's side,831 he had opted for the rebels. His support for Robert might have come at a price: in 925, he appears as abbot of Saint-Amand, 832 formerly held by Robert, who might have given up his abbacy before his death.833 Finally, like Roger, Bishop Abbo of Soissons can be made out to have joined Robert's side at an unknown point, in his case before 25th January 923, when he appears as Robert's archchancellor.834

The turning point in favour of the rebels, and in the support Charles still enjoyed in Francia, seems to have been reached around Pentecost.835 Just before, when Charles' and Robert's armies had been encamped in close vicinity to each other, nobles from Charles' army negotiated with Robert and Raoul. 836 Charles' reaction was twofold: he moved his army to the vicinity of Reims837 and tried to act like a king, demonstrating that he was the protector of the Church and willing to compensate those loyal to him. To do so, he reenacted the assembly that had taken place in June 899 at Tours-sur-Marne: at the same place, around the same time of the year, he again issued numerous charters for recipients from Septimania, the churches of Narbonne and Girona, *fideles* and abbeys in the region,838 as well as a


diploma for the abbey Saint-Thierry of Reims,839 granting Bishop Guigo, the only noble actually attending, "what little is left at the disposition of the royal power" because of the fidelity shown by him.840 The message was clear and might have drawn more nobles back to his side, yet after his attack on Reims failed and Laon was captured by Robert's men, the latter's forces "increased each day while those of Charles diminished."841 This was probably the moment when nobles like Heribert II of Vermandois842 went over to Robert. The opposition against Charles had now reached a point when it became strong enough to claim the right to choose a new king. With Charles having withdrawn to Lotharingia, the "Franks chose Robert as *senior* and committed themselves to him", thus making him their king.843

The Lotharingian nobility seems to have been as divided as the West Frankish. While Flodoard, without going into any further detail, repeatedly emphasises the role of Charles' Lotharingian connections,844 he also reports that, on two occasions, Lotharingian nobles left Charles' army to return to their homes845 and that on another some of them negotiated separately with Robert846 only to rejoin Charles soon after.847 Among those remaining loyal to Charles appears to have been Count Theoderic, who, shortly after Charles' failed advances on Reims and Laon, received a diploma from the king.848 Likewise, Archbishop Roger of Trier seems to have remained loyal to Charles up until this moment since he features as archchancellor in this very same diploma. Count Gislebert, Reginar Longneck's son, and Count Otto, son of Count Ricuin of Verdun, on the other hand, were obvious allies for the West Frankish rebels since they were already in conflict with


<sup>839</sup> DChS 115 (31st May 922, Tours-sur-Marne).

the king.849 And indeed Gislebert is mentioned to have met first with Hugh and then with Robert just after the rebellion had broken out.850 Later in 922, Hugh led an advance to relieve Gislebert's *castrum* at Chièvremont, which was besieged by Charles. 851 However, Gislebert's position in the rebellion is less clear than it seems on first glance. Indeed it appears that the king and the count had concluded a peace treaty at some moment before the siege—albeit not a lasting one.852

The rebellion, as we have seen, cut deep through Charles' network of support. Setting Robert aside, the extent of the breach between Charles and the nobles becomes clearest when considering the positions of Heribert II of Vermandois and Bishop Abbo of Soissons. They both had belonged to Charles' inner circle and Abbo even appears to have been one of the most influential nobles at court. After the battle of Soissons, despite the death of Robert, Charles was abandoned by his remaining followers while his opponents refused to return into his peace.853 As in 897, when Charles had faced a similar situation, he now turned to his last resort: the Northmen of the Loire.854 However, as in 897, this alliance did not help to further his cause. The Frankish nobles, having called upon Raoul of Burgundy, blocked the Northmen's path and soon after Charles was captured and imprisoned by Heribert II.

A comparison between the attitude of the Lotharingian nobles to Charles and that of those from the West Frankish realm seems hardly possible due to the nature of Flodoard's account. While emphasising the importance of the Lotharingians for Charles, he fails to mention those West Frankish nobles who continued to support the Carolingian but concentrates on the leading figures, Robert and his son Hugh. As it seems, Charles faced the same problems in Lotharingia that he was facing in the Western realm. From early on, he was losing support and was finally abandoned after the battle of Soissons. However, there also appears to be a distinctive difference. Nobles from Lotharingia, not least Gislebert, returned into the king's peace even after they had turned away from him. The reason for this difference was probably their attitude towards Robert. Robert's own network extended over Neustria, Francia and Burgundy and it was the nobles from these regions who chose him as their king.855 In Lotharingia, on the other hand, his influence and the support for his claim, were rather limited. Here, he was not perceived as king but as a potential ally in private conflicts with Charles. Only once Charles was imprisoned, did the Lotharingians look out for a new ruler and again the different interests of the various nobles surfaced. A group of nobles,


<sup>849</sup> Flodoard, Annales 922, 7. On the conflict, see chapter VI.3.

<sup>850</sup> Flodoard, Annales 922, 8.

<sup>851</sup> Flodoard, Annales 922, 11.

<sup>852</sup> See chapter VI.3.

among them Bishop Wigeric of Metz, turned to Raoul while soon after Gislebert and Archbishop Roger of Trier called upon Henry the Fowler.856

The question remains however: what caused this dissolution of Charles' network? In 920, when the Frankish counts first articulated their reproaches concerning Hagano, they challenged royal power but not Charles' right to rule. Charles' relations with Hagano certainly did play a major role both in 920 and in 922, yet the nobles' concerns do not seem to have been directed against the king's companion. The settlement agreed upon seven months after Soissons did not include Hagano's removal from court and there still seems to have been room for negotiation in 922. Yet, the rebellion of 922 reveals a deep mistrust against the king, a mistrust so deep that, even after Robert's death, the nobles preferred to turn to Raoul instead of returning to Charles. Given the length of Charles' reign at that moment—24 years—and the fact that the rebellion broke out only in 922 and not in 920, this mistrust must have been a rather recent development, possibly setting in around 918, when the first signs of a cooling down in the relations between Charles and Robert become visible, and rapidly increasing from 920 onwards. The reasons for this mistrust are probably to be found in Charles' actions in the various conflicts during these years, a question that we will address in our last chapter.

## **IV. Relations with other rulers**

So far, royal politics have been dealt with as the result of the interaction between the ruler and the nobles around him. While in general correct, in some cases this model proves too narrow since there are also other political actors to take into consideration. Late Carolingian royal politics did not stop at the borders of a ruler's realm but took place in the greater Frankish world (and beyond). Noble networks, including contacts of the rulers to nobles, reached over the borders into the neighbouring realms. Next to political or ecomonical interests this also included family ties, last but not least also between the Carolingian rulers. This means that in certain situations the balance between a ruler and the nobles around him could be influenced by relations with actors outside his realm. Other rulers could become allies as well as enemies, they could intervene in the politics of a realm in one way or another. Archbishop Fulk was all too aware of the possibilities and threats this wider framework offered and tried to play it as best he could when leading Charles the Simple's struggle against Odo. Not only rulers, also the nobles themselves could seek support from the outside. Reginar Longneck seeking Charles the Simple's aid against Zwentibold was only one example of how conflicts between a noble and his ruler could provide a window for intervention. This chapter is dedicated to the different effects of this wider framework on the various West Frankish rulers' politics and to their efforts to make use of the possibilities it provided.

## **IV.1 Creating cooperation: Louis the Stammerer**

As the first chapter of the treaty concluded between Louis the Stammerer and Louis the Younger at Fouron states, the *amicitia* agreed upon was only of temporary character because of "some hindering causes."1 This phrasing points to the ongoing problems that had arisen over the succession of Louis II of Italy, whose *regnum* had then yet to be divided under the remaining rulers.2 Louis the Stammerer was pursuing a policy aimed not only at becoming the ruler of Italy, but in particular at aquiring the imperial crown previously held by his father. When he had met with Pope John VIII at the Synod of Troyes, he had made an initial attempt to secure the crown for himself when he asked the pope to confirm his

<sup>1</sup> MGH Capit. II, N° 246, c. 1, 169: *Ut, quia firmitas amicitiae et coniunctionis nostrae modo quibusdam praepedientibus causis esse non potuit, usque ad illud placitum, quo simul ut conveniamus statutum habemus, talis amicitia inter nos maneat Domino auxiliante de corde puro et conscientia bona et fide non ficta, ut nemo suo pari vitam, regnum aut fideles suos vel aliquid, quod ad salutem sive prosperitatem ac honorem regni pertinet, discupiat aut forsconsiliet.*

<sup>2</sup> MGH Capit. II, N° 246, Preamble, 169.

father's order handing the realm on to him.3 It is certain that Louis was not asking for John to approve his rule in the West Frankish kingdom, for he had already won this when the pope had crowned him a short time earlier; rather, he was seeking John's support in succeeding Charles the Bald as emperor.4 Although the pope refused to fulfil the request, the two of them appear to have come to an agreement in private talks shortly afterwards.5 In a letter John later sent to Louis, he refered to a *secretum* which the two of them had agreed upon at Troyes: Louis was to support the pope militarily, while the pope would fulfil his wishes in return.6 The first portion of Louis' aid appears to have consisted of Louis sending Boso to accompany John back to Italy, where he remained for several months.7 As Johannes Fried has convincingly argued, Louis in turn became John's candidate for the imperial crown.8 Louis' imperial ambitions must have been brought up during the negotiations preceding the treaty of Fouron, and it is hardly surprising, given the progress he had already made, that he was unwilling to renounce them. Nevertheless, the overall tenor of the treaty of Fouron was one of friendly cooperation. Both rulers agreed to support each other in the event of Viking incursions9 and to protect the property of the churches of each other's realms,10 indicating that the intention was to develop a wide-ranging spirit of co-operation between the two rulers.

This image is further strengthened when considering yet another chapter of the treaty that was of special interest for Louis the Stammerer. Not only was the succession of his sons Louis and Carloman secured in this passage, but also the succession of any future sons "whom God's mercy might be willing to give him."11 This was a *de facto* acknowledgment by Louis the Younger of the official position of the West Frankish court on the validity of both of Louis the Stammerer's marriages.12 Yet there is even more to the passage. For one, it is stated that all of Louis' sons should be able to hold the paternal *regnum* by right of inheritance, passing over the role of the nobility in the question of succession. This impression is strengthened by the second part of the passage, which committed Louis the


Younger to counsel and aid Louis' heirs.13 This agreement seems to reflect Louis the Stammerer's own problems at the time when he had succeeded his father and faced the displeasure of the leading nobles, who resented his attempts to create his own power base independent from them. Were Louis to die, this treaty would not only secure the succession of all of his sons, but also provide them with a powerful ally against the nobles if need should be.

Two other passages, although certainly in line with earlier treaties,14 appear to have been motivated by current events. Both kings agreed not to receive "any whisperers and disparagers, envious of our peace and unable to bear the realm's being peaceful, wish[ing] to sow quarrels and contentions and discords between us,"15 adding later on that "the peace and quiet of the realm keeps being shaken by rootless men who lack respect for anything and behave like tyrants"; these men were also to be rejected and either "brought to right reason" or pursued by both parties.16 For Louis the Stammerer, these passages, at a minimum, secured Louis the Younger's neutrality in his ongoing conflict with Bernard of Gothia, against whom he prepared a campaign early the following year.17 According to the treaty, Bernard would henceforth find neither support nor shelter in the Eastern realm, while Louis could request his cousin's help should the *marchio* turn there.

Thus, the treaty of Fouron strengthened Louis the Stammerer's position considerably, securing not only the succession of his sons but also ensuring that his neighbour would not interfere in his internal affairs. The agreements, of course, were mutual and would deny Louis the opportunity of taking advantage of any problems Louis the Younger might have had at the same time. However, this appears to have been a price he was willing to pay, especially considering that Fouron was only a preliminary meeting, held in preparation for a second, larger one that was to include Louis the Younger's brothers Carloman and Charles. 18 It seems that the intention was to create a large network that included all the Carolingian rulers

13 On the development of the phrase *consilium et auxilium* applied here, see Devisse, Consilium.


and was designed to stabilise their realms by solving the Italian problem as well as by cooperating against rebellious nobles.

#### **IV.2 Carolingian networks: Louis III and Carloman II**

Louis the Stammerer's early death led to exactly the situation that he had tried to avoid with the treaty of Fouron. His sons became objects of the rivalries between the noble factions around Hugh the Abbot and Gauzlin, which in turn led the latter to invite Louis the Younger to intervene in the Western realm and thus to the support of an external power for a group within the realm. The situation is not as clear cut as it would appear from this characterisation, however. In his last actions, Louis the Stammerer had made it clear that he intended only his oldest son Louis to succeed him while the faction around Gauzlin aimed at realising the treaty of Fouron.19 It could even be argued that by his intervention in the West, Louis the Younger acted in accordance with the treaty by coming to the aid of his cousin's sons and ensuring that both of them would succeed their father. However, his motives appear not to have been altogether altruistic. Hincmar sees Louis the Younger's wife Liutgard as a driving force, who, he insinuates, held ambitions to take over the entire Western realm.20 While this part of Hincmar's report might simply have served to denounce Gauzlin21 by grossly exaggerating the danger to the Western realm and Louis the Stammerer's sons, it nevertheless points towards another result of Louis' intervention: the Western realm lost the part of Lotharingia that had been acquired by Charles the Bald via the treaty of Meersen in 870 to Louis the Younger.

It is interesting to note the difference between the accounts of two of the main sources on the subject. According to Hincmar, it was Gauzlin and his allies who offered the *regnum* to Louis,22 while the *Annales Vedastini* report that Hugh the Abbot offered Lotharingia to Louis, not to make him come, but to make him leave.23 Whether it was Gauzlin or Hugh who was the originator of the cession, it appears to have been supported by both parties—the price that was to be paid for


<sup>19</sup> See chapters I and II and Werner, Gauzlin.

<sup>20</sup> Annales Bertiniani 879, 238: *Audiens autem hoc uxor illius, satis moleste tulit, dicens quia si illa cum eo uenisset, totum istud regnum haberet.*

Louis' intervention. This might have been due to the terms under which the western part of Lotharingia became part of Louis the Younger's realm. Hincmar tells us that after Louis the Younger's death, "the leading nobles from the part of this king's *regnum* which had been given to this Louis as a lease"24 were coming to meet Louis III. This phrasing is repeated shortly onwards when Hincmar reports how Hugh the Abbot sought out Charles the Fat, requesting that the emperor "should restore to Carloman, as he himself had promised to do, that part of the kingdom which his brother Louis had received as a lease."25 The two reports specify that western Lotharingia had not been ceded in perpetuity but only temporary—as a lease—and that (as evidenced by Hugh the Abbot's request that Charles the Fat return the region) this lease was probably supposed to end with Louis the Younger's death, at least in the Western leaders' point of view.26 Charles the Fat indeed appears to have been involved in the agreement from an early point,27 probably to avoid any future tensions over the subject. This circumstance is probably the best example of the relations between the various parties involved and their goals. Louis the Younger wanted to make personal gains from his intervention in the West, a demand that was accepted by the factions around Hugh and Gauzlin as well as by Louis the Stammerer's sons. Yet, at the same time, the idea of cooperation that had been so dominant in the treaty of Fouron still wielded a strong influence over the different parties.

That this cooperation was still highly valued was probably due to the common threats the different rulers now had to face. In 879 the Vikings had returned to the


continent,28 in Lotharingia Lothar II's illegitimate son Hugh had rebelled29 and finally, most importantly for the Western realm, Boso had initiated his own coronation at Vienne.30 As Simon MacLean argues, the summits that took place between the various rulers from late 879 until June 880—among them three involving the new kings of the Western realm, Louis III and Carloman II—were meant to settle ongoing disputes and coordinate the actions taken against these threats.31 The measures decided upon unfolded over the following months: first, Louis III and Carloman II, with the support of forces provided by Louis the Younger, launched an attack on the rebellious Hugh. From there, this united army turned south into Burgundy against Boso where it was joined by Charles the Fat. 32 Meanwhile, another army was placed under the command of Gauzlin and other nobles to deal with the Northmen who had installed themselves on the Scheldt.33 The *Annales Vedastini* provide a detailed account of this last campaign which reveals that the extent of cooperation in this theatre of war did not differ from that of the others: the attack led Gauzlin and the others deep into Lotharingian territory—hence into Louis the Younger's realm—where they coordinated their actions with the local forces.34 United against common enemies, the joint armies operated over the borders of the *regna*: this is how far the various kings were willing to cooperate, and shows how much importance they attributed to dealing with these threats.

This cooperation continued over the following years. When Charles the Fat left the siege of Vienne early on, he did so with the approval of Louis and Carloman after both sides had sworn oaths,35 oaths that probably also included agreements about Charles' succession36 and the two brothers' acknowledgment of his claim to the imperial crown. Later, he rejoined the efforts against Boso, sending forces to deal with the remnants of the rebellion.37 And when asked by the pope to release Engelberga, the widow of Emperor Louis II who had been imprisoned to prevent her from supporting Boso, Charles pointed out that he needed the consent of Louis and Carloman before doing so.38 Meanwhile in the north, as the *Annales Fuldenses* 


<sup>28</sup> See chapter V.1.1.

report, Louis the Younger and Louis III held a "fitting meeting" at Gondreville.39 Unfortunately, the annals do not give any details about the discussions. Yet, as Simon MacLean argues, the meeting is reported immediately following to the account of the rebellion of Lothar II's son Hugh, thus making it "fitting" if both kings met to coordinate their measures against this new threat.40 These numerous signs of intense cooperation also cast some light on an occurrence already mentioned above. When Louis the Younger died, Lotharingian nobles approached Louis III to commend themselves to him.41 As we have argued, the western part of Lotharingia had only been temporarily leased to Louis the Younger and was to revert back to the Western realm upon his death. Thus, while it would have been legitimate for Louis III to accept their commendation, he nevertheless refused their offer. Since Charles the Fat had also been involved in the agreement concerning Lotharingia, it seems safe to assume that it had been negotiated that the *regnum* should not automatically revert to the West, but that Charles should return it publicly,42 an act that Louis did not want to anticipate in order to avoid any unnecessary tensions. Fulfilling his side of the bargain, he, nonetheless, appears to have maintained a certain degree of caution. While he did send back the Lotharingian nobles, he also dispatched an army under Count Theoderic of Vermandois, "as if to help them against the Northmen",43 as Hincmar writes. His account, strongly biased against Louis III and the nobles surrounding him,44 may not be entirely reliable in this case, falsely implying that there were also other motives behind this campaign. Nevertheless, it seems reasonable to assume that Theoderic's intervention was not only meant to aid the Lotharingians, but also to strengthen links with the local nobility and to emphasise Louis' claim on his father's and grandfather's lands.

This episode also reveals the problems attendant on this cooperation between Louis and Carloman. While they profited enormously from the aid provided by Louis the Younger and Charles the Fat in their fight against Boso, it also limited their ability to pursue their own interests when doing so would have threatened the foundations of this same cooperation. These limitations become even more apparent when taking into consideration Hugh the Abbot's attempt to regain Western Lotharingia from Charles for Carloman II, soon after Louis III's death:


42 This would also explain why Hugh the Abbot, soon after trying to negotiate the return of western Lotharingia to Carloman, reminded Charles the Fat of his promise. Annales Bertiniani 882, 249: *A quod placitum Hugo abbas, quibusdam sociis secum assumptis, perrexit pro petitione partis regni quam frater suus Hludouuicus in locarium acceperat, ut, sicut ipse Karolus olim promiserat, Karlomanno restitueret.*

43 Annales Bertiniani 882, 246. After Louis had met with nobles from Lotharingia: *[Hludouuicus] scaram hostilem, cui praefecit Theodericum comitem, quasi in adiutorium illorum contra Nortmannos disposuit.*

<sup>39</sup> Annales Fuldenses (Mainz continuation) 881, 96: *Rex cum suo nepote Hludowico apud villam Gundolfi congruum habuit colloquium; inde transiens omne tempus aestivum in Baioaria moratus est.* 

<sup>44</sup> See chapter II.1.

reminded of his own promises, the emperor simply played for time by not commiting himself.45 For him, cooperation was seemingly less important than it had been for Louis and Carloman, especially since by now Hugh and Boso had been defeated. Carloman, on the other hand, was not only weakened by the resistence of the nobles of the northern part of his realm to his rule, but also at the same time confronted with the Northmen.46 Thus, he lacked the power to challenge the emperor and force him to keep his part of the bargain.

#### **IV.3. Structural weaknesses: Odo**

In comparison to his Carolingian predecessors, Odo's dealings with other rulers started under completely different conditions. Whereas Louis the Stammerer and his sons negotiated as equals, at least in rank, Odo's first meeting with Arnulf of Carinthia only took place because the latter summoned him to a *placitum*. 47 According to the *Annales Fuldenses*, Odo "adopted the sensible plan of saying that he would prefer to hold his kingdom in peace by the grace of the king than to rebel in pride contrary to his fidelity, and coming there humbly to the king he was received with grace."48 While certainly the exaggeration of an East Frankish source, its message, nevertheless, concords with that of the *Annales Vedastini*, which adds some more details. They report that Arnulf received Odo honourably at Worms and, after they had concluded an *amicitia*, Arnulf sent Odo back to his *regnum*, asking him to forgive those who had come to him.49 Both sources agree that Arnulf was the higher-ranking ruler of the two. This superiority in rank was certainly a question of power, as Arnulf was already a well-established ruler; yet Arnulf 's Carolingian blood and consequently more legitimate position must also have played a role.50

However, this ignores the question of why Odo would so willingly acknowledge Arnulf as a ruler superior to him in rank. When Odo met with the East Frankish ruler, his own claim to the throne was still challenged by Archbishop Fulk of Reims, Abbot Rodulf of Saint-Vaast and Count Baldwin II of Flanders, who had turned to Arnulf, inviting him to come to the Western realm and take the realm


that was lawfully his.51 By gaining Arnulf 's acknowledgement, Odo was able to eliminate another claimant to the throne and to deliver a serious blow to the faction opposing him. During the negotiations between Odo and Arnulf preceeding the actual meeting, Baldwin abandoned his allies and commended himself to the king,52 a clear sign of the success of this strategy. Apart from this, Odo had even more to gain from Arnulf. As his own ascent to the throne rested on military prowess rather than blood and heritage, and his kingship was being challenged by a part of the nobility and external competitors, he needed to strengthen his position by finding other ways to legitimise his claim. Arnulf 's Carolingian descent made him a source of legitimation, a source that Odo could use to negate the difference between himself and his Carolingian predecessors.53 Just how important Arnulf 's acknowledgment was for Odo is further revealed by his second coronation with a crown sent to him by Arnulf. 54

The extent to which not only Odo but also Arnulf adjusted their mutual relations in accordance with their own necessities is revealed by events during the time of the struggle between Odo and Charles the Simple. In 894, Arnulf agreed to Charles coming to him and acknowledged him as king—something that cost him nothing but allowed him to demonstrate his superiority over a fellow Carolingian king during a time when he needed to strengthen his own position.55 This did not mean that Arnulf now stopped considering Odo a legitimate king as well. Similarly, Pope Formosus had accepted Charles' coronation, congratulating him on his elevation.56 Yet, at the same time, Formosus also wrote to the bishops of the Western realm, asking them to join Odo, to keep an armistice and to work on the restoration of peace.57 If we can have faith in Flodoard's *regesta* of the archbishop's letters, even for Fulk, Charles' coronation had not deprived Odo of his crown. In a letter to the pope, Fulk spoke of "the conflict which was going on between the kings Odo and Charles."58 While Odo's legitimacy did not suffer by Charles being acknowledged as king, his position nevertheless took a severe blow, as Charles' claim was now sanctioned as legitimate by the dominant Frankish ruler. To make things worse for Odo, Arnulf went even further and now supported Charles openly with Lotharingian forces,59 thus shifting the balance of power against Odo and making Charles' side even more attractive for other potential allies.


This was undoubtedly the reason why, the following year, Odo obeyed Arnulf 's summons to Worms although this meant that he not only would have to acknowledge Arnulf 's superiority once more, but that he also ran the risk that Arnulf would try to act as judge over him and Charles, who had likewise been called because Arnulf "wanted to put an end to this misery between them."60 By coming to Worms, Odo risked that the conflict would be settled to his detriment. Odo gambled, but it paid off when Charles did not come to Worms. Bearing generous presents, he "was received honourably and sent back home with delight."61 Thus he was able to publicly demonstrate that his own claim to the throne was as valid as Charles'—or maybe even more valid, since Charles had not turned up—and that Arnulf 's support for Charles had ended.

The whole affair reveals how vulnerable Odo's position had become and how the internal problems of his rule influenced his relations with other kings. The second meeting between Odo and Arnulf was the direct result of the latter's intervention in the affairs of the Western realm, an intervention that forced Odo to again acknowledge Arnulf 's superiority in rank and take an enormous risk that might have severely threatened his position. This vulnerability becomes even more apparent when, after the meeting at Worms, Charles allied himself with Zwentibold, who then intervened in the Western realm.62 In contrast to his father, Zwentibold appears not to have been interested in gaining prestige but in gaining territory, for that was promised to him by his new allies. This difference allows us to draw some conclusions about Odo's military strength in comparison to his eastern neighbour. Odo, faced with the alliance between Charles and Zwentibold, withdrew over the Seine, indicating that his forces were not sufficient to challenge his opponents. Soon afterwards, however, Charles' supporters and Zwentibold fell out with each other, leading to Odo's return and Zwentibold's quick withdrawal to his own *regnum*. Odo was strong enough to deal with an isolated opponent, but not with an alliance between the two.

Thus, Odo's relations with other rulers are revealed to have suffered from two major problems: Arnulf 's superior rank and the conflict with Charles. While Odo was at first able to profit from the former by strengthening his own legitimation via Arnulf 's acknowledgement, later on this same constellation turned against him when Arnulf recognised Charles' claim as legitimate and allied himself with him. Now it was Odo's Carolingian rival who profited. Moreover, the conflict between Odo and Charles led to Zwentibold's intervention in the Western realm,

<sup>60</sup> Annales Vedastini 895, 75: *Constricti vero hi qui sequebantur Karolum*—*nam Odo rex eis quicquid in Francia habuerant tulerat—Burgundiam acriter depopulati sunt. Venitque clamor eorum ad aures Arnulfi regis. Qui missos in Franciam mittens iussit, ut Odo et Karolus ad eum venirent, quatinus tantae calamitatis malum inter eos finiret.*

<sup>61</sup> Annales Vedastini 895, 75: *Odo vero rex strenuis secum assumptis viris ire ad regem perrexit Arnulfum multisque honoribus eum honoravit. Rex vero illum cum honore excoepit atque cum laetitia ad sua remisit, filiumque suum rex Arnulfus in praesentia Odoni regis nomine Zuendebolchum benedici in regem fecit eique concessit regnum quondam Hlotharii.*

<sup>62</sup> Annales Vedastini 895, 76.

threatening not so much Odo's rule as the territorial integrity of his realm. This was the exact situation that Louis the Stammerer had feared and tried to avoid with the treaty of Fouron, a situation that is strongly reminiscent of Louis the Younger's intervention in the Western realm, which lead to the loss of the western part of Lotharingia.

#### **IV.4 Possibilities and limits of royal power: Charles the Simple**

As in Odo's case, Charles the Simple's relations with other kingdoms initially aimed at legitimising his claim to the throne by gaining Arnulf 's acknowledgement. A number of Archbishop Fulk's letters reveal the problems Charles was facing in this context. In the first one, addressed to Arnulf himself, Fulk expresses his concerns about Arnulf being biased against Charles. 63 The failure of his endeavour becomes apparent from a second letter, this time addressed to Pope Formosus, asking for his support on the same matter,64 while a third one shows that Arnulf had not only refused to come to Charles' aid but even taken a hostile attitude towards his supporters and got his hands on property belonging to the church of Reims.65 Yet, while getting Arnulf 's acknowledgement was certainly important for Charles, this goal was not the principal factor in their early relations. In contrast to Odo, Charles had a strong claim to the throne based on his Carolingian descent, a claim that was immediately recognised, if not by Arnulf, then at least by the pope who congratulated him on his coronation.66 While Arnulf 's acknowledgment was very still valuable for Charles, Fulk's letters reveal a second aspect that seems to have carried more weight: the need for the actual material support that Arnulf could provide.

When Arnulf changed course in 894 and invited Charles to come to Worms, this was a great success for the young king, even if, like Odo before and again soon after, he had to acknowledge Arnulf 's superiority in rank. The *Annales Vedastini*'s description of this occasion emphasises this factor even more than the East Frankish account provided by the *Annales Fuldenses*. Seemingly, Arnulf granted Charles his father's *regnum.*67 This phrasing is most interesting since it reveals the complexity of the relations between the two rulers. Firstly, the mention of Charles


<sup>63</sup> Flodoard, HRE IV, c. 5, 380−383: *Arnulfo regi Transrenensi litteras mittens pro causa regis Karoli, quem parvulum adhuc unxerat in regem, reddit causas eius provectionis eo, quod audierat motum fuisse animum ipsius Arnulfi contra se pro hac perpetratione…*

<sup>67</sup> Annales Vedastini 894, 74: *Arnulfus vero rex benigne suum excepit consobrinum eique regnum paternum concessit adiutoresque ei delegavit hos qui erant ex superiori Francia.* Annales Fuldenses (Ratisbon continuation), 894, 125: *Wormacia habitum est generale conventum; ibi inter alia Karolus puer indole iuventutis, Hludowici Karoli de occidentali Francia regis filii filius, nepos regis, ad eum veniens, quem rex cum dilectione suscepit et ab solvit.*

receiving his father's realm demonstrates that his claim was independent and not reliant on receiving it from Arnulf. Secondly, the circumstance that Charles received the realm from Arnulf once again underlined the latter's superior position as well as his status as a source of justice and legitimation, the same source that Odo had used five years earlier. Thirdly, the act appears to have created a bond between Arnulf and Charles, requiring Arnulf to impose the judgement he had just spoken. And indeed Arnulf now provided Charles with "helpers" from Lotharingia to aid his cause. The price Charles had to pay for this help was even higher than the acknowledgement of Arnulf 's superior rank. As another letter from Fulk reveals, he also had to make a *promissio* about which nothing further is known.68

In the following year, when Arnulf called both Odo and Charles to come to him, Fulk's letter and the reference to this *promissio* was at first the only response from Charles' side. The reasoning behind this decision was probably that Charles had nothing to gain from the renewal of his public submission under Arnulf and that his independence would be questioned if he obeyed the summons. Only when it became known that Odo had in fact gone to Arnulf did Fulk react, quickly preparing for a journey to Worms, yet losing his baggage with the gifts for Arnulf to Odo, who was already returning to the West.69 Fulk's visit was probably meant as a compromise: Charles would again acknowledge Arnulf 's superiority, yet lose none of his own prestige by avoiding a meeting in person. In any case, Fulk set out too late, and the assembly of Worms was already over. The winner on this occasion was Odo, whose claim was publicly strengthened by the demonstration of the good relations between him and Arnulf.

Setting aside the fact that Odo's position had been strengthened, the setback for Charles appears to have been only a minor one. While not receiving any more help from Arnulf, he was able to secure an alliance with Zwentibold, who immediately intervened in the Western realm and joined his forces with Charles, forcing Odo to retreat behind the Seine.70 Just like Arnulf 's, Zwentibold's aid came at a price, in his case in the form of a promise for territorial gains. This again reveals the underlying problem of Charles' relations to other rulers during these years: while his claim was quickly acknowleged, he depended on external aid to actually impose it against Odo. This considerably weakened his position in negotiations, forcing him to make rather large concessions. Furthermore, while he depended on his partners, they could cease their support for him without the risk of losing anything. This was why Zwentibold could plot against Charles during the siege of Laon before retreating to Lotharingia: he had nothing to fear from Charles but could hope to gain even more by strengthening his ties to Charles' other supporters. There was, however, one exception to this rule. In late 895, after the alliance

<sup>68</sup> Flodoard, HRE IV, c. 3, 383: *Promissio nem quoque, quam rex suus Karolus eidem Arnulfo, qui regnum sibi contradiderat, promisisset manere inconvulsam…*

<sup>69</sup> Annales Vedastini 895, 75−76; Regino, Chronicon 895, 143.

<sup>70</sup> Annales Vedastini 895, 76.

with Zwentibold had been dissolved and negotiations between Charles and Odo had commenced,71 Charles met with Emperor Lambert and King Rudolf of Upper Burgundy at Remiremont.72 All three of them were under considerable pressure at that moment and in dire need of allies,73 so for the first time Charles' position was as strong as his partners'. However, nothing appears to have come from this meeting. This was probably not only due to the fact that none of the three rulers had any forces to spare, but also because Lambert and Rudolf had a different enemy than Charles: Arnulf. Setting aside this outcome, the meeting, nevertheless, points again to another characteristic of Charles' relations to other rulers during these years: that he faced no problems in being accepted as one of them.

#### **Lotharingia**

Charles' major problem during these years, his dependence on external help, was solved the moment he came to an accord with Odo and soon after became the sole ruler of the Western realm. His politics changed immediately: after a request from Reginar Longneck and his allies, he invaded Zwentibold's *regnum*. 74 Managing to liberate the besieged Durfos, he marched on Aachen, Nimwegen and Prüm, taking possession of the old Carolingian palaces,75 but when faced with Zwentibold's newly gathered forces, he had to conclude an armistice and return home. While the campaign demonstrates Charles' determination to make the most of the opportunities presented to him, it also reveals the limits of his power. The basis of his support in Lotharingia consisted only of Reginar Longneck and his allies, and he was unable to rally more of the local nobility to his cause. Moreover, his own forces were apparently insufficient to challenge Zwentibold's rule over the *regnum*.

However, the affair did not end when Charles left Lotharingia. Early in 899, Zwentibold met with legates sent by Charles and Arnulf at St Goar. The exact proceedings of the conference are unknown. Regino of Prüm only states that "later events brought more clearly into the light what was discussed at this meeting in private, when the king wasn't present."76 Nevertheless, some observations can be made. Charles' legation consisted not only of his archchancellor Anskeric, but also of Count Odacar, who had fallen out with Zwentibold alongside Reginar. 77 Odacar's presence points to one of the reasons behind the meeting: the ongoing

71 Annales Vedastini 895, 76 and 77.

<sup>72</sup> Liber Memorialis Remiremont, 21, fol. 11v: *Karolus rex iuuenis, Lanbertus imperator, Rodolfus rex, Rampo, Vuitbertus, Rotrudis, Adeldrudis, Siifridus, Gotdofridus, Manases, Eldigarius ep., Folco ep., Uuilerius, Lehutaldus.* For the dating and identifications, see Hlawitschka, Lotharingien, 147−152.

<sup>73</sup> See chapter I.2.6 and Hlawitschka, Lotharingien, 145−155.

<sup>74</sup> Annales Vedastini 898, 80−81; Regino, Chronicon 898, 146.

<sup>75</sup> Schneidmüller, Lothringenpolitik, 6−7.

<sup>76</sup> Regino, Chronicon, 899, 146−147: *Zuendibolch colloquium habuit cum optimatibus Arnulfi et Caroli et suis apud sanctum Goarem; ex regno Arnulfi interfuerunt Hattho archiepiscopus, Cuonradus et Gebehardus comites, ex parte Caroli Haschiricus episcopus et Odacar comes. Quid vero in eodem conventu seorsum sine presentia regis pertractatum sit, postea eventus rei luce clarius manifestavit.*

<sup>77</sup> See also chapter III.2.1.4.

conflict between the Lotharingian king and the group around Reginar. Charles' being invited to take part in the meeting allows us to conclude that his alliance with Reginar had not ended with his withdrawal from Lotharingia but continued on, and that he was considered as a power in his own right who needed to be involved in efforts to restore peace in the *regnum*. Charles' actual influence on the proceedings at St Goar is impossible to discern. It would seem that no final peace was made with Zwentibold, as both kings met later that year at Cambrai to make peace.78 The conflict between Zwentibold and Reginar and his allies appears to have remained unresolved as well, since the king renewed his attacks on them at Durfos and soon after tried to have them excommunicated.79 Finally, it also appears highly unlikely that Charles took part in the negotiations over Zwentibold's future as Regino hints at in his phrasing.80 If he had been involved in these secret proceedings, and therefore had known that Zwentibold's fate had already been decided upon, he would hardly have deemed it necessary to conclude a separate treaty with him some time after. Hence, Charles appears to have played only a secondary role at St Goar, a conclusion further emphasised by the location of the meeting on the Rhine, the border between Lotharingia and the East Frankish kingdom, indicating that Zwentibold and Arnulf were the primary participants. For Charles, however, this probably hardly mattered. For him, the situation had changed completely. Now it was he who could take advantage of the others' weakness. While his anwer to Reginar's and his allies' call had brought him no immediate gain, it had not harmed his position either.

Having dealt with Charles' first incursion into Lotharingia, we must now address one of the central questions concerning his reign, his ambitions to add this *regnum* to his rule. Charles had a long history of relations with this region. During the years of the fight against Odo, Lotharingia had served as a refuge on more than one occasion; in 898, as we just have seen, Charles responded to Reginar's appeal for aid; and in late 911 he finally acquired the *regnum*. Consequently, scholars have often argued that Charles had a strong ambition to add the lands of his Carolingian ancestors to his realm.81 Since these ambitions would have had a major impact on Charles' politics, it seems necessary to reevaluate the arguments on which these conclusions are based and to compare Charles' position in regard to Lotharingia with those of his predecessors.

"Count Reginar, however, went to King Charles and swearing his allegiance to him, he persuaded him and his *fideles* to invade Zwentibold's realm."82 According

79 Regino, Chronicon 899, 147.

<sup>78</sup> Annales Vedastini 899, 81.

<sup>80</sup> For the connection between Regino's phrasing and Zwentibold's downfall, see for example Hartmann, Lotharingien, 136−137.

<sup>81</sup> For example Parisse, Lotharingia, 313; Schneidmüller, Lothringenpolitik, 6−7; Bruand, Francie, 25; Koziol, Politics, 484.

<sup>82</sup> Annales Vedastini 898, 80: *Ragnerus vero comes venit ad regem Karolum et fidem ei promittens suasit illi atque suis fidelibus invadere regnum Zuendebolchi.*

to this report of the *Annales Vedastini* and the very similar one composed by Regino of Prüm,83 in 898 the initiative for Charles' invasion of Lotharingia lay not with the king, but with Reginar Longneck. Reginar, until that moment one of the most influential nobles at Zwentibold's court, had not only lost his position at the Lotharingian king's side, but also been relieved of his *honores* and banned from the realm.84 That he turned to Charles is hardly surprising for a number of reasons. Relations between Charles and Zwentibold had probably been tense ever since the siege of Laon, when the alliance between the two kings quickly deteriorated before it finally broke over rumours that Charles was to be murdered,85 even if Charles was still able to seek refuge within Zwentibold's realm. The tensity of their relation becomes apparent from one of Charles' early diplomas, issued for the abbey of Saint-Mihiel close to Verdun.86 In this diploma, at the request of his mother, Charles granted property in the Verdunois and the Charpeigne to the monks. Charles' grant was no more than a confirmation of an earlier diploma of Zwentibold issued about three years earlier at Trosly-Loire, west of Laon,87 at the time of Zwentibold's campaign into the West Frankish realm. This can hardly be a coincidence. Given that Zwentibold's diploma was issued either at a time when the two kings were allied or at least shortly after, it is most likely that Charles knew of the earlier grant. The monks of Saint-Mihiel had nothing to gain from a second grant of the same property made by a king who was not even their ruler. On the other hand, for Charles the issuance of the diploma made sense. By doing so, he could send Zwentibold a message that the events of Laon had not been forgotten and also that he was able to lure Zwentibold's supporters from his side, just as the Lotharingian king had done with Charles' allies Reginar and Baldwin. Saint-Mihiel's cooperation on this occasion can be easily explained: its abbot, Stephen, had strong ties not only to Charles himself, being related to him by marriage,88 but also to Archbishop Fulk, with whom he appears to have cooperated to become bishop.89 When Reginar sought aid against Zwentibold outside of Lotharingia, he could not only point to his earlier marriage to Charles' sister—a rather weak argument, given that she had already died and that he had abandoned Charles at Laon—he could also count on their common enmity towards Zwentibold.

While Charles' motives for invading Lotharingia can be partly sought in the fallout from events at Laon, his actions also reveal that he intended to make the most out of the opportunity presented to him by the invitation of Reginar and his allies. According to Regino, he marched to Aachen, then to Nijmegen before

84 Regino, Chronicon 898, 145. On Zwentibold's turn against Reginar see Beumann, Kurswechsel.

86 DChS 11 (13th February 898).

89 Flodoard, HRE IV, c. 7, 396.

<sup>83</sup> Regino, Chronicon 898, 146: *Rege ab obsidione recedente prefati comites Carolum adeunt et eum cum exercitu in regnum introducunt.*

<sup>85</sup> Annales Vedastini 895, 76.

<sup>87</sup> DZ 3 (14th August 895).

<sup>88</sup> In DChS 81, 181, Stephen is adressed as *nostre consanguinitati affinis dilectissimi* by Charles, indicating him being a relative by marriage.

finally turning to Prüm.90 By taking control of the old Carolingian palaces, especially the one at Aachen, which was so closely associated with Charlemagne and Louis the Pious, 91 he could legitimate his claim to rule over Lotharingia.92 His position in the *regnum*, however, appears to have been rather isolated and limited to his alliance with the group around Reginar. The rest of the Lotharingian nobility seems to have either remained loyal to Zwentibold who was able to muster sufficient force to confront Charles and bring about an armistice;93 or to have been oriented towards the East Frankish realm.94 This becomes apparent at the meeting of St Goar during the following year when Charles, as we have argued above, was not included in the negotiations between the Lotharingian nobles and Arnulf 's legation concerning the deposition of Zwentibold. Charles' retreat to the West and the subsequent treaty of Cambrai hence appear as the result of a realistic assessment of his overall position in the *regnum*, leading to the conclusion that there was nothing to gain for the moment. This assessment probably also influenced his decision not to intervene in Lotharingia when the local nobles turned away from Zwentibold. When Zwentibald's downfall had been decided upon at St Goar, it is likely that the future course of the *regnum* and its allegiance to the East Frankish ruler also had been determined at that moment.95

Regrettably, we lack corresponding narrative accounts for 911, the year when Charles finally acquired Lotharingia. Only the *Annales Alemannici* report that during this year, the leading Lotharingian nobles broke away from Louis' reign, which, since the annals note Louis the Child's death only in 912, seems to indicate that the Lotharingian nobles might have renounced Louis the Child as their king even before his death on 24th September 911.96 Whether this was the case or not, over a month appears to have passed until the question of succession was resolved. The *Annales Prumienses* note the 1st November as the start of Charles' rule over


Lotharingia,97 which coincides with the time frame between 10th October and 27th November deduced by Philippe Lauer based on an analysis of Charles' diplomas;98 while Conrad's coronation at Forchheim only took place between (probably) 7th and 10th November.99 However, Charles only arrived at Metz 1st January 912,100 so over three months after Louis' death and two after he had become king of Lotharingia. For example in 869, Charles the Bald had quickly reacted to Lothar II's death 6th August 869 at Piacenza and arrived at Metz already 5th of September.101 Charles the Simple, it would seem, was simply not prepared to move towards the east, indicating that he was rather surprised by the events and hence was not directly involved in them.

Lotharingia submitting to Charles' rule at that point indicates that some major changes had taken place since the earlier period. The first change was the rise of Reginar Longneck, who had become the most influential noble within Lotharingia over the course of Louis the Child's reign.102 His rise had been directly tied to the death of Count Gebhard, who had been described by one of Louis' diplomas as *dux* of Lotharingia,103 a rise that undoubtedly brought him in conflict with Gebhard's family, the Konradiner, the family dominating the circle around the East Frankish king.104 In this conflict, Reginar was joined by the Matfrid family, whose attempt to extend their influence in Lotharingia had been thwarted by Louis and his allies.105 Two of the three major factions within the *regnum* would hence not support the succession of Conrad, 106 and the third, the church of Trier, remained hesitant. Furthermore, a number of changes in important positions had taken place. At Liège, one of Charles' relatives, Stephen, 107 had become bishop. The latter, formerly abbot of Saint-Mihiel, had also been close to the Matfrid family,108 a connection he now undoubtedly brought into Charles' new network.109 In ad-


dition, Bishop Dodilo of Cambrai who had always followed a more independent policy110 and supported Zwentibold against Charles, 111 had been succeeded by another Stephen. This new bishop was seemingly more cooperative, being among the first to receive a diploma from Charles. 112 Stephen of Cambrai also represents yet another group of nobles who had earlier on belonged to Zwentibold's network but then lost their influence under Louis the Child's reign—nobles that now joined Charles' side.113

Hence, within Lotharingia there were a number of factions that would opt for Charles for different reasons: rivalry with the Konradiner, personal relations with Charles, dissatisfaction with their own influence at court. These different factions prepared the field for Charles and provided him with the support necessary not only to take over the rule, but also to defend it against the Konradiner attempts to regain control over the *regnum* or at least over some of its parts. Two of these attempts can be traced by the *Annales Alamannici*: a first encounter between the two kings appears to have taken place close to Strasbourg, where Conrad issued a diploma in March 912,114 when both kings seem to have concluded a treaty.115 However, since the annals refer to a breach of faith around the same time, this treaty appears not to have lasted long and was followed by an attack of Conrad on Aachen.116 Meanwhile, at Strasbourg a strong pro-Carolingian party appears to have existed—maybe the reason why Conrad had moved there: to maintain control over the city. With Conrad's absence from Strasbourg, Charles' allies in the south appear to have become active, attacking and devastating the city,117 albeit without being able to hold it: Conrad returned there in March 913.118 Nevertheless, within Strasbourg, the pro-Carolingian party seems to have won the upper hand. After the inhabitants had killed Bishop Otbert, they elected Charles' nephew Gauzfrid.119 Gauzfrid died soon after and was replaced with Ricuin, a Lotharingian noble who in 916 at the synod of Hohenaltheim, while himself being absent, was

consider 913 to be the latest possible year during which the diploma was delivered is not clear. 915 corresponds with the year of Reginar Longneck's death at whose request the diploma was issued). 110 See Archbishop Fulk's remarks against him. Flodoard, HRE IV, c. 6, 390 and 392.

111 DZ 23. See also Schieffer, Kanzlei, 32 and Bauer, Lotharingien, 106. Bauer refuses Schieffer's assumption that the diploma reflects Dodilo's aid for Zwentibold against Charles, but agrees on it being aimed against the king of the Western realm.

112 Vercauteren, Note, 102. Vercauteren also brings forward the hypothesis that it was Charles who installed Stephen on the siege of Cambrai. While this may indeed be possible, considering that Cambrai was part of the archdiocese of Reims, it cannot be further proven.


accused by Conrad's supporters of being elected uncanonically,120 which seems to indicate that he as well prefered to side with Charles. 121 The loss of Strasbourg also marked the end of Conrad's efforts to regain control over at least parts of Lotharingia. It is clear that local support played a central role in Charles' success in taking over and defending Lotharingia.

However, the question remains as to whether Charles pursued an active policy to influence conditions in his favour. Given the lack of narrative accounts, we have to rely on the reconstruction of the chronology to draw any conclusions. As we have seen, a rather long period of time passed between Louis' death and Charles' arrival in his new *regnum*, indicating that he was not prepared to act quickly. This is not to say that Charles was not using his contacts within the Lotharingian nobility to influence their decision in his favour, yet, nevertheless, the initiative appears to have lain with the nobles, not with him.122

The events of 898 and of 911 follow the same pattern: Lotharingian nobles approached Charles and offered him the rule over the *regnum*, a call to which Charles reacted positively, ready to defend his claim. The circumstance that the nobles took the initiative does not mean that Charles did not carry any ambitions towards Lotharingia, however. Quite the contrary, the nobles turning to him was undoubtedly due to the fact that they expected a positive response from his side. However, the question is how grand his ambitions actually were. In this context, a number of arguments have been brought forward. Based on Charles' itinerary and the number of Lotharingian nobles in his diplomas after the acquisition of the *regnum*, it has been assumed that Charles felt a deep attachment to the lands of his ancestors.123 This seems problematic for two reasons. First, how, based on the sources we have, can we distinguish between sojourns and contacts motivated by personal attachment and those necessitated by political circumstances? Second, as we have argued in the previous chapter, setting aside the times when those stays were definitely due to political necessities, contacts with Lotharingia appear to been on a reasonable level, while those with nobles from the Western realm remained stable.124

Another argument is based on the genealogy dictated by Charles to the canons of Saint-Corneille of Compiègne.125 According to Geoffrey Koziol "its details are unmistakably oriented to Charles' Lotharingian aspirations. Thus, while not-


ing all Louis the German's sons, it only traces the descent of Carloman, through whom Lotharingia passed via Arnulf to Zwentibold and Louis the Child. Further attesting to Charles' insistence on legitimacy, it also notes Zwentibold's illegitimacy."126 While the genealogy indeed only follows Carloman's lineage and ignores the offspring of Louis the German's other sons, this has nothing to do with any Lotharingian aspirations on Charles' side. Indeed, there is a far simpler explanation for why the genealogy is constructed the way it is: it is focused on the West Frankish lineage, which is described in large detail, naming all the offspring of the successive rulers including the names and titles of their mothers. All the other lineages only name those sons who were crowned rulers, ignoring daughters and illegitimate children.127 Also, while the genealogy does mention Zwentibold's birth by a concubine—indeed revealing Charles' sense of legitimacy—this is hardly a sign that he claimed Lotharingia. By naming Zwentibold, Charles added him to those he acknowledged not only as belonging to the family, but also as a legitimate ruler.

In the context of Charles' Lotharingian ambitions, Koziol furthermore points to the names of Charles own illegitimate offspring, his sons Arnulf and Drogo, both carrying names "distinctive to the Lotharingian side of his family."128 Both names certainly have a Lotharingian touch, as their first bearers had been Arnulf, the Carolingian progenitor, and Drogo, an illegitimate son of Charlemagne, both of whom had been bishops of Metz. However, in both cases there are alternative readings. According to Karl Ferdinand Werner, of seventeen illegitimate sons of Carolingian rulers, four were called Arnulf, three Hugh and two Drogo, while all other names only appear once.129 While there certainly were other potential choices, as is shown by the name of Charles' third son, Rorico, Arnulf in particular appears to have been a far too common choice of name to use as an argument for Lotharingian aspirations on Charles' side. In the case of Drogo—his name being rather uncommon—it is worth pointing out another possible reason. The first bearer of this name was bishop of Metz and one of Louis the Pious most important advisors130 and it might have been that Charles hoped for his illegitimate son to occupy a similar role under his own, yet unborn, heir,131 although this has to remain mere speculation.

While the sources thus do not provide any further insight into Charles' actual ambitions, it is worth comparing his endeavours to acquire Lotharingia with those of his brothers. As we have noted above, the western part of the *regnum*,

126 Koziol, Politics, 480 with n. 84.


<sup>127</sup> See chapter III.1.3.

<sup>131</sup> Becher, Arnulf, 667, concerning Charles the Bald's choice to name one of his twins by Richilde Drogo.

which Charles the Bald had acquired for himself in the treaty of Meersen,132 had fallen to Louis the Younger in the wake of his intervention in the Western realm. On two occasions, these sons made efforts to regain control of this area which had apparently been intended to revert to them upon Louis the Younger's death.133 Louis, on the one hand, refused the Lotharingian nobles who were trying to commend themselves to him, but, on the other hand, sent an army to the *regnum* under Count Theoderic of Vermandois, one of his most important advisors, supposedly to protect Lotharingia from the Northmen. This, however, was probably not the only reason. By showing his face in Lotharingia and demonstrating his willingness to act in a royal manner by fighting the Vikings, he ensured that his claim on his father's part of Lotharingia remained visible. While he did not want to jeopardise the cooperation established between the various Carolingian rulers by acting without the consent of the new East Frankish ruler, Charles the Fat, he nevertheless betrayed his ambition to regain the lost territory. After Louis' death, it fell to his brother Carloman to secure his father's heritage. Under his rule, Hugh the Abbot sought out Charles the Fat to remind him of the treaty his father had concluded and to have the western part of Lotharingia returned to him. Like Charles the Simple, his brothers Louis and Carloman had strong ambitions to re-acquire parts of Lotharingia. In the light of these earlier developments, it even seems possible that the new addition to the dating clauses of Charles' diplomas, the *largiore vero hereditate indepta* ("attaining [his] larger inheritance"), did not so much refer to a general claim based on him being the last remaining Carolingian king, but rather to a very specific one, based on the fact that at least the western part of Lotharingia had belonged to his father and should never have remained part of the East Frankish realm. This is of course highly speculative, yet given Charles' highly developed sense for history and legitimacy134 it may not be too far fetched to assume that these thoughts may have played a role in the creation of the dating formula. Be that as it may, Charles appears to have carried the same ambitions as his brothers, the main difference being the means he was able and willing to use.

#### **The treaty of Bonn**

As with Lotharingia, the context of Charles' ambitions also plays a central role in the assessment of his last treaty135 with another ruler, Henry the Fowler, the treaty of Bonn concluded in November 921. The text of the treaty itself, preserved in the West Frankish version,136 is rather short, consisting of only four chapters. The first

132 MGH Capit. II, N° 250, 191−192.


<sup>133</sup> See chapter IV.2.

of them gives the time and place of the meeting, the second and third the reciprocal promises of an *amicitia* between the two kings, and the fourth names the witnesses. Very much like the meeting at Cambrai, the meeting at Bonn—actually on the Rhine itself—was at the border between the *regna* of the rulers, demonstrating the equality of Charles and Henry. This equality is also portrayed in the text itself, addressing both kings with very similar titles: Charles is the *domnus et gloriosissimus rex Francorum occidentalium*, Henry the *domnus et magnificentissimus rex Francorum orientalium*, which emphasises that both rulers were considered as Frankish kings of the same rank.137 This negation of the difference between the Carolingian Charles and the non-Carolingian Henry has often been interpreted by scholars as a "defeat" for Charles who, by this treaty, would have had to give up any plans of restoring the old Frankish empire, as well as the claim of being superior to his East Frankish neighbour.138 This reading depends on two assumptions: that Charles indeed attempted to extend his realm in the east and that he did not acknowledge Henry as his equal.

In this context, Charles' new title after the acquisition of Lotharingia has often been read as a sign of his imperial ambitions, with the *rex Francorum* expressing not only his claim on Lotharingia, but also on the Eastern realm.139 Herwig Wolfram, on the other hand, pointed out that *rex Francorum* might indeed have had a different meaning. According to him, *rex Francorum* drew its meaning much less from Charlemagne than from the union of the old Carolingian core lands between Seine and Rhine, *Francia.* Read in this context, the new title would hence not express a claim on the Eastern realm but a "voluntary self-restraint."140 Given the ambiguity of the new title, it is best to analyse whether Charles made any efforts to extend his rule over the East Frankish realm. Two episodes are most revealing about this issue. After the acquisition of Lotharingia, Charles successfully managed to install his own candidate in the see of Strasbourg, thus securing the city against Conrad I, 141 a policy that has been read as a sign that he attempted to add


<sup>137</sup> See also Ehlers, Anfänge, 30−31.

<sup>141</sup> See above.

Alsace, belonging to the East Frankish realm, to his rule.142 However, Alsace was one of the "Bruchlinien" mentioned by Jens Schneider,143 a region on the border between Lotharingia and the East whose status was ambiguous.144 Charles' efforts to gain control over this region were hence based on a maximalist interpretation of the borders of Lotharingia and were not an endeavour to extend his rule over the East Frankish realm.

The second episode directly preceeds the treaty of Bonn and sees Charles attacking in the area of Worms. According to the continuator of Regino's chronicle, Charles, by this attack, wanted to usurp Alsace, the parts of Francia along the Rhine, and Mainz.145 The reliability of this account seems problematic. He misdates the event to 923 and fails to put the attack in its proper context by claiming to have had insight into the king's motivation for the attack. In fact, Charles' campaign was only the response to a preceeding intervention by Henry in Lotharingian affairs. When Charles advanced against Reginar Longneck's son Gislebert in late 919, according to Richer of Saint-Remi, the latter fled to Henry's court.146 It seems safe to assume that early in 920, when Charles had been abandoned by the West Frankish counts, Henry supported Gislebert's return to Lotharingia. During the same time he was furthermore involved in the succession of Liège, where, as Charles' letter to the bishops of the realm claims, he pressured Archbishop Hermann of Cologne into ordaining Gislebert's ally Hilduin as bishop.147 Hence, Charles' advance against Henry was hardly due to an ambition to extend his rule over the East Frankish realm but a reaction to Henry's attempts to increase his influence within Lotharingia during a moment of Charles being vulnerable. This is of course not to say that Charles might not have harboured any hopes to do so and, given how he acquired Lotharingia.148, it seems indeed highly likely that if such an opportunity had presented itself to him he would have seized it. Yet, despite any claims indicated by his new title, his actual policies do not betray any claims over more than Lotharingia.

The very circumstance that Charles made no attempt to extent his rule to the East does not mean that he did not claim a superior status to Henry based on his Carolingian heritage. As we have argued in the previous chapter,149 Charles possessed a strongly developed sense of legitimacy when it came to his own family, yet this same sense did not carry over into a special emphasis of his heritage


within royal diplomas in comparison to those of his predecessors. To determine how Charles perceived the non-Carolingian Henry, it is best to evaluate the ways Charles dealt with other non-Carolingian rulers before the treaty of Bonn. As we have seen, Charles had already allied himself with two of these rulers during the years of his fight against Odo, Emperor Lambert of Spoleto and Rudolf of Upper Burgundy. The entry made at Remiremont records *Karolus rex iuuenis*, *Lanbertus imperator* and *Rodolfus rex*, making, apart from the emphasis on Charles' youth, no distinction between the three rulers.150 Furthermore, Charles' contact to Emperor Berengar needs to be mentioned. A letter from Pope John X to Archbishop Hermann of Cologne concerning the succession at Liège indicates that Berengar intervened for Charles' candidate Richer in front of Pope John X, 151 most probably at Charles' request even if the letter itself fails to mention this circumstance.152 Berengar was a logical choice if Charles wanted someone to represent his interests at the papal court. John and Berengar entertained good relations with each other that had already resulted in an alliance between the two probably in late 914 and Berengar's coronation as emperor by the pope a year later.153 By the time of the dispute at Liège, both cooperated against a certain Hubert who appears to have been in conflict with the church of Rome.154 Berengar as a son of Louis the Pious' daughter Gisela155 had Carolingian blood himself, yet in how far this influenced the relation between him and Charles stands open to debate—he does not appear in the genealogy of the Carolingians dictated by Charles to the canons of Compiègne.156 In any case, Berenger intervening in favour of Charles' candidate indicates that the two of them were probably also on good terms with each other. Nothing further is known about this contact, yet it seems reasonable to assume that Berengar's lack of a Carolingian father again did not considerably determine their relation. Finally, Charles' handling of his own non-Carolingian predecessor Odo needs to be taken in consideration. As mentioned, there are only four mentions of Odo in Charles' diplomas, in addition to a number of dating clauses early in his reign reading *in successione Odonis* ("in succession to Odo").157 Three of these mentions are confirmations of diplomas issued by King Odo (*Odo rex*)158 while the fourth is to include Odo in a memorial service ordered by Charles. 159 While some of these mentions were most likely due to the influence of Odo's brother Robert, at least on one occasion Charles confirmed one of his predecessor's diplomas without Robert


being present.160 The Remiremont entry, Charles' dealing with Berengar as well as his diplomas hence show no signs of Charles claiming any kind of superiority over non-Carolingian rulers. Charles dealing with Henry the Fowler as a peer seems to be in line with his general policy.

With the reading of the treaty of Bonn as a defeat for Charles dismissed, a new assessment based on the political circumstances can be attempted. For Charles, the treaty came at the end of a development which had started with the confrontation with the Western nobles at Soissons in early 920. After they had finally come to a settlement, he moved to Lotharingia, coming to an accord with Count Gislebert, who had taken the opportunity to have himself elected *princeps* in Charles' absence. Charles then went on to attack Worms which was, according to Regino's continuator, a futile attempt. The following year saw Charles active in Lotharingia again, neutralising Gislebert's ally Count Ricuin of Verdun before concluding a truce with Henry. 161 In the context of the unresolved conflict with Gislebert and his allies—shortly after the treaty of Bonn Charles again attacked the count162 the *amicitia* with Henry gains new importance. First, Henry acknowledged that Lotharingia was under Charles' rule,163 thus securing the *regnum* from further incursions of the East Frankish king. Second, Charles undercut Gislebert's alliance with Henry by making his own alliance with the East Frankish king.164 The treaty of Bonn, despite lacking any precise agreements about further cooperation, considerably increased Charles' room for manoeuvre, allowing him to pursue an aggressive policy against Gislebert. Similarly, the treaty also allowed Henry to concentrate on his unresolved internal problems with Burchard of Swabia and especially Arnulf of Bavaria, who had not yet acknowledged Henry as king.165 However, the treaty should not be reduced to a mere tactical measure for both sides. Charles and Henry did indeed intend to cooperate on a wider scale. In 922, a synod assembled at Koblenz, organised by the two kings.166 With Coblenz being located within Lotharingia, it seems that originally an assembly far larger than the one that actually took place had been intended, for only Archbishop Hermann of Cologne was present from Charles' side. By the time of the synod, the rebellion of the West

160 DChS 60 for Notre-Dame de La Grasse.


Frankish nobles had probably already commenced, so that Charles and his allies' focus was drawn away from the meeting.167

However, this rebellion also demonstrates the limits of the new cooperation between the two kings. Some time after Robert's coronation, he and Robert met at the Ruhr, in turn concluding an *amicitia*. 168 In the wake of this *amicitia*, it is probable that an entry in the *Liber Memorialis* of Remiremont was created, featuring not only Robert and Henry, but also King Rudolf II of Upper Burgundy, 169 which demonstrates how quickly Robert's kingship was acknowledged by other rulers and reveals how well he was connected even beyond the borders of the West Frankish realm. Karl Schmid, based on a number of other entries in the *Liber Memorialis*, has managed to cast some light on the background of this *amicitia*. In the wake of the death of a certain Count Guntram 25th December 921, a considerable number of nobles from the West and East Frankish realms as well as Lotharingia appear to have met at Remiremont to memorate the deceased. Among these nobles, Schmid identified not only Robert's son Hugh the Great, but also the sons of Richard the Justiciar as well as members of the Konradiner and other important East Frankish noble families.170 If Schmid's identifications are correct, Robert possessed influential allies at Henry's court, most importantly members of the Konradiner family.171 The Konradiner in turn, as we have seen, had lost their leading position in Lotharingia when it had fallen to Charles172 and it seems likely that they now mediated between Henry and Robert to loosen the West Frankish control over the *regnum*. However, the strength of this network should not be overemphasised either. Henry did not come to Charles aid, although, according to Widukind, the latter had asked him for his support, sending him a relic of Saint Dionysos after the lost battle of Soissons,173 probably a reference to the treaty of Bonn where they had sworn their oaths on relics.174 Yet, neither did Henry intervene on Robert's behalf. Thus, Henry's treaty with Robert is strongly reminiscent of Arnulf acknowledging both Odo and Charles as rulers and in fact the *amicitia* between the two may


not have meant more than Henry recognising Robert as king. While the treaty of Bonn had not brought a new age of cooperation for Charles, it had at least secured Henry's neutrality.

Thus, two different phases of Charles' relations with other rulers can be distinguished: before and after he became sole ruler of the Western realm. Before 898, as with Odo, the acknowledgment of his new rank by other rulers played a certain role, yet in contrast to Odo, this role was of rather minor importance. Charles' own claim was strong enough that further legitimation was a bonus, not a necessity. Instead, his relations were marked by his need for actual material support, weakening his position in negotiations and forcing him to make significant concessions to his potential partners. However, once became sole ruler this image changes radically and Charles became the first West Frankish ruler since Charles the Bald who was able to exploit opportunities presented to him by conflicts between neighbouring rulers and their nobility. In this context, comparison with his brother Louis III, who had refrained from doing so, allows insight into the way the possibilities and limits of royal power in regard to relations to other rulers differed between the various rulers. Where his brother had had to take into consideration the impact of his actions on the cooperation established between him and the other Carolingian kings, Charles was at liberty to pursue his own interests. However, at the same time Charles also missed out on the stabilising effects such cooperation could have had on his position. When the tide began to turn and his own rule was weakened by a number of conflicts with the nobles, it was Henry the Fowler who took the opportunity presented to him by intervening in Charles' realm. Charles, as it seems, was very well aware of this problem and its potential effect on his own rule. His reaction was the treaty of Bonn, a treaty that aimed at preventing Henry from intervening in West Frankish affairs, but also at establishing a new cooperation from which Charles, if need should arise, could draw further support. Therefore, Bonn seems like an attempt to revive the period following the treaty of Fouron, increasing Charles' room for manoeuvre in relations with his nobles while limiting his possibilities to exploit opportunities to extend his rule—a price he seems to have paid willingly, given that there is no indication he had ambitions to do so even before the treaty.

#### **IV.5 Conclusion**

A number of factors determined the possibilities and limits of royal power when dealing with other rulers. First, the strength of a ruler's claim to the throne had a large impact on their relations with others. This is especially apparent in Odo's case, as his claim's weakness made it necessary for him to seek Arnulf 's acknowledgement as another source of legitimation to strengthen his position within the West Frankish realm. In this context, the support of the nobles for a ruler also played a crucial role. Charles the Simple, in need of allies, had to acknowledge Arnulf 's superiority in rank and had to make further concessions, just as Odo did. If a lack of support turned into open conflict between the king and the nobles, situations were created that gave other rulers the opportunity to intervene in these conflicts. This could of course work both ways: while the succession question in the West Frankish realm after Louis the Stammerer's death, as well as the rebellion against Odo and the conflict between Charles the Simple and Gislebert led to interventions from Louis the Younger, Arnulf, Zwentibold and Henry the Fowler, Charles the Simple profited from conflicts within Lotharingia and the East Frankish realm to extend his rule over the neighbouring *regnum*. However, the success of such interventions depended not only on the weakness of the opposition, but most of all on the support the intervening ruler could muster himself. Charles was only able to acquire Lotharingia when a large number of nobles from the *regnum* turned to him. However, relations between the various rulers were not only marked by rivalry and attempts to extend their influence. On the contrary, the treaty of Fouron between Louis the Stammerer and Louis the Younger shows strong signs that it was aimed at a large scale cooperation, including the attempt to prevent nobles from seeking support in neighbouring kingdoms against their rulers. This very same cooperation was then realised under Louis III and Carloman II against the threat posed by Boso of Vienne. However, while strengthening their position against Boso considerably, this alliance also had its downside, as it prevented Louis and Carloman from pursuing their interests in the western part of Lotharingia against Charles the Fat. Unhindered by such restrictions, Charles the Simple was able to acquire the *regnum*, yet he also lacked the support he could have drawn on from a similar alliance during his last years—a situation he tried to ameliorate with the treaty of Bonn. In the end, the necessity for such alliances depended on the support the king received from his nobles. If these relations were strong, such cooperation was unnecessary. If they were weak, they could serve the king in the ensuing conflicts.

## **V. The Viking problem**

When Charles the Simple became king, Viking warbands had already been raiding Frankish shores for over a century. In the 790s, small groups had started ravaging trading centres and monasteries.1 The overall size of these groups started increasing from the 840s onwards, soon ranging up to armies of several thousand men.2 At the same time, the number of active groups increased, with the Vikings from then on attacking several places simultaneously and penetrating the continent deeper and deeper via the rivers while ranging on horseback into the countryside.3 Instead of returning home after each campaign, they now remained in the country, erecting secure camps for their ships and themselves to pass the winter months, which indicates their high level of organisation.4 In the late 860s in England, one of these groups grew even more in strength, forming what is now known as the "Great Army." When the resistance organised by King Alfred became stronger and the army suffered a defeat at Edington, this army turned towards the continent, arriving on the Scheldt in 879.5

However, one hundred years of ravaging and plundering also meant that there had been one hundred years of contact between the Northmen and the Franks. From the early 9th century onwards, Viking leaders were present at Frankish royal courts and pledged their allegiance to Frankish rulers.6 Almost all of them became Christian, baptism in combination with godparenthood not only serving to establish peace, but also to create a personal bond between the parties involved, putting the baptised in an inferior position to his godfather.7 In most cases, these ties proved to be effective, with the respective Northmen becoming loyal supporters of the kings whose courts they had joined.8 However, these cases only represent a small part of the contacts that existed between the Vikings and the Franks: both sides treated with each other and exchanged emissaries, treaties were concluded

1 The standard work on the Viking incursions is still Musset, Invasions. Since then (but also before) an abundance of studies have been published. Most important for our own work are Sawyer, Welt; D'Haenens, Invasions; Zettel, Bild; Searle, Kinship; Boyer, Vikings; Lund, Allies; Bauduin, Normandie; Baudin, Monde and Mohr, Wissen. Collections of articles include Bauduin, Fondations; Flambard Héricher, Progression and Pierre, Naissance. A good overview is offered by Coupland, Vikings.

2 For the number of ships reported by the sources, see Coupland, Vikings, 194. On the Viking ships and the number of men they carried, see D'Haenens, Invasions, 69−72 and Bill, Navires, 41. On the organisation of the Viking bands, see Mohr, Wissen, 198−204.

3 Coupland, Vikings, 195−196. A good example is the Loire, where Viking attacks are reported at Nantes in 843, at Tours in 853, Orléans in 856 and at Fleury in 865.


<sup>4</sup> Musset, Invasions, 209.

and efforts were made to convert the Northmen.9 In fact, as Pierre Bauduin argues, the Northmen were increasingly integrated into Frankish society, with their leaders entering into competition with the elites of the realm and these latter in turn becoming intermediaries between the Vikings and the king.10

The sheer extent of the Viking incursions meant that, over the course of the 9th century, dealing with them became one of the most important challenges faced by Frankish kings. Kings were presented with a number of options for action. Campaigning against the Northmen was one of them,11 yet insufficient considering that the Vikings were often simultaneously active in different theatres, most importantly in the Western realm along the Loire and the Seine.12 This soon necessitated the delegation of defence to individual nobles who were put in charge of newly created marches protecting the rivers and the adjacent regions.13 One of them was along the Loire where Charles the Bald installed Robert the Strong as leader, providing him with the resources and authority to defend the valley independently from the king.14 However, in both cases direct military actions seeking to encounter and defeat the enemy proved rather fruitless. The Northmen, aiming for profit and not for battle,15 preferred to avoid engagements by travelling through inaccessible areas and retreating when facing a hostile army. When it actually came to battle, victories were certainly won, yet proved insufficient due to the Northmen simply scattering and later regrouping.16 An alternative strategy was to prevent the Vikings from entering vulnerable areas, for example by occupying the river banks when a Viking fleet arrived or by blocking the river with fortifications.17 The last option the rulers had was not to fight the Northmen, but to use existing contacts to negotiate a tribute—a strategy that in fact turned out to be the most effective way to remove the Vikings from the realm.18 While this solution was unpopular among the churches who had to provide a large part of the sums paid to the Northmen, it, nevertheless, appears to have been a valid political option actually sought by the


<sup>9</sup> Bauduin, Monde franc, esp. 225−341.

rulers.19 Fighting and negotiating were thus closely connected to each other. If the former could not drive the Northmen from the realm, at least an advantage could be won over them that could be used when treating with them.20

These were also the strategies available to the West Frankish rulers succeeding Charles the Bald who were faced with the Great Army that had formed in England over the course of the past decades and now turned to the continent. Louis the Stammerer and his sons, Charles the Fat, Odo and Charles the Simple all had to decide in which way they wanted to react to this incursion. In this regard, Fernand Vercauteren described a shift of responsibility for defence, or at least for the actual measures taken, from the various rulers towards the local nobility,21 a view that has been challenged by Simon MacLean for Charles the Fat. 22 Measures taken by the king remained important and in the following two sections we will analyse these measures, their effectiveness and the role of the nobles in their implementation.

#### **V.1 Viking incursions and royal measures in protection of the realm**

#### **V.1.1 Going on the offensive: Louis the Stammerer—Charles the Fat**

#### **Louis the Stammerer**

During Louis the Stammerer's reign, still preceding the arrival of the Great Army on the continent, Viking activities are only reported along the Loire where they had been installed since around 862.23 Despite the defence of the region falling under the purview of the local *marchio*, Hugh the Abbot,24 in early 878 Louis crossed the Seine for a campaign against the Northmen. Hincmar's report about the circumstances around the campaign casts some light on this decision. According to him, Louis' intervention was not the result of royal initiative, but of a request for aid from Hugh the Abbot himself.25 Yet, the call for aid against the Northmen appears to have been only one of the reasons for Hugh's appeal, in addition to


<sup>19</sup> Keller, Négociation. Both scholars rightly emphasise that tribute payments were not necessarily the result of Frankish weakness leaving the rulers negotiating them no other choice, but a valid political option.

<sup>20</sup> See Abel, Danegeld, 184−186 for Alfred the Great's dealings with Guthrun, aguing that "the key to Alfred's successful peacemaking was a demonstration of military power sufficient to cow Guthrun into at least temporary submission."

the conflicts that had broken out with the sons of Gauzfrid and Emeno. The multitude of problems occurring at the same time seems to have overstrained Hugh's resources, necessitating Louis' intervention in the region. Indeed, while Hincmar mentions nothing about any measures taken against the Northmen, he does report that Louis solved the conflict with Gauzfrid, thus casting some light on the king's priorities. Soon after, Louis received word that Pope John VIII had crossed the Alps. Wishing to meet with him, he left the Loire again.26 While his departure was certainly due to John's arrival, it, nevertheless, appears that Louis considered the region secure enough to leave the defence to the local forces under Hugh. This impression is further strengthened when taking into consideration the measures the king initiated after the meeting with the pope at Troyes. Instead of returning to the Loire, he first met with Louis the Younger at Fouron before starting to prepare a campaign against Bernard of Gothia. 27 For Louis, the Viking threat was of low priority, his actions revealing only that he was able to intervene if local defences were insufficient to deal with the current situation.

This image also persists during the period immediately following Louis' death. The *Annales Bertiniani* report one campaign in the same region towards the end of 879, after the meeting between Louis and Carloman with Charles the Fat at Orbe: "When they returned home from their journey, it was reported to them that the Northmen who were in the Loire were devastating the land. And setting off immediately, they met them on the day of Saint Andrew and many of them were killed and even more drowned in the river Vienne and according to God's will the Frankish army returned home victorious and safe."28 There are a number of points to be made about this campaign. First, considering the young age of Louis and Carloman, it seems hardly likely that it was they who mobilised the army and led it into battle. As we have argued, just after Louis the Stammerer's death the brothers were in the hands of the group around Hugh the Abbot, the same man who was in charge of the march along the Loire.29 The reaction to the Viking raids and the mustering of forces was hence most probably his and his allies' doing. Furthermore, it is interesting to note the timing of the campaign. Gauzlin and Count Conrad had already called Louis the Younger into the realm and only the sickness of the latter's brother Carloman had prevented him from coming to the Western realm where, to complicate matters even more, Boso's rebellion had al-

*markionis, Ebrocensem ciuitatem usurpans, multas depraedationes circumcirca in illis regionibus exercebat, insuper et Eiricum more Nortmannico depraedari praesumpsit.*

26 Annales Bertiniani 878, 222−223; Annales Vedastini 878, 43.


ready started. Given this background, it may well have been that the campaign was not only meant to deal with the threat posed by the Northmen—a task that Hugh had to fulfil in any case—but also to strengthen the position of Hugh's group by demonstrating their ability to protect the realm. The Northmen at the Loire were probably the best target for such a purpose: a well-known threat, yet not one that was overly strong, given the low priority that Louis the Stammerer had attributed to dealing with them. In any case, this early campaign under Louis and Carloman, like the one of their father, was probably rather limited, without any larger mobilisation of forces, a mere prologue to the challenges soon to be posed by the Great Army.

#### **Louis III**

When the first forces of the Great Army arrived on the continent in 879, they entered the Scheldt and installed themselves at Ghent to pass the winter, using this base to plunder the surrounding area,30 a group of them going even as far as besieging Reims, as one of Hincmar's letters indicates.31 This explains the readiness of the new West Frankish kings and the nobles around them to cooperate with Louis the Younger on this matter. The first sign of this cooperation followed directly on from the treaty of Ribemont in February 880, when Louis and his army, on the way back from the meeting, engaged Viking forces at Thiméon. While the reports of Hincmar, the *Annales Vedastini* and Regino of Prüm seem to indicate that this encounter was the result of chance rather than planning,32 the *Annales Fuldenses* suggest the opposite. According to them, after Ribemont Louis the Younger turned his army to drive out the Northmen from the Scheldt.33 This reading would explain why Hugh the Abbot was with the army and indicate that the Viking incursion at the Scheldt was perceived as a common threat that needed to be dealt with by joint forces. This view is further encouraged by the next campaign during the summer of the same year. In the wake of the large-scale


cooperation against Lothar II's son Hugh in Lotharingia and Boso in Burgundy, the various rulers had also agreed to join forces against the Northmen installed at Ghent. This time, the major part of the forces appears to have come from the Western realm: Gauzlin and many others were ordered to advance against the Northmen. Arrived at the Scheldt, Gauzlin tried to coordinate his attack with the local forces. However, the manoeuver appears to have failed and the army was routed. Gauzlin now appears to have come to the conclusion that his remaining forces were insufficient to take any further measures and disbanded them, leaving the north-east of the realm open to the Northmen.34 While the circumstance that both Western kings went south seems to indicate that fighting Boso was given priority over the Viking threat, the fact that one of the leading nobles of the realm led the army against the Northmen nevertheless indicates how seriously this campaign was taken.

Gauzlin's defeat led to a change in these priorities. Louis III now returned to Francia, leaving Boso to his brother Carloman. 35 It is interesting to note the change of tone in the sources from this moment on. The *Annales Vedastini* now emphasise the king's role in gathering a new army and leading these forces against the Northmen.36 While this does not necessarily indicate a change in his relations to the nobles surrounding him—as Thilo Offergeld has shown, from the moment of their coronation, Louis and Carloman appear as independent, mature rulers in the narrative sources despite the continuing dominance of the nobles37—there, nevertheless, seems to have been a significant change following Gauzlin's defeat. In their extremely different reports of the battle of Saucourt, both Hincmar and the *Annales Vedastini* ascribe a crucial role to Louis' leadership. According to Hincmar, when the battle was developing in the Franks' favour, it was "Louis himself [who fled] with his men […], though no one was pursuing them," thus revealing the divine judgement.38 The *Annales Vedastini*, on the other hand, report that it was the king himself, who turned the flight of his men into a victory by bringing them to stand and fight.39 While this certainly did not mean that Louis now

34 Annales Vedastini 880, 47 and 48−49. For the coordinated attack on Hugh and Boso, see chapter IV.2.

35 Annales Vedastini 880, 49. On the raids following Gauzlin's defeat, see Vogel, Normannen, 270−272.

36 Annales Vedastini 881, 50: *Interim Hludowicus rex gravi dolore contristatus, videns regnum deleri, convocato exercitu praeparat se ad proelium.*

37 Offergeld, Reges pueri, 381−382.

38 Annales Bertiniani 881, 244: *De quibus non modicam partem occisis ceterisque fugatis, et ipse Hludouuicus una cum suis retrorsum, nemine persequente, fugam arripuit, diuino manifestante iudicio quia quod a Nortmannis fuerat actum non humana sed diuina uirtute patratum extiterit.* Translation by Nelson, Annals, 222.

39 Annales Vedastini 881, 50: *Contra quos rex ire perrexit obviavitque eos in pago Witmau, in villa quae dicebatur Sathulcurtis, et commissum est proelium. Moxque Nortmanni fugam ineunt atque dictam villam deveniunt; quos rex insecutus est gloriosissimeque de eis triumphavit. Et patrata victoria ex parte coeperunt gloriari suis hoc actum viribus et non dederunt gloriam Deo. Paucique Nortmanni ex dicta villa egressi, omnem exercitum vertit in fugam, pluresque ex eis, videlicet usque ad centum homines, interfecerunt; et nisi rex citius equo descendens locum resistendi et audaciam suis donaret, omnes turpiter ex eodem loco fugiendo abirent.*

acted completely on his own—according to Hincmar, soon after the battle Louis, urged by some of his advisors, constructed a fort at Estruy,40 indicating again the influence of the nobles around the king—his crucial role in the battle appears as a sign of increased responsibility in the affairs of the realm. The battle of Saucourt undoubtedly was a major success for Louis,41 demonstrating his ability to muster sufficient resources to successfully challenge the Northmen, yet it appears not to have led to any agreements with the Northmen, who now retreated to Ghent before turning to the Maas where they erected a new camp at Elsloo.42

Louis' next campaign, this time not against the Northmen at the Scheldt, against whom he detached an army under Theoderic of Vermandois, 43 led him across the Seine towards the Loire to expel the Northmen from his realm. The region should have been under the protection of Hugh the Abbot's march. The king's activity in this region may be explained by two reasons: First, like under Louis the Stammerer, the forces of the march may not have sufficed to deal with the Northmen and the Bretons at the same time, thus necessitating the king's intervention. Second, the absence of Hugh the Abbot from the region, him being Carloman's main advisor, may have led to a lack of leadership that was now compensated by Louis taking command himself. Instead of using force, Louis now opted for a diplomatic strategy and concluded an *amicitia* with the Viking leader Hasting. 44 A number of factors may have contributed to this decision. Having split his forces, Louis' army may simply not have been sufficiently large for a military encounter. In fact, it seems likely that Louis had intended to reinforce his troops by forming an alliance with the Breton leaders—according to Hincmar, meeting with them was also a reason for him to cross the Seine.45 Finally, it seems that for him the Loire theatre was only of secondary importance to Louis. Having secured the Loire by turning Hasting and his men away from the region to aim for other regions at the sea,46 probably the Breton north coast,47 Louis next seems to have planned to move again against the Northmen in the north-east who had again extended their raids into the Western realm. At the moment of his unexpected death, the nobles of Francia had already assembled an army ready for a campaign, an army that was now taken over by Louis' brother Carloman, who left his own forces in Burgundy to continue the fight against Boso.48 It may have been due to this combination—a lack of resources to achieve

40 Annales Bertiniani 881, 244−245.


a military solution combined with Louis' focus on defending the Francia—that, in this case, he chose to opt for a seemingly purely diplomatic solution, allowing him to quickly return to the East.

#### **Carloman II**

In contrast to his brother's, Carloman's measures against the Northmen were strongly affected by his relations with the Northern nobles.49 According to Hincmar, in September Carloman was still with the army that had been assembled for his brother's next campaign against the Northmen.50 However, in November, when noting Hugh the Abbot's embassy to Charles the Fat, he remarks that the latter's "absence brought the utmost harm to this realm, since Carloman lacked the resources to mount resistance to the Northmen once certain magnates of this kingdom withdrew their support," leaving the realm open for the Northmen who now attacked Laon and Reims.51 This would indicate that the nobles turned away from the king at some point between September and early November, leaving the king, according to the archbishop of Reims, without the necessary forces to defend the realm. However, the *Annales Vedastini* report that in October Carloman moved against the Northmen who had installed themselves in the Condé, from where they ravaged up to the Somme.52 Carloman's campaign without the support of these nobles lead to a victory at Avaux, helping to assert the gravity of Hincmar's reproach against Hugh the Abbot.53 Carloman was by no means left defenceless, but was able to challenge the Northmen. However, as both Hincmar and the *Annales Vedastini* note, the king's victory did not put an end to the incursion. Carloman instead withdrew to Compiègne, leaving the Northmen to extend their ravaging up to the Oise. Hence, the threat posed by the king and his army was not sufficient to force the Northmen from the realm and he also appears either not to have pursued a diplomatic solution or failed to implement one.

Reinforcements brought by Hugh the Abbot reenabled Carloman to campaign again during the winter of 882/3, yet again without forcing the Northmen out of the realm or leading to successful negotiations while instead Saint-Quentin and Arras were plundered.54 The Northmen ravaging Flanders during the summer, in autumn 883 Carloman raised a new army to attack them in the Vimeux. This time the Northmen did not retreat but turned on the king both from land and from the river, forcing Carloman to quickly withdraw over the Oise and thus opening the

<sup>49</sup> See chapter II.2.

<sup>50</sup> Annales Bertiniani 882, 247.

<sup>51</sup> Annales Bertiniani 882, 249−250: *Sed absentia illius in isto regno maximum detrimentum fecit, quia Karlomannus non habuit unde Nortmannis posset resistere, quibusdam regni primoribus ab ipsius auxilio se retrahentibus. Quapropter usque circa Laudunum castellum uenerunt, et quae in gyro ipsius castelli erant depraedati sunt et incenderunt, disponentes Remi uenire, indeque per Suessiones et Nouiomagum pergentes, ad praefatum castellum expugnatum redire et regnum sibi subicere.*

<sup>52</sup> Annales Vedastini 882, 52.

<sup>53</sup> Annales Vedastini 882, 53; Annales Bertiniani 882, 250.

<sup>54</sup> Annales Vedastini 882, 53 and 883, 53.

river up to the Seine to them.55 The catastrophic outcome of the campaign led to a massive shift of power within the realm, both in regards to the relations between the different noble factions as well as those factions and the king. The report of the *Annales Vedastini* is straightforward: "Meanwhile, because the king was a juvenile, all of the magnates assembled at the palace of Compiègne to confer about what they should do."56 In regard to the disaster of the campaign, Carloman, whatever his position might have been before, was now set aside and the nobles, or, as we have argued, the faction around Hugh the Abbot as well as Gauzlin, but not yet the nobles from Francia,57 took over. This points to the complexity of the situation. Without the support of the local nobility, the royal forces were not sufficient to deal with the Viking threat. This left two options: either to integrate the nobles who had been so far excluded from the circle around the king into the rule, or to change the strategy to address the Viking problem. In fact, they implemented both, negotiating a treaty with the Northmen that offered them a tribute of 12.000 pounds of silver in exchange for them leaving the realm as well as an armistice until the payment was due58 and integrating the nobles from Francia into its implementation.59 However, if they were now willing to set aside their rivalries, why would they not have continued to seek a military solution with their newly united forces? It is hard to imagine that such a solution was not brought up during the meeting at Compiègne; yet, it would seem that they had evaluated the probable outcome of such an endeavour and had come to the conclusion that they would fare better by seeking a diplomatic settlement. If this was the case, this reveals how powerful the Northmen had become and how royal power was now limited even if it could rely on the united support of the nobles.

Carloman's role in these negotiations is difficult to ascertain. The report of the *Annales Vedastini* clearly states that the initiative for the change of course lay with the leading nobles. However, was Carloman really too young at that time not to fulfil his role as king? Examining this passage of the annals, Thilo Offergeld has pointed out that Carloman was already 17 years old and, hence, old enough to rule. According to him, exaggerating the king's young age was a pretext to overrule him in this matter—up until that moment, Carloman is portrayed as able to rule and lead an army.60 Following Offergeld's reading, this would mean that the aggressive strategy previously used against the Northmen was indeed the king's choice, yet that after his failure to protect the North a shift had taken place in the circle around him, leading to him being overruled. Nevertheless, the king appears

<sup>55</sup> Annales Vedastini 883, 54.

<sup>56</sup> Annales Vedastini 884, 55: *Interim, quia rex iuvenis erat, omnes principes Compendio palatio conveniunt tractaturi, quid illis esset agendum.*

<sup>57</sup> See chapter II.2.

<sup>58</sup> Annales Vedastini 883, 54 and 884, 55. Both entries report an assembly of the Frankish nobles and a mission of the baptised Viking Siegfrid to the Northmen (late 883 and early 884 respecitvely). This is certainly a duplication of the events.

<sup>59</sup> See chapter II.2.

<sup>60</sup> Offergeld, Reges pueri, 396−397.

to have been involved in the proceedings shortly afterwards. According to Regino of Prüm's chronicle, upon Carloman's death, despite having been paid the tribute, the Northmen returned to the Western realm arguing that they had concluded the treaty with the king and that his successor was to pay the same amount if he wanted them to leave.61 This points to a more central role of the king in the proceedings than the previous account of the *Annales Vedastini* indicates. It would seem that, since Carloman was considered as a partner in the deal, he had adapted his own course to that of the nobles and played his part in the negotiations. However, his position in regard to the nobles does not appear to have recovered entirely. When the tribute for the Northmen was collected, the Franks assembled an army in case that the Northmen did not leave the realm despite having received the tribute. Command over the army was then given to Carloman, who followed the Northmen up to the moment they embarked. Yet, there is a slight change in the phrasing used by the *Annales Vedastini* describing these events. It was "the king and the Franks" who followed the Northmen and it were "those Franks who were with Carloman" who returned to their homes once the Northmen were gone.62 The same phrasing is also used by the annals for Charles the Simple and the nobles supporting him during the fight against Odo, thus indicating the dominance of the circle around him.63 It would have been interesting to see how Carloman's relation to the nobles might have developed from this point on, especially after the threat posed by the Northmen had been lifted from the realm. However, his death shortly afterwards meant not only that Charles the Fat now became ruler of the Western realm, but also that the Northmen immediately returned to the Western realm.

Carloman's reign demonstrates the dependence of the king on the support offered by the nobles in the case of his politics towards the Northmen. While he was able to assemble forces sufficient to challenge them in the field without the support of the nobles from the northern part of the realm, these forces, nevertheless, appear to have been too small to prevent the Northmen from their ravages into the realm and, if he had had such intentions, to improve his position in negotiations. Failures in protecting the realm could lead to a massive shift in the balance of power between the nobles and the king, which, in turn, could result in a takeover of responsibilities by the nobles around the king. At the same time, this takeover, and the subsequent change of strategy leading to a diplomatic solution, also reveal how limited military means actually were even if the nobility stood united.

<sup>61</sup> Regino, Chronicon 884, 122.

<sup>62</sup> Annales Vedastini 884, 55−56: *Tandem soluto tributo, mense Octobrio finiente adunantur Franci, ut, si Nortmanni inmutari fidem vellent, eis resisterent. Nortmanni vero sua castra incendunt atque ab Ambianis recedunt. Rex vero et Franci transito Hisa lento itinere eos insequuntur. Praedicti vero Dani iter agentes Bononiam veniunt ibique agentes consilium, quid sibi faciendum esset, pars illorum mare transiit, atque pars Luvanium in regno quondam Hlotharii, ibique sibi castra statuunt ad hiemandum. Franci vero qui cum Karlomanno fuerant redierunt ad sua loca.*

<sup>63</sup> Offergeld, Reges pueri, 454−456.

Seemingly, they estimated their chances to secure the realm by military means too small to pursue this strategy any further.

#### **Charles the Fat**

For Charles the Fat, these framework conditions differed considerably. First, the Northmen, upon leaving Carloman's realm, had split into two groups, one sailing to England, the other one erecting a camp at Leuven and probably uniting with other Northmen who had, up until this moment, remained beyond the Scheldt,64 thus reducing the threat they posed, at least to a certain degree. Furthermore, Charles could rely not only on West Frankish forces, but also on those from Lotharingia, the East Frankish realm and Italy.65 However, at the same time, this also meant that he had to protect a much larger area and deal with a multiplying number of problems. Mobilising troops from one part of his realm to protect another meant that they had to operate far from their homes, putting a strain on supplies and requiring difficult coordination. Charles' first measures reveal the whole complexity of this situation. Although he was absent, he ordered forces from the Western realm and Lotharingia to move against the Northmen at Leuven.66 When these Northmen turned towards the Seine, another army was assembled, consisting of forces from Neustria and Burgundy and put under the command of Ragenold of Maine,67 yet it dissolved after his death.68 In the absence of the emperor, the West Frankish nobles now initiated their own measures, constructing a fort at the Oise to block the river and putting it under the command of Count Aledramnus as well as fortifying Paris.69 The role of the emperor in these different actions is hard to grasp. While the *Annales Vedastini* clearly state that the early mobilisation and attack on the Northmen at Leuven was initiated by Charles, when it comes to the assembly of the second army under the command of Ragenold, they remain silent. The large-scale cooperation of forces from different regions under the command of a noble otherwise invisible in the sources, however, does seem to indicate that this was the emperor's doing as well. On the other hand, the defensive measures taken after Ragenold's death seem to have been taken at the initiative of the nobles—at least in the case of the fortification of Paris, the *Annales Vedastini* indicate Gauzlin was responsible.70


This interplay between measures initiated by local leaders and the emperor also characterises the events around the siege of Paris. While the defence of the city was led by Gauzlin and Odo, the emperor mobilised forces from his other *regna* to relieve the besieged. His first response to the Viking assault appears to have been quickly organised, with an army lead by Duke Henry arriving at Paris in spring 886,71 again revealing Charles' ability to delegate the defence to trusted men if it seemed necessary to him. When the first campaign failed, in the late summer of 886 he himself led a second army to the Western realm, his presence indicating the importance he now attributed to the situation.72 Charles forces appear to have been of considerable size, strong enough to force the Northmen to abandon one of their camps and to withdraw their forces over the Seine. However, instead of attacking them, he initiated negotiations, agreed to pay a tribute and opened up Burgundy for them to spend the winter.73 Charles' decision to negotiate has often been interpreted by scholars as a sign of his military weakness and related to his deposition a year later.74 However, Simon MacLean argues that there are no signs in the contemporary sources that Charles' reputation had suffered from this change of strategy. In fact, as he continues, the treaty must have been the result of a strategy discussed with the leading West Frankish nobles.75 As lack of military strength was apparently not the issue, there must have been other reasons for this decision. The *Annales Vedastini* report that negotiations were initiated because winter was approaching.76 Given the size of Charles' army, supply would have become a serious problem during the following months,77 making it necessary to disband at least some of his forces, if not send the entire army home. Hence, negotiating might have been a necessity dictated by the circumstances, the only means to ensure the end of the siege if it had not been the emperor's intention from the beginning. In this context, it is worth pointing out that Gauzlin had already opened negotiations with one of the Viking leaders, Siegfrid, during the siege and concluded an *amicitia*. 78 This group of Vikings seems not to have taken part in Charles' treaty since they did not turn towards Burgundy but entered the Oise,79 leaving the realm before the tribute had been delivered.80 Hence,


paying off the Northmen was considered a valid option by at least some of the leading West Frankish nobles—support Charles could count on when opening negotiations.

The special situation of Charles the Fat's reign, with the emperor ruling over all of the Carolingian *regna*, reveals three layers of measures that dealt with the Viking threat. In the wake of royal absence from the West, the defence organised by the local nobility surfaces much more than in the previous periods. In this context, the case of the fortress constructed at the Oise is of particular interest since command over it was given to Count Aledramnus.81 In the early 880s, Aledramnus and his brother, Count Theoderic of Vermandois, controlled a block of *honores* north of Paris and repeatedly appear in the sources as leaders of operations against the Northmen, indicating that they were key figures in the defence of the region.82 This development was continued under Odo when Theoderic's *honores* were passed on to Heribert I of Vermandois, who subsequently often appears as one of the leaders of, as Karl Ferdinand Werner phrased it, the Oise line.83 At the same time, both of them, as well as Heribert later on, can be seen in close contact with the emperor,84 thus revealing how closely linked these different layers of organisation actually were. In support of the local defence, Charles furthermore mobilised troops from other parts of his realm, delegating command to suitable commanders while he was needed elsewhere. The top layer, finally, consisted of Charles' direct intervention with even greater forces. While there was certainly no lack of effort made in the defence of Paris, its outcome reveals the limits of such purely military measures. Even with the resources of the entire empire at his disposal, forcing the Northmen to leave the realm appears to have been impossible, making a diplomatic solution preferable to Charles and the leading nobles of the West Frankish realm.


<sup>81</sup> Annales Vedastini 885, 57−58.

<sup>82</sup> MacLean, Charles, 86−89 and MacLean, Kingship, 106−107.

#### **V.1.2 Strategies of containment: Odo**

Odo appears as the diametrical opposite to Charles the Fat, a ruler whose resources were mostly based on his Neustrian possessions, with limited support from the nobility and a legitimacy deficit. Odo's first campaigns reveal the restrictions placed on him by these structural problems. While his victory against the Northmen at Montfauçon boosted his reputation considerably, resulting in Arnulf inviting him to a placitum and acknowledging him as king,85 this battle appears not to have had any major impact on the activities of the Northmen. Meaux was besieged and burnt down, its remainders serving as base for the Northmen over the summer.86 In the meantime Odo fell back on a defensive strategy, using his army to protect Paris from the advancing Northmen without initiating any further measures against them.87 At this point, his lack of offensive action may have been due to his need to stabilise his reign: from his defensive position at Paris he moved to Reims to be crowned with a crown sent by Arnulf while early in the next year he went to Aquitaine to treat with Ramnulf of Poitiers and receive his submission.88 This does not mean that Odo completely abandoned the defence against the Northmen. When he moved to Aquitaine, the *Annales Vedastini* report that he only took a few men with him—which may indicate that he left his army in the North to monitor Viking activities—and, furthermore, that "he hastened back to Francia" because of them.89 However, at the same time he seems to have been unable to prevent the Northmen from entering the Loire where they now devastated "Burgundy, Neustria and a part of Aquitaine without encountering resistance."90 The measures Odo took against the Northmen upon their return to the Seine mark the beginning of a slightly adjusted approach to the Viking problem. While he moved against them, he opened negotiations and bought their withdrawal from the city, making them turn towards Brittany.91 Over the course of the past years, especially during the siege of Paris, Odo must have been able to closely study the

85 Annales Vedastini 888, 65: *Sed dum illi haec agerent, contigit Odoni regi per Dei misericordiam inopinata victoria. Nam die nativitatis sancti Iohannis baptistae cum parvo exercitu Danorum exercitum obviavit super Axonam fluvium commissoque proelio mox victor extitit. Quae victoria non modicam illi contulit gloriam. Post haec ab Arnulfo convocatur ad placitum.* See also chapter IV.3.

91 Annales Vedastini 889, 67−68.

<sup>86</sup> Annales Vedastini 888, 66.

<sup>87</sup> Annales Vedastini 888, 67: *Circa autumni vero tempora Odo rex adunato exercitu Parisius venit, ibique castra metati sunt prope civitatem, ne iterum ipsa obsideretur. Nortmanni vero per Maternam in Sequanam regressi indeque navigantes et iter per terram facientes, Luvam fluvium ingressi, circa eius littora sedem sibi firmant. Odo vero rex Remis civitatem contra missos Arnulfi perrexit, qui ei coronam, ut ferunt, misit, quam in aecclesia Dei genitricis in natale sancti Brictii capiti impositam, ab omni populo rex adclamatur.*

<sup>88</sup> Annales Vedastini 889, 67.

<sup>89</sup> Annales Vedastini 889, 67: *Aquitanios itaque rex ex parte receptos, festinavit propter Nortmannos redire in Franciam.* 

<sup>90</sup> Annales Vedastini 889, 67: *Dani vero more suo Burgundiam, Neustriam atque partem Aquitaniae, nullo resistente, igne et ferro devastant.*

Northmen and come to a realistic assessment of the options he now had in regard to their strength and his own resources, preferring to act defensively rather than to attack and to seek diplomatic solutions when possible.

This more careful approach to the Viking threat also becomes visible over the course of the following year when the Northmen returned to the Seine, splitting in two groups, one entering the Oise, the other under Hasting operating at the Somme. Odo appears to have fancied standing a better chance against these weaker forces, as he tried to attack a part of the first group before again falling back to his defensive strategy, assembling an army at the Oise to prevent the Northmen from roaming deeper into the realm.92 While this strategy had its advantages, effectively protecting a large part of the realm by limiting Viking mobility, it also had its drawbacks as it revealed Odo's incapacity to drive them away from the areas they were currently threatening. This resulted in a conflict of interests with the local nobles, as demonstrated by the case of Abbot Rodulf of Saint-Vaast, who now concluded an agreement with Hasting, allowing the latter to circumvent Odo. 93 While Rodulf changed his mind shortly afterwards and instead attacked the Northmen, the case, nevertheless, reveals how problematic Odo's strategy was. It was probably also due to the pressure placed on him by this event that during the following year Odo made new offensive efforts, trying to deal with the various Viking groups. The results were hardly encouraging. While he managed to deprive one group of their plunder (although without actually being able to pin them down, possibly forcing them into Lotharingia),94 his attempt to attack a second group at Amiens led to his army being put to flight.95 These campaigns reveal the dilemma of Odo's situation. Containing the Northmen was the best strategy, yet one that needed the support of the local nobility, support that appears to have been increasingly uncertain, especially in regard to the conflicts breaking out over the course of the following year. This lack of support from the local nobility was probably why Odo now reverted back to an offensive strategy, a decision that was facilitated by the Northmen splitting up their forces. However, at the same time, the Northmen avoided open battle: they moved through almost impassable terrain and scattered


when pinned down, only to regroup soon after while at the same time exploiting any opportunity offered to them. In the end, the withdrawal of the Northmen was not brought on by Odo's measures, but by a famine in Francia.96

Odo's caution surfaces again four years later when the Northmen returned to the Seine. At this point, his attention was still turned to the conflict with Charles the Simple, a circumstance that the author of the *Annales Vedastini* does not fail to remark on: "And because the king busied himself with other affairs, he increased the great evil coming on him and the realm."97 Abbo of Saint-Germain went even further, accusing Odo of not caring at all and neglecting "the sheep sent to him by Christ," thus forsaking his own honour.98 While Abbo's harsh critique certainly was an overstatement, Odo indeed only became active again during the following year when he tried, without success, to attack a group returning from their raid. The group appears to have been rather small since "for fear of the size of [Odo's] army" they retreated into the Seine.99 However, Odo's general assessment of his chances against them appears to have remained unchanged. From the Seine, the Northmen ravaged over the course of the summer, apparently again without encountering much resistance. Later that year, Odo, while now having pacified the realm and hence being able to give the Viking threat his full attention, nevertheless decided not to challenge the Northmen but to, as in 889, conclude a treaty "to ransom the realm."100 While this last treaty has often astonished scholars who attribute its conclusion to Odo being weakened by sickness101—an explanation that would be backed by his death following soon after—it appears to be in line with the earlier measures taken by Odo. He had mostly favoured a defensive strategy, attacking only splinter groups while paying off larger armies without seeking battle. The Northmen returning to the Seine appear to have soon increased in


strength, shifting the balance of force in their favour. Odo, given the limitations of his resources, always carefully weighed his options. As the military ones seemed too risky, he chose the diplomatic solution.

#### **V.1.3 Return to the old strategies: Charles the Simple**

Charles the Simple abandoned the caution shown by Odo and reverted back to the approaches shown by his brothers, trying to attack the Northmen. When they returned from the Loire, where they had passed the winter, Charles mustered an army and pursued them. While the *Annales Vedastini* describe the size of his forces as a "small army," it seems to have been strong enough to be considered a serious threat by the Northmen who, after a first encounter that left many of them wounded, averted further contact by once again marching through barely passable terrain.102 Similarly, it may have been due to his return to Francia after his campaign into Lotharingia over the summer that they chose to pass the winter in Burgundy where they were defeated by Richard the Justiciar and forced back to the Seine.103 Furthermore, Charles also appears to have been able to pin them down to the river since the *Annales Vedastini* only report further raids in November 899, when Charles had to deal with Baldwin II of Flanders, who taken possession of Péronne.104 This observation makes it possible to draw some conclusions about the balance of power between the king and the Northmen. Charles' army, if the account of the *Annales Vedastini* is not to be completely dismissed, was rather small. Yet, at the same time, it was strong enough to be considered a serious threat by the Northmen, who only started to move again when Charles had to turn his army away from them. Apparently, the Viking forces Charles had to deal with in Francia were not as strong as the ones Odo had faced in the preceding years. Unfortunately, the narrative sources remain silent about any groups leaving the army, but it seems possible that a part of the Northmen had either not returned to Francia from the Loire in 898 or had left the realm over the course of the winter. Hence, Charles' room for manoeuvre appears to have increased considerably, especially when he was able to join his forces with those of Robert, Richard and Heribert in early summer 900 when they met at the Oise to deal with the Northmen remaining there.105 However, at this point, his attention was drawn away from the

105 Annales Vedastini 900, 81−82.

<sup>102</sup> Annales Vedastini 898, 79−80: *Nortmanni vero verno tempore rediere ad naves, vastatam Aquitaniae partem atque Neustriam, insuper plurima eversa castra, interfectis habitatoribus. […] Post haec rex Karolus cum exercitu parvo Nortmannis a praeda revertentibus in pago Vitmau iuxta quandam... insecutus, aliquibus suorum interfectis plurimisque vulneratis, Nortmanni more solito loca inoportuna tenentes rediere ad naves.*

<sup>103</sup> Annales Vedastini 898, 80. For Richard's battle at Argentueil see also Chronicon S. Benigni Divionensis, 113; Annales S. Columbae Senonensis 899, 104; Chronicon S. Petri Vivi Senonensis, 70. See Vogel, Normannen, 381, n. 2 for further references.

<sup>104</sup> Annales Vedastini 899, 81.

Northmen once again when the conflict with Robert broke out. Hence, Charles, during these early years of his reign, appears to have been capable of dealing with the Viking threat, yet, like his predecessors, still did not possess the resources to force the Northmen to leave the realm for longer periods of time. Whether cooperation with Robert and Richard would have changed the balance of power in his favour has to remain open. However, the cooperation of these nobles points also to another observation. During these early years, nothing indicates that Charles attempted to negotiate with the Northmen. Yet, since Robert, Richard and Heribert provided forces for the return to the rather offensive strategy previously mainly shown by his brothers, this seems to indicate that they supported this strategy and might even point to them being reluctant to negotiate, as Odo so often had preferred to do.

With the end of the *Annales Vedastini*, only small traces of the Viking incursions and the royal measures to keep them in check remain.106 Edward the Elder's victories over the Northmen in 909 and 910 probably led to an increase in Viking activity on the Continent around the same time.107 Burgundy appears to have been targeted in 910, when the abbey of Saint-Colombe of Sens was fortified108 and Bishop Madalbert of Bourges was killed by the Northmen.109 This incursion was probably the prelude to the campaign of 911, which lead to the battle of Chartres and the subsequent treaty of Saint-Clair-sur-Epte. While a number of sources mention the battle,110 the only one providing further detail is Dudo of Saint-Quentin to whose account we will turn shortly. What seems to be clear from these sources is the leading role of both Robert of Neustria and Richard the Justiciar (as well as, according to Dudo, Ebalus Manzer of Poitiers)111 in the victory. The participation of the first two seems certain since there is also a letter preserved demonstrating their efforts to coordinate their forces before the battle.112 None of these sources provide any indication about Charles' involvement in the campaign. However, the question is which form such an involvement would have taken. Louis the Stammerer, when crossing the Seine, only reacted to a call from Hugh the Abbot, Louis III's intervention in the area also included negotiations


with the Bretons. In fact, Robert appears to have been quite capable of defending the Loire on his own. Probably in 903, Tours was attacked and burnt down by Vikings113 while from 915 onwards the Loire again appears to have been penetrated by the Northmen who used the river mainly for raids into Brittany.114 This latter group was likely the same one that Robert fought in 921 before ceding Brittany and the pagus of Nantes to them in return for the conversion.115 On none of these occasions do the sources report any involvement from Charles' side, the last one even seems to indicate that Robert enjoyed some liberty when it came to negotiating peace deals.116 This liberty was nothing new.117 Already Robert the Strong had negotiated a tribute payment with a group of Northmen allied with the Breton Salomon118 and also Hugh the Abbot appears to have concluded a treaty without the king, baptising a number of Northmen.119 At least on the latter occasion, Charles the Bald appears to have given his consent only afterwards. Furthermore, as we have seen, also Gauzlin, in absence of Charles the Fat, negotiated an agreement with Siegfrid, one of the Viking leaders besieging Paris, paying him a tribute for agreeing to an *amicitia*. Just like in the former two cases, also Robert may have intended to ask for Charles the Simple's confirmation of the deal. Thus, in the case of the campaign leading to the battle of Chartres, it seems likely that while Robert and Richard were in charge of the forces, they, nevertheless, acted in Charles' name and probably with him an overseeing role.120 As we have argued in the previous chapter, based on one of Robert's private charters,121 it appears that Robert acted on the king's behalf when fighting the Northmen. If this was indeed the case, the campaign leading to the battle of Chartres would be an example of how Charles was able to coordinate forces from different parts of his realm in an attempt to organise an appropriate response to the incursion. Charles' ability to do this would be confirmed if the diplomas issued in June 914 during a couple of days are indeed the signs of a campaign against the Breton march in the wake of the


breakdown of order in Brittany due to a Viking incursion, with the participation of forces from Neustria and Burgundy.122

While the examples of 900, 911 and 914 demonstrate Charles' ability to mobilise and coordinate forces in defence of the realm, on another occasion he does not seem to have been able to do this as easily. Flodoard's account in the *Historia Remensis Ecclesiae* concerning the Hungarian invasion of Lotharingia reports that Charles' call to arms was only answered by Archbishop Heriveus of Reims.123 As we have argued above, Flodoard's report appears to have been grossly exaggerated and constructed in a way to portray Heriveus in the best light possible.124 Nevertheless, it would appear that, on this occasion, the response to Charles' call was only a limited one. His first call would undoubtedly have gone to his Lotharingian supporters and only after that to his West Frankish men because the Hungarian attack came from the east. Indeed, Flodoard only speaks of the "great men of the Franks" (*proceres Francorum*) failing to respond, a designation that does not necessarily include the nobles from Lotharingia.125 It may even be the case that Heriveus joined the king's forces early on not because he had been called, but out of his own interest since the church of Reims had property in Lotharingia.126 Be that as it may, Flodoard's report remains unsatisfactory, not even reporting whether the king's measures against the Hungarians were crowned with success or resulted in failure.

However, military campaigns were not the only means the king used to protect his realm from the Northmen. As we have mentioned above, the Anglo-Saxon chronicle reports a Viking war leader in England, Earl Thurcetel, preparing for a venture into the West Frankish realm with the consent and aid of King Edward. 127 In this context, Charles' marriage with Edward's daughter Eadgifu takes on a new meaning, namely that of an alliance between the two kings in the shadow of the Viking incursions which prevented Edward from giving further aid to Northmen aiming for Charles' *regnum*. This marriage reveals that Charles was willing to follow other ways than his predecessors, to break out of the old patterns of military


campaigns and paying tribute by thinking further ahead and depriving at least some Northmen of their base and hence hampering their raids into his realm.

Thus, Charles dealt with the Viking threat on several levels. In terms of military measures, he appears to have been able to mobilise sufficient forces to challenge the Northmen and pin them down. If necessary, he was able to rely on forces from the entire realm, revealing the full extent of his possibilities. However, at the same time, these large scale operations also reveal his limitations: carrying them out depended on the cooperation of the leading nobles. If one of them stood against him, his ability to defend the realm suffered a serious blow, not only because he then was missing part of his support, but also because, as in the case of Baldwin, he needed to shift his attention away from the Northmen and concentrate on the nobles opposing him. As under his predecessors, military operations appear to have yielded mixed results at best and were not able to achieve a durable solution. It is altogether more interesting that Charles appears to have come to that same conclusion and tried to pursue new strategies. Depriving the Northmen of their bases in England was one of these means. Another was the adoption of a new diplomatic strategy, to which we will now turn.

#### **V.2 Diplomatic solutions**

#### **V.2.1 Dudo's** *De moribus et actis primorum Normanniae ducum*

Raising armies and trying to defeat the Northmen were only one means of dealing with the Viking threat. Another one, seemingly much more effective, was to come to an agreement with them. As we have seen above, apart from Louis the Stammerer, each of the kings concluded treaties with the Northmen, involving the payment of a tribute for the safety of the realm.128 Charles the Simple is not reported to have made such payments, yet given the absence of narrative sources for almost 20 years of his reign, this is hardly surprising, and it cannot be taken as evidence that Charles' politics differed from those of his predecessors. In fact, Charles concluded at least one treaty with the Northmen, the treaty of Saint-Clairsur-Epte, which was probably agreed upon in the wake of the battle of Chartres in 911 and followed by the settlement of the Vikings in Normandy. The main source for these events is Dudo of Saint-Quentin's work *De moribus et actis primorum Normanniae ducum,*129 a highly problematic narrative account of the foundation of the duchy of Normandy. Dudo belonged to a Frankish cultural milieu, had strong links to a church in the old Carolingian heartlands130 and seems to have been

<sup>128</sup> A concisive account of the treaties between Northmen and Francs is provided by Neifeind, Verträge. As he (Verträge, 22) has pointed out, none of these are preserved in written form.

<sup>129</sup> Lair, De moribus.

<sup>130</sup> Labande, Historiographie, 759−760 and Duby, Ordres, 108 argue for Reims. Shopkow, World, argues for Liège.

strongly influenced by Bishop Adalbero of Laon to whom he dedicated his work.131 A canon at Saint-Quentin, he was sent by Albert of Vermandois as an ambassador to Richard I, for whom he then served as deacon. After Richard's death, he served his son and successor Richard II as chaplain and chancellor. Thus, he was integrated in the Norman court and acquainted with its procedures.132 Richard I had commissioned the *De Moribus*, and the commission was renewed after his death in 996 by his son Richard II and half-brother Raoul, count of Ivry. The first version was probably drafted between 996 and 1015, with a revision following afterwards.133 Dudo used written accounts, first of all Flodoard's *Annales*, but his chief informants were members of the ducal family: Raoul of Ivry, Richard I's half-brother, the duchess Gunnor and probably also Archbishop Robert. 134

Dudo's *De moribus* was criticised vehemently by Henry Prentout, who described him as "Poète et rhéteur, amateur de pittoresque, voilà trois défauts impardonnables chez un historien; ajoutons-y le goût du délayage, le besoin de raconter en six pages ce qui pouvait se dire en quelques lignes, la manie d'introduire dans sa narration des dialogues, des discours…"135 Since then Dudo has been treated more kindly by scholars136 who have revealed the ideological background of *De moribus* as a lineage-story of the Dukes of Normandy.137 While this emphasis on the literary character of Dudo's work in general has shifted the focus away from the question of how far his account is based on historical events,138 the problem has not been forgotten entirely by scholars, leading to very different assessments.139

An interesting case study which serves to shed some light on this problem is Pierre Bouet's on Dudo's depiction of the Viking Hasting, the main character of *De moribus*' first book. He argues that none of Hasting's actions were invented by Dudo. In fact, allusions to most events Dudo has Hasting participate in can be found in other sources. However, Dudo not only changed the chronology, but also used Hasting as a symbolic figure of wrongdoing by ascribing the devastation caused by other Viking leaders such as Oscar, Siegfrid, Björn, Ragnar and Hundeus to him.140 This, of course, renders his whole account extremely problematic as a source of hard facts. But we should not rely on the case of Hasting to discern the


reliability of the entire account of *De moribus*. Did Dudo apply the same technique to the main character of his second book, Rollo? According to Dudo's account, Rollo first appears in the Frankish world when he installs himself at Walcheren,141 an event that probably reflects the arrival of the Great Army from England in 879.142 His subsequent move to the Scheldt and the occupation of Condé during the winter143 are also reported by the *Annales Vedastini* for the years 880−882, referring to an incursion led by Siegfrid and Gottfrid. 144 Count Reginar being captured by Rollo on this occasion undoubtedly stems from Regino's account, who reports the same for a certain count Eberhard in 880.145 The following turn of Rollo towards the Seine, dated to 876 by Dudo, 146 also finds a double correspondence in the contemporary sources. For the very same year the *Annales Bertiniani* note the arrival of about 100 Viking ships entering the river 16th September,147 although in accordance with the following events described by Dudo more match the advance of the Northmen on Paris in 885.148 In reaction to the incursion, the Franks are then said to have gathered an army under the command of Ragenold and to have sent Hasting to treat with Rollo149—Ragenold is also named as the Frankish leader by the *Annales Vedastini*150 while Hasting is undoubtedly a reference to Siegfrid, who was sent to deal with the Northmen attacking Amiens the year before.151 Even an allusion to the mockery of Dudo's Northmen can be found in the *Annales*, namely at Leuven in 885.152 Rollo's march on and siege of Paris, including the death of Ragenold, 153 are equally well known from other sources,154 although Rollo does not appear in them. Thus, like Hasting, Dudo's Rollo appears to be constructed out of several other, often unknown Viking leaders raiding the Francia during the 870s and 880s. The general tenor of this construct, however, is diametrically opposed to that of Hasting. Dudo's description of Rollo raiding into Burgundy and along the Loire preceding the battle of Chartres for example can also be traced in the contemporary Frankish sources.155 Yet, Dudo adds some new aspects that portray Rollo, although still a pagan, in the best light possible: "Some of his men went


to Burgundy; they sailed up the Yonne to the Saône, laying waste the lands next to the rivers everywhere as far as Clermont, and pushed on into the province of Sens, and returned to meet Rollo at Saint-Benoît, ravaging everything. However, on seeing the monastery of Saint-Benoît, Rollo refused to defile it, nor would he allow that province to be despoiled, for Saint Benedict's sake."156 Even the battle of Chartres is reinterpreted, since in Dudo's presentation it is not a defeat but rather a non-victory in which Rollo ordered a tactical retreat.157 As Eric Christiansen puts it: Rollo was Dudo's "Viking Aeneas, following a destiny first laid down by Vergil for the Roman precursor," developing through baptism "into a Clovis, benefactor of the churches and divinely-sanctioned conqueror."158

#### **V.2.2 The treaty of Saint-Clair-sur-Epte**

Dudo's technique of rearranging and relabelling his sources while mixing them with actual events makes his account of the treaty of Saint-Clair-sur-Epte highly problematic, which means that each term needs to be evaluated in the context of other sources. According to Dudo, the terms agreed upon consisted of a donation of the land between the Epte and the sea, to be held in perpetuity as personal property, as well as Brittany to live from since their new lands had been left uncultivated for a long time. Rollo then swore an oath of fealty by placing his hands between those of the king and in turn received Charles' daughter Gisela as his wife. The ceremony was to be concluded by Rollo kissing the foot of the king, however, as he was unwilling to do so, he ordered a warrior to act in his stead. Instead of kneeling down, the warrior remained standing, raised Charles' foot to his lips and kissed it, causing the king to fall on his back.159 Directly connected to the treaty was Rollo's baptism which took place some time after the meeting at Saint-Clair-sur-Epte. 160

<sup>156</sup> Dudo, De moribus II, c. 21, 161: *Sui autem in Burgundiam pergentes, perque Ionam in Sigonam navigantes, terrasque amnibus affines usque Clarum-Montem undique secus devastantes, Senonis provinciam invaserunt, atque cuncta depopulantes ad Sanctum Benedictum contra Rollonem revenerunt. Videns autem Rollo monasterium Sancti Benedicti, illud contaminare noluit, nec praedari illam provinciam propter sanctum Benedictum permisit.* Translation by Christiansen, Dudo, 42.

<sup>157</sup> Dudo, De moribus II, c. 23, 162: *Rollo vero, more soliteo constanter irruit super illos atque, in primo apparatu belli, superavit, fortiter dimicans illos. … Cernens autem se Rollo inter utrumque exercitum stare, seque non praevalare suosque decrescere, transiens per medium illorum, coepit ab eis declinare, ne praeoccuparetur morte.*

<sup>158</sup> Christiansen, Dudo, XIX. On the origins of the historical Rollo, see Douglas, Rollo; Musset, Origine and Renaud, Vikings, 47−55.

<sup>159</sup> Dudo, De moribus II, c. 28, 168−169 and c. 29, 169. On this kiss, see Hattenhauer, Aufnahme, 22−37 and Bauduin, Rituel. Both admit that this ritual could have taken place, but disbelieve the form reported by Dudo, a position that we share.

<sup>160</sup> Dudo, De moribus II, c. 30, 170.

In general, baptism seems to have been a prerequisite for a treaty between Carolingians and Northmen.161 And, in fact, Flodoard notes that Christianisation was part of the agreement.162 The conversion itself is well documented and has been made subject of a study by Olivier Guillot.163 Here, two sources are central, a letter from Archbishop Heriveus of Reims to his colleague, Wido of Rouen164 and another one from Pope John X to Heriveus. 165 Both letters are part of an exchange originating at Rouen, in which Wido first seems to have asked the other archbishop what to do with the Northmen who, despite having been baptised, repeatedly fell back on their pagan habits. Heriveus then brought the matter to John X and, having received his reply, composed his advice in form of 23 chapters. It is hardly surprising that the conversion of the Northmen did not come without difficulties and took at least several years. As Guillot has emphasised, this was an evolutionary process, and, in the end, a successful one.166

Flodoard also mentions another part of the treaty: a donation of land, several *pagi* by the sea devastated by the Vikings, as well as the city of Rouen and all its dependencies.167 This is further confirmed by a royal diploma issued in 918, indicating that a part of the possessions of the abbey of Croix-Saint-Ouen in Mérezais at the Eure had earlier been given to Rollo and his companions.168 Lucien Musset has tried to reconstruct the borders of the territories ceded to the Vikings and deduced that they were constituted more or less by the rivers Bresle, Epte, Eure and Avre in the east and south, while nothing is known about the western frontier.169 The devastation of this area as described by Flodoard cannot be doubted, given the frequency of the Viking incursions of the preceding decades.170 According to Jacques Le Maho, during the early years of Odo's reign, due to this constant threat,


Rouen was transformed into a "ville-refuge."171 The fortifications supposedly were restored and a large part of the rural population sought refuge there, while the domains they had abandoned fell under royal control.172 If this was indeed the case, this would undoubtedly have facilitated the treaty concluded between Charles and Rollo. Royal control over a large part of the territory, as well as its deserted state, would have ensured that the interests of Frankish nobles were hardly affected and the abandoned villages could then have been occupied by the Northmen, if they had not already done so.173

The cession of land is commonly assumed to have been tied to another condition, the protection of the Seine from further Viking incursions by Rollo and his men.174 The basis for this assumption is Charles' diploma of 918 which states that the monks of Saint-Germain were given the land of Croix-Saint-Ouen, "apart from the parts which we have given to the Northmen at the Seine, that is to say Rollo and his companions, for the protection of the realm."175 However, the phrasing "for the protection of the realm" is rather ambiguous. While it may indeed indicate that Rollo was charged with preventing other Northmen from entering the Seine, it may as well simply be a phrasing used to legitimise the ceding of land *per se*—to protect the realm from Rollo himself.176 This question leads us to examine what role Rollo played within the political hierarchy of the realm from this point onwards.177 For Karl Ferdinand Werner, Rollo was made count of Rouen and, the *pagus* being part of Neustria, put under the jurisdiction of Robert. 178 Pierre Bauduin, on the other hand, interprets Rollo as a *marchio* in his own, who was


installed by Charles to act as counterweight to Robert. 179 Both readings assume that in the wake of the treaty Rollo had become a fully integrated member of the Frankish nobility. Yet, Charles' diploma does not refer to him as *comes* or *marchio* but plainly as Rollo. The same is also true of the main narrative source of the time, the writings of Flodoard, who is usually very precise about the application of titles. As Eleanor Searle points out, for him Rollo and his son William Longsword are "chieftain [*princeps*] of the Norse," but never *comes*. 180 Concerning the integration of Rollo into the structures of the realm, we should also take the actual content of the diploma into consideration. Pierre Bauduin argues that the land was given to Rollo *in alodo et in fundo* ("as an allod and property"), a phrasing that implies the unconditional and permanent character of the donation.181 However, these expressions are only applied by Dudo182 and not, as Bauduin seems to imply, by Charles' diploma. Hence, the legal foundations of the ceding are entirely unknown. From what the sources are able to tell us, it seems that Rollo's position in the realm was in fact far less formalised than often assumed. This is not to say that Rollo did not become or had not already become integrated into Frankish society. Indeed, he was no stranger to the leading nobles of the adjoining counties. Several years before 911 he had already taken a woman called Poppa183 as his companion who had given birth to his son William. According to Dudo, Poppa was the daughter of a certain Berengar, possibly an Unruoching.184 Following the tradition of the *Annals of Jumièges* however, Poppa was the daughter of Count Wido of Senlis and sister to Bernard.185 The question of her heritage has yet to be answered,186 but, as Pierre Bauduin has emphasised, the important point is that by this relation Rollo was introduced into the Frankish networks of alliances and parentage.187

These networks lead to the final part of Dudo's report of the treaty of Saint-Clair-sur-Epte, namely the marriage between Rollo and Charles' daughter Gisela. Marrying his daughter off to Rollo would certainly have tied the latter even more strongly to the king than his conversion and the donation of land. Yet, was Rollo really important enough to enter into a marriage alliance with Charles? Dudo certainly does his best to portray Rollo as an all-powerful Viking warleader. Yet, the political reality appears to have been rather different. When Charles opened


negotiations with the Northmen after the battle of Soissons, his principal target appears to have been Ragenold, the leader of the Vikings at the Loire. When he reacted to Charles' call, Flodoard reports that was joined "by many from Rouen"188—Rollo is not mentioned at all. Later that year, these Northmen from Rouen concluded a treaty with Heribert II of Vermandois, Archbishop Seulf of Reims and other Frankish nobles, granting them more land beyond the Seine which was confirmed by King Raoul shortly afterwards.189 Flodoard does not mention Rollo either at this point or during the following year, although the Northmen he speaks of are clearly from Rouen.190 Even when these Northmen broke the treaty the following year, Flodoard again only calls them "Northmen from Rouen," again without mentioning Rollo191 and continues to do so for the subsequent campaign against them.192 Only when Raoul directly attacked the Northmen within their territory is Rollo finally mentioned, this time as their *princeps*, sending troops from Rouen in support of a *praesidium* threatened by the Frankish army.193 Flodoard's account seems to indicate that Rollo's control over the Northmen at Rouen was rather limited at this point, that he was only one leader among others and that these others were able to operate independently from him.194 Viking armies were heterogeneous groups, consisting of numerous warbands under their own leaders who had joined for some time, but could always fall apart again to pursue their own


goals.195 Hence, it seems plausible to assume that early on, Rollo's position among the Northmen now settling around Rouen was not a dominant one and that his rise over the other leaders only took place when their new home was threatened by the Frankish forces and he took charge of the defence.

If Rollo's position among the Northmen was indeed a weak one, a marriage to Charles' daughter appears rather unlikely. Of course, such a marriage would not have been unprecedented. In 882, the Viking leader Gottfrid, upon concluding a treaty with Charles the Fat, was not only ceded territory in Frisia in return for becoming a Christian, but also married to a Carolingian princess, the illegitimate daughter of Lothar II and Waldrada. 196 Yet, this same case also underlines why Dudo's account is so problematic when it comes to the marriage: Gottfrid's bride was called Gisela. It would be absolutely clear that Dudo was appropriating Gottfrid's career for his image of Rollo if the genealogy dictated by Charles to the canons of Compiègne did not mention a Gisela among his own daughters.197 While it is impossible to tell whether Dudo's account is correct or not, we have to refrain to point out that the marriage was rather unlikely.

Hence, the treaty of Saint-Clair-sur-Epte appears to have consisted of two main parts: the cession of land, centred on Rouen; and the Christianisation of the Northmen. Given that the treaty was concluded in the wake of the victory at Chartres and Rollo's apparently weak position amongst the Northmen up until the early 920s, it seems doubtful that becoming Christian should have been the only condition for the grant of land. Settling the Northmen at Rouen, controlling the lower Seine and access not only to Paris but also to the river systems leading deep into the realm, put Rollo's Northmen in a position from where they could block the river to other invaders. At the same time, strong personal bonds were created by means of baptism and the grant, which tied the Viking leader close to Charles and his supporters. There was a risk that the Northmen would break the treaty and return to their old habits, yet at the same time this risk appears to have been calculable. The battle of Chartres had proven that West Frankish forces were now sufficiently strong to defeat the Northmen while Rollo himself was too weak to disturb the balance of power within the realm. Hence, it seems plausible that protecting the realm was indeed part of treaty.

#### **V.2.3 Using an old solution in a new way**

Saint-Clair-sur-Epte was not the first treaty concluded between the Franks and the Northmen. After the death of Charles the Bald, five are recorded by the sources.

<sup>195</sup> Keller, Négotiation, 103.

<sup>196</sup> Annales Vedastini 882, 51−52; Annales Fuldenses (Mainz continuation) 883, 100.

<sup>197</sup> Witger, Genealogia, 303.

When dealing with the Northmen along the Loire, Louis III concluded an *amicitia* with the Viking leader Hasting, making them leave the realm.198 After Carloman II's disastrous campaign of 883, the nobles took over and negotiated the payment of a tribute for the withdrawal of the Northmen.199 Charles the Fat, having arrived with a large army to lift the siege of Paris, did the same, additionally opening Burgundy for them to pass the winter.200 Odo dealt with them twice, paying a tribute for their withdrawal from the realm,201 and the second time (similarly to Charles the Fat) giving them access to the Loire as well.202 These treaties reveal a remarkably consistent pattern. First of all, they were all aimed at removing the Northmen from the realm. This diplomatic strategy proved remarkably successful, yet only helped to bring short- to mid-term relief. For example, Carloman II's untimely death lead to the immediate return of the Northmen to Francia who claimed that the treaty just concluded had lost its validity with the king's death. Similarly, Odo's first treaty only lasted for a short period, for the Northmen returned to the realm after having been repulsed by the Bretons. Second, all of these treaties appear to have been concluded under very similar circumstances. When Charles the Fat bought off the Northmen, he did so because the upcoming winter season rendered the continuation of the campaign impossible. Louis III's treaty with Hasting appears to have been concluded because the king had split up his army, leaving him with insufficient forces. Similarly, Odo fell back on diplomatic solutions because he estimated himself too weak to militarily expel the Northmen. Finally, Carloman II's treaty was the consequence of his disastrous defeat. In all of these cases, the rulers had reached the limits of their military capacities, rendering them incapable of achieving their goal of removing the Northmen from the realm by means of force. Saint-Clair-sur-Epte was different to these treaties in every respect. Negotiated during a moment of military superiority, it was not aimed at making the Northmen leave, but at making them stay and protect the realm. The treaty was not without precedents. The treaty concluded between Charles the Fat and Gottfrid after the siege of Asselt appears to have been quite similar: apart from marrying into the Carolingian family, Gottfrid was also baptised and given control over Frisia.203 Yet, even this treaty was the consequence of having to abandon the siege because of supply problems and a disease outbreak in the Frankish camp.204 Hence, Saint-Clair-sur-Epte appears as a conscious decision to try a new strategy in dealing with the Viking problem, a strategy that apparently paid off for the next decade. According to Flodoard, the Northmen only broke their part of the bar-

200 Annales Vedastini 886, 62. See chapter V.1.1.

<sup>198</sup> Annales Vedastini 882, 52.

<sup>199</sup> Annales Vedastini 883, 54 and 55. See chapter V.1.1.

<sup>201</sup> Annales Vedastini 889, 67−68.

<sup>202</sup> Annales Vedastini 897, 79.

<sup>203</sup> Annales Vedastini 882, 51−52; Annales Bertiniani 882, 247−248; Regino, Chronicon 882, 119−120; Annales Fuldenses (Mainz continuation) 882, 98−99. On Gottfrid and Frisia, see also Coupland, Poachers, 108−112, and Besteman, Danish rule.

<sup>204</sup> Leyser, Warfare, 44−45.

gain in 923, and this was only because they were then incited to do so by Charles himself.205

Such a change in policy raises the question of which role of the nobles around the king played in its prevalence. Geoffrey Koziol argues that the treaty primarily reflects Charles' own policy towards the Northmen, pointing out the alliances the king concluded in 897 and 923.206 However, these two examples are extremely problematic. On both occasions, Charles was in a desperate situation, having lost most of his supporters whilst facing overpowering enemies. Only under these circumstances did he reach out to the Northmen as potential allies. As soon as he became king, the measures he took in the defence of the realm against the Northmen did not differ from those of his predecessors. Robert, on the other hand, is portrayed by Koziol as having been a resolute enemy of the Northmen and opposed to any treaty with them.207 Yet, Robert concluded a treaty extremely similar to Saint-Clair-sur-Epte in 921 with the Northmen at the Loire, granting them Brittany and the *pagus* of Nantes in return for their conversion.208 More importantly, Robert was directly involved in the proceedings at Saint-Clair-sur-Epte where he acted as Rollo's godfather209 and thus became the personification of the bond between the Northmen and the Franks. If Robert had indeed opposed the treaty, Charles would certainly have taken this position himself—he had done this already earlier in the case of Hundeus.

Yet, Hundeus' case points to some other changes in the attitude of leading nobles towards the Northmen. Archbishop Fulk had harshly condemned Charles'


alliance with the Northmen,210 seemingly opposing any dealings with "the enemies of God." However, after Fulk's death his successor Heriveus appears to have pursued a programme of converting the Northmen, seemingly even preceding the treaty of Saint-Clair-sur-Epte, as is apparent from Flodoard's *Historia Remensis Ecclesiae*: "[Heriveus] vigorously laboured towards the appeasement and the conversion of the Northmen, until finally, after the battle which Robert fought against them at Chartres, they started to receive the Christian faith…"211 It seems that by the early 900s a strong lobby had developed within the circle of the most important nobles around the king who now supported the very ideas the treaty of Saint-Clair-sur-Epte stood for. Within the context of the possibilities and limits of royal power, Charles' options for dealing with the Viking threat had increased in comparison to his predecessors.

#### **V.3 Conclusion**

When dealing with the Vikings, the cooperation of the nobles played a central role for the rulers, although in military matters it was far from decisive. Carloman's and Odo's victories were won by small armies while Carloman's second campaign ended in disaster despite him now being supported by Hugh the Abbot; and Charles the Fat, although he probably had the largest army of this time at his disposal, chose to negotiate instead of fighting. In any case, the results of seeking direct confrontation were mostly limited when it came to securing the realm against the Northmen—even Louis III's victory of Saucourt only lead to a temporary shift of Viking activity to Lotharingia. One way to deal with this problem was a defensive strategy aimed at containing the Northmen in a specific area, as practised by Odo. Here however, the support and cooperation of the nobles became indispensable, as Odo's attempts failed when Abbot Rodulf opened private negotiations. When it came down to making the Northmen actually leave the realm, negotiations with the Northmen proved to be far more successful than the use of force. Yet, apart from Odo, the rulers appear to have reverted back to this strategy only under specific circumstances. Hasting, with whom Louis III concluded an *amicitia*, operated along the Loire, which was apparently only considered a secondary theatre of war by the king. Carloman's treaty with the Northmen was the result of the nobles taking over after a disastrous campaign. Charles the


Fat may have chosen to pay tribute to the Northmen because he probably would have run into severe problems in supplying his army. Odo, on the other hand, acting mainly defensively, seems to have considered negotiations an important part of his strategy to protect the realm from further incursions. In general, the rulers appear to have enjoyed the support of the nobles when negotiating—or, as Carloman's example demonstrates, may have even been forced by them to enter into them. However, Charles the Simple renouncing Odo's cautious strategy and returning to the more aggressive approach, as previously shown by his brothers and Charles the Fat, was supported by the leading nobles of his realm, among them also Odo's brother Robert. It seems that there were limits on how far the nobles were willing to accept negotiations as an essential part of the measures taken against the Northmen and that these limits had been reached under Odo. The cooperation of these nobles with Charles also points to another observation: similar to his predecessors, Charles the Simple was able to coordinate large-scale operations. That Charles was not in military command of at least the campaign in 911 did not mean that these men acted on their own accord, but that Charles had delegated the task of defending the realm to them, similar to Gauzlin under Louis III and Carloman or Henry under Charles the Fat. It was also due to their support that Charles was able to implement a distinctive change in strategy. The treaty of Saint-Clair-sur-Epte, negotiated from a position of strength in the wake of a major military victory, was no longer aimed at having the Northmen leave the realm, but at their settlement and conversion.

## **VI. Conflicts, rebellions and the role oftrust**

Over the past decades, the way conflict situations are interpreted has significantly changed. Conflicts are no longer read as a sign of the weakness of the medieval state, but acknowledged as an important part of the existing social order and a common experience.1 For example, Charlemagne's reign was, as Janet Nelson phrases it, "one goddamned crisis after another."2 In fact, conflicts between noble factions competing with each other or the king was one of the driving forces of politics throughout the entire Carolingian period.3 This change in reading is directly linked to the importance now attributed to social ties. Within conflicts, the opponents were usually not strangers to one another, but, as members of the same noble society, connected to each other by a complex system of social relations, which made solving the conflicts much more a question of social interaction than of legal dispute.4 Equally important, both parties were also part of larger social networks, networks they could manipulate to their own benefit "by mobilising groups that could be formed on the basis of kinship, lordship, common residence, association with a cult centre […] or other social ties"5 since these ties were often considered more important than the question of whether a claim was legally justified or not.6

The first section of this chapter focuses on two ways in which these social ties determined the possibilities and limits of royal power in conflict situations, analysing the way they influenced the rulers' strategies of bringing the respective conflicts to an end. Firstly, it is worth emphasising again that only a very limited number of nobles had direct access to the ruler.7 These nobles constituted the links between the ruler and his opponents, thus enabling communication between the opposing parties at any given time. Furthermore, they participated in the decisionmaking process at court,8 meaning that they were able to directly influence the way the king addressed the conflict in question. Secondly, since the studies of Gerd Althoff, the existence of certain rules, or rather norms,9 in the dealings between king and nobles has been generally accepted by scholars.10 These norms, although never written down but commonly only referred to in very general terms, constituted a framework when it came to deciding upon actions.11 Thus, in conflicts,


these norms constituted a common ground which allowed the opposing parties to come to terms with each other.12 In the interpretation of these norms there certainly was room for manoeuvre,13 yet it was common for both kings and nobles to overstep these boundaries,14 which, in turn, led to an atmosphere of mistrust.15

In modern research, trust as a concept and the role it played in medieval rule have received surprisingly little attention, and only recently German scholars have paid it particular attention.16 Trust, or rather, interpersonal trust17 (given that Carolingian rule was primarily based on relations between individuals), is defined as an expectation concerning the future behaviour of another person. This expectation serves as a basis for one's own decisions in regard to actions which may be influenced by the other person, taking into consideration the risk that the other may indeed act differently but assessing this risk as minor.18 Thus, when deciding about future actions, trust narrows down one's options by providing a guideline which


helps to predict the behaviour of other people.19 According to Jan Hirschbiegel, interpersonal trust depends on four prerequisites which cover the different layers between an individual and the community in which it is embedded: first, a general readiness to trust someone else; second, a specific readiness to trust a person which is based on one's experiences with this person and their reputation; third, the so-called structural trust, that is to say the framework provided by the social groups the individuals are part of; and fourth, systemic trust, which is defined by the social, cultural, religious, political or otherwise impersonal frameworks characterising society as a whole.20

In regard to Carolingian rule and our own study, the second point is of particular interest since it refers to a certain dynamic in the development and loss of trust, a dynamic which could be influenced by the ruler's actions. These actions do not only have an impact on the degree of trust between the ruler and the particular noble involved, but also on the way he was perceived, that is to say how trustworthy he was considered to be by other nobles.21 Actions, even if they were only directed at a single noble, could damage the trust between the ruler and wider circles of the nobility and thus lead to a crisis of trust, potentially threatening his rule in general. Scholars have identified two dimensions which determine how an individual's trustworthiness is perceived: first, their competence or reliability in performing a certain task or action in an appropriate manner and second, their integrity, honesty and commitment to take other parties' interests into consideration.22 In the first chapter we argued that the rebellion against Odo in 892 was in part due to a crisis of trust. One of the examples brought forward was the case of Baldwin, who refused Odo's offer regarding the return of his deceased cousin Rodulf 's abbeys when the king requested him come to court and rely on the royal generosity. This refusal aptly illustrates the nature of this crisis: it was rooted in doubts about Odo's integrity as a king. Yet, there is also an example of a crisis of trust caused by lack of competence: when the nobles took over from Carloman II in late 883, this was the consequence of his disastrous campaign against the Northmen in the Vimeux23—Carloman was seemingly still trusted as far as his behaviour conformed to societal norms; yet, he was not trusted when it came to leading against the threat posed by the Northmen.

This brings us to the following sections of this chapter, which revolve around the role of trust in the case of rebellion. While conflicts usually developed over changes within the hierarchy of nobles, often revolving around the distribution of


<sup>19</sup> Luhmann, Vertrauen, 23−24; Marzano, Confiance, 53−54 and Weltecke, Gab es Vertrauen, 73. Against the pure rational choice approach pursued, for example, by Harding, Trust, Marzano rightly emphasises the non-rational aspects of trust such as the unconditional trust of children.

<sup>20</sup> Hirschbiegel, Nahbeziehungen, 60−61.

<sup>21</sup> On third-party effects of trust relations, see Burt and Kneez, Kinds and Hardin, Trust and Trustworthiness, 139−142.

*honores*, questions of rank and—directly connected with these two—political influence at the royal court,24 the parties involved accepted the ruler's final authority. However, this changed in Odo's case, when Fulk and his allies crowned Charles the Simple king. They questioned the former's right to rule and attempted to replace him with another king—in other words, they rebelled.25 Régine Le Jan has described these rebellions as signs of a crisis of royal power, ascribing them to the reaction of the noble elites to royal policies attempting to interfere with what they perceived as their natural privileges.26 While there certainly was a crisis of royal power, at least in the case of the rebellion against Odo's there appear to have been other reasons too—not least a lack of trust in him. In addition to the rebellion against Odo, Western Francia, from the death of Charles the Bald to the end of the reign of Charles the Simple, saw two more such attempts to replace the ruler, with Boso being crowned in opposition to Louis III and Carloman II; as well as Robert of Neustria's rebellion against Charles the Simple.27

In the second section of this chapter, conflict situations are analysed in regard to the kings' actions and whether these actions may have contributed to a loss of trust in the respective rulers by wider circles of the nobility. However, royal actions could not only damage a ruler's reputation but also help in building trust. This could be done in a number of ways: creating kinship was one means, concluding an *amicitia* another,28 and symbolic actions could also be used in such ways.29 The third section of this chapter deals with a special type of these symbolic actions which are closely tied to conflict situations: rulers aiming to do justice in various ways by making use of capitularies, judicial assemblies and restorational acts. Doing justice allowed the king and the magnates to cooperate and thus to strengthen the bond tying them together since justice was spoken in consensus.30 By doing so, they legitimised the existing social order as well as the position of the various protagonists within the hierarchy.31 This last point emphasises the inherent symbolical power of pronouncing justice. The king, receiving his realm from God, was responsible for its welfare and order which made him the supreme judge. The ideal of ruling was identical with doing justice, that is to say, respecting the law, being a judge and fighting injustice.32 Thus, doing justice could be used as a tool to legitimate rule and a means to restore lost trust in the king. Finally, the last section


deals with the rebellions breaking out against Louis the Stammerer's sons in 879 and against Charles the Simple in 922, analysing them in regard to the role trust in the king played in their emergence.

#### **VI.1 Solving conflicts: The role of interest groups**

A perfect example of how nobles present at the royal court could influence the outcome of conflicts between nobles and the king is provided by the case of Count Gauzfrid and Louis the Stammerer. Probably early in 878, Count Gauzfrid's sons seized the stronghold and *honores* of the deceased Count Odo from his son. In response, Hugh the Abbot requested Louis the Stammerer's aid against them.33 The conflict appears to have been quickly resolved. "Some of the royal counsellors who were friends of [Gauzfrid]"34 mediated between him and the king. Gauzfrid and his sons presented themselves at the royal court, returned the stronghold and *honores* and received both of them back by a royal grant. In return, Gauzfrid brought a group of Bretons into the service of the king. A number of observations can be made here. First, the problem leading to the conflict between Gauzfrid and the king was not actually his usurpation of property, but the fact that he had circumvented the royal prerogative of controlling the distribution of *honores*. Once Gauzfrid acknowledged the position of the king, the conflict could be resolved. Second, the crucial role of Gauzfrid's connections at court can be discerned. Karl Ferdinand Werner was undoubtedly correct in pointing out that one of these connections was Gauzfrid's brother, Gauzlin. 35 Seemingly, Gauzlin's influence at this point of his career was strong enough to outweigh any opposition presented at court by the connections of Count Odo's son, which made the entire matter a question of acknowledging Louis' own interest, something that was acceptable for Gauzfrid.

The same mechanism, namely the influence of a powerful group of nobles at court, can also be discerned in another conflict during Louis the Stammerer's reign, although with the opposite effect. Early in his reign, Bernard of Gothia and his brother Emeno became embroiled in conflict with the king, the former preventing Archbishop Frotar of Bourges from entering his city36 and the latter seizing Évreux and devastating the surrounding country.37 Both were excommunicated at

<sup>33</sup> Annales Bertiniani 878, 222. On Count Odo, see Levillain, Essai, 174−177. On this request, see also chapter II.1 and V.1.1.

<sup>34</sup> Annales Bertiniani 878, 222: *Sed miserante Domino aliquantulum [Hludouuicus rex] conualescens, satagentibus quibusdam consiliariis suis et amicis Gozfridi, uenit ad eum isdem Gozfridus, adducens secum filios suos, ea conditione ut castellum et honores quos inuaserant Hludouuico regi redderent et postea per concessionem illius haberent. Tunc Gozfridus partem de Brittonibus ad regis fidelitatem conuertit.*

<sup>35</sup> Flodoard, HRE III, c. 24, 324. Werner, Gauzlin, 417.

<sup>36</sup> MGH Epist. VII, N° 149, 126. On this conflict see also Oexle, Bischof, 202−204.

<sup>37</sup> Annales Bertiniani 878, 222.

the synod of Troyes,38 and Bernard at least lost his *honores*, which were distributed amongst Louis the Stammerer's supporters;39 he also became the target of a military campaign organised by the king.40 In 880, after Louis' death, the *Annales Fuldenses* note Bernard's submission,41 which happens to be also his last appearance in the sources. Apparently, the conflict was resolved by means of force. If negotiations did take place, they failed.

All this is surprising since it seems that the highest circles responded with sympathy to Bernard's cause and this support could have been used to come to terms with the king, as in Gauzfrid's case.42 Frotar's own position at Bourges was a rather difficult one. Starting off as bishop of Bordeaux, Frotar had abandoned his see in flight from the Northmen, first transferring to Poitiers, then to Bourges.43 These transfers had been highly disputed. Frotar found a backer in Charles the Bald44 and Pope John, 45 yet his case was still open in 878 when it was discussed again at the synod of Troyes.46 Bernard could use this dispute as an argument for his own case and, indeed, he defended his actions by claiming that he wanted to prevent Frotar from opening Bourges to the enemies of the king.47 Pope John VIII, while certainly not agreeing with Bernard's measures, appears at least not to have been biased against him. In a letter from June 878, he did ask Bernard to keep his vassals from hindering Frotar and to wait for the judgement of the synod; yet he addressed him as "[our] beloved son Bernard, most noble *marchio,*" thus indicating that Bernard still enjoyed his favour.48 Only afterwards did John change his position and request Bernard several times to present himself with some of his men to be judged at the synod.49

In contrast to the pope, Hincmar appears to have remained sympathetic to Bernard's cause until the very end. The archbishop appears to have despised Frotar,


who had supported the primacy of Archbishop Ansegisus of Sens against his own claim, and denounced him as a lickspittle.50 For him, Frotar was the Archbishop of Bordeaux, not of Bourges.51 His opposition to Frotar may be the reason why he seems to have taken a favourable position towards Bernard. He certainly did not actively support him for a letter to Gauzlin clearly shows that he did in fact want Bernard's endeavour to end.52 Yet it is remarkable that he omits Bernard's excommunication from his account at the proceedings at Troyes, noting only the distribution of his *honores* during a meeting between Louis and the group around Hugh the Abbot.53

Like Gauzfrid, Bernard's and Emeno's most important connection at court was Archchancellor Gauzlin, their uncle.54 In his letter, Hincmar urged Gauzlin to influence Bernard to cease his efforts,55 which indicates the existence of a closer relation between the two. Probably not long after, Gauzlin and Bernard appear to have acted together in Burgundy where they are recorded as having visited the abbey of Saissy together.56 The failure of this connection to secure a settlement with the king was undoubtedly directly related to Gauzlin's rivalry with the group around Hugh the Abbot which managed to dominate the circle around the king approximately at the same time as Bernard's and Emeno's excommunication.57 This conclusion is confirmed by Hincmar's comments on the matter, which indicate that Emeno was excommunicated at the instigation of certain bishops and with the king's backing,58 which is certainly a reference to Hugh and his allies. The rivalry between the two groups and the outcome of the conflict reveal the mechanisms active at the royal court. Louis acted in concord with a group of nobles powerful enough to outweigh Bernard's support. Louis himself might have accepted a settlement with Bernard—in a diploma for the church of Barcelona issued after


Bernard's excommunication he is still depicted as "our *marchio*"59—yet this group undoubtedly blocked any attempt made in this direction. This would also explain why Bernard ignored the summonings to the synod of Troyes: faced with the growing opposition against him, he probably feared his was an already lost cause.

Powerful interest groups at court which blocked the resolution of a conflict can also be made out under Charles the Simple. According to the *Annales Vedastini*, early in 899 Count Baldwin II of Flanders invaded Péronne "against the will of the king."60 The wording chosen here seems to indicate that before Baldwin had taken up arms, he had claimed at least this town at court, yet been refused by Charles. The fate of Péronne appears to have been closely related to that of Saint-Quentin. Again according to the *Annales Vedastini*, Saint-Quentin had belonged to the son of one of Odo's closest supporters, Count Theoderic of Vermandois, before it was captured by Baldwin's brother Rodulf. 61 Odo had reacted to this capture by planning an attack not only on Saint-Quentin but also on Péronne,62 a circumstance that led Karl Ferdinand Werner to the assumption that the latter had also belonged to Theoderic's son.63 If this assumption is correct, this would explain why Charles refused Baldwin's request. At some moment after Odo had taken Saint-Quentin, it ended up in the hands of Count Heribert I of Vermandois,64 either after the latter had abandoned Charles and submitted to Odo or after Charles had become sole king. Given that it seems most likely that Baldwin would have renewed his claim when the *honores* were about to be redistributed, it does appear probable that it was Charles who granted them to his close supporter Heribert. However, Baldwin and Heribert had already been enemies before, possibly because of Rodulf 's death at Heribert's hands. When the Frankish nobles assembled after Odo's death to put Charles back on the throne, Baldwin abstained from appearing in person, sending only a legation, according to the *Annales Vedastini* because of Heribert's presence at the royal court.65 Charles granting Heribert the *honores* was undoubtedly the consequence of their close relationship66 and served to secure royal control in the area. At the same time however, this same relation also blocked Charles from accommodating Baldwin if he did not want to estrange Heribert.

This relation probably also explains the aggressiveness of Charles' strategy against Baldwin. Saint-Vaast was quickly besieged and Baldwin's men defending the *castrum*, after having been excommunicated, were forced to give up the

59 DLS 17 (9th September 878, Troyes), 55: *…Bernardus marchio nostro..*.

60 Annales Vedastini 899, 81: *Balduinus vero contra regis voluntatem Perronam invasit, sed sub celeritate amisit.*


abbey.67 Baldwin, in turn, appears to have attempted to strengthen his position by seeking help outside the realm from Zwentibold, who, after Charles' campaign into Lotharingia during the previous year, had only concluded an armistice with the king and was likely to need allies if the hostilities were renewed.68 This connection would explain why Baldwin was present at Cambrai when Charles and Zwentibold met a little after the fall of Saint-Vaast. However, if Baldwin had expected actual support from Zwentibold he was to be disappointed. Zwentibold's own political problems inside Lotharingia encouraged him to conclude a peace treaty with Charles and Baldwin was consequently forced to submit to the king, reconcile with Heribert, give up Saint-Vaast69 and return further property to the church of Noyon.70

This defeat, however, did not prevent Baldwin from renewing his efforts. During the following year he approached Charles to try to regain his lost *honores*. Yet, this time he was opposed not only by Heribert, but also by Archbishop Fulk, 71 hence by two of Charles' most important supporters. After Baldwin's submission at Cambrai, Charles had granted Saint-Vaast to the archbishop, who, in turn, had traded it with Count Altmar for Saint-Medard of Soissons.72 Fulk was, nonetheless, undoubtedly concerned by Baldwin's approach: should Charles return Saint-Vaast to the count, his deal with Altmar would undoubtedly be revoked. While Charles had tightened his control over the north-east of the realm by giving the abbey to one of his most trusted advisors, he had also strengthened the opposition against Baldwin at his court and thus he had limited his options for coming to a lasting agreement with the count even more. The circumstance that Baldwin, despite facing such powerful opposition, nevertheless made this attempt, points to an important factor in his relations to the king. It seems unlikely that he would have made the effort to approach Charles if he had not imagined the chance of having his *honores* returned to him, especially since he had already made a similar attempt during the previous year. His chances of achieving a positive outcome set aside, his attempts, nevertheless, indicate that his relation with the king was better than the ongoing conflict might otherwise convey. Despite having lost Saint-Vaast in the previous encounter, Baldwin hence still appears to have been confident that


there was a chance that Charles would act in his best interest—in other words, he trusted the king.

However, Charles' second refusal may have overstrained this trust, since only a couple of days later Baldwin's men murdered Fulk. 73 Surprisingly, this murder did not result in an escalation of the conflict, but was followed by a settlement between Charles and the count. Only the actual murderers were excommunicated by a synod under Fulk's successor Heriveus74 while Baldwin himself received Saint-Bertin, an abbey he had previously claimed for himself,75 from the hands of the king.76 This complete change in policy can be explained by a number of factors. Fulk's death had weakened the opposition against Baldwin at court and provided Charles with Saint-Bertin, an abbey that Baldwin had also attempted to bring under his control. Furthermore, Fulk's murder led to political turmoil and the crisis with Robert of Neustria.77 Confronted with Robert, Charles could probably not afford a conflict with Baldwin; at the same time, he needed to grow his basis of support in order to stabilise his rule. Therefore, compensating Baldwin was a necessity, the logical consequence of the political circumstances, and reveals how changes in the circle around the king could open up new options. However, the settlement should not be read as Charles finally yielding to Baldwin. While the count did receive Saint-Bertin, he had to give up his claim on Saint-Vaast, which remained in the hands of Count Altmar. 78

Whereas the conflict between Charles and Baldwin was again marked by the presence of strong interest groups at the court blocking an earlier agreement, its prehistory, the conflict between Baldwin and Odo, demonstrates the importance of trust between king and noble as a determining factor. After the death of Abbot Rodulf of Saint-Vaast and Saint-Bertin, the monks of the Saint-Vaast had sent Count Ecfrid to the king to enquire after his will. However, only some days later, at the instigation of a certain Everbert, instead of waiting for Odo's response, they invited Rodulf 's cousin Baldwin to take over their abbey,79 thus pre-empting the royal will. This, and not the actual control of the abbey, now became the central issue. After having arrived at Saint-Vaast, Baldwin turned to the king, asking for his consent, not only to hold Saint-Vaast, but also Rodulf 's other abbey, Saint-Bertin. 80



<sup>75</sup> See below.

According to the *Annales Vedastini*, Odo responded "that first [Baldwin] should let [Odo] assume power over his property that God had given to him and come to him, confident to find him in benign mood."81 This phrasing seems to indicate that Odo, while insisting on his royal prerogative of controlling the distribution of *honores*, was nevertheless willing to acknowledge Baldwin's claim and proposed a solution that strongly resembles the settlement between Louis and Gauzfrid. 82 However, Baldwin refused this proposal and the king's subsequent attempts to negotiate a settlement failed,83 with the count's interest in acknowledging Odo's prerogative seemingly having been very limited. According to the annals, Baldwin's legation to Odo was only the result of Everbert's counsel while, after Odo had refused him, all further attempts to negotiate appear to have originated at the royal court.84

As we have argued, Baldwin's position appears to point to a deeper problem with Odo's rule at this moment, namely the existence of a strong sentiment of mistrust against the king within the nobility.85 Baldwin's refusal to come to court, to give up control over Saint-Vaast and to rely on the royal generosity indicates that he did not trust Odo to protect his interests, that is to say, to acknowledge his claim over his cousin's *honores*. This view was probably reinforced by developments at Rodulf 's other abbey, Saint-Bertin. While the monks of Saint-Vaast had opted for Baldwin, those at Saint-Bertin turned to Odo upon hearing that the count had claimed their abbey in order to express their opposition. Supported by Archbishop Fulk, they obtained the right to elect their new abbot, with their choice falling on Rodulf 's predecessor—the very same Fulk.86 Meanwhile, Baldwin's position at Saint-Vaast was probably also not as strong as he would have liked it to be. As the initial legation under Count Ecfrid shows, there was a party among the monks of Saint-Vaast whose loyalty towards Baldwin was questionable and who instead favoured a royal decision on the matter, thus providing an opening for Odo, should he wish to deny Baldwin's claim. If Baldwin had already harboured doubts about Odo's intentions, the development at Saint-Bertin had certainly not improved his hopes that, should he give up control over Saint-Vaast, the king would grant him the abbey.

Probably around the same time as the conflict for Rodulf 's abbeys was developing, Odo appears to have attempted to install his ally Altmar as count of Poitiers, which led to a confrontation with the Ramnulfides and thus diverted a part

<sup>81</sup> Annales Vedastini 892, 71: *Odo vero rex respondit, ut sineret illum prius esse potestativum de suo, quod Deus illi concessit, et veniret ad se, fidens benignum erga illum se inventurum fore.*

<sup>82</sup> Another example for this practice is the restitution of estranged possessions to the pope by Wido of Spoleto in 882 at an assembly at Ravenna under Charles the Fat, only to receive them back immediately after. Dümmler, Geschichte III, 186 and 218−219.

<sup>83</sup> Annales Vedastini 892, 71.

<sup>84</sup> Annales Vedastini 892, 71.

<sup>85</sup> See chapter I.2.3.

<sup>86</sup> Folcwin, Gesta Abbatum S. Bertini Sithensium, c. 98, 624.

of the royal attention away from Baldwin.87 In addition, Baldwin was able to win further support from one of Odo's own relatives, his uncle Adalelm's son Walker, who the king had entrusted with the *castrum* of Laon.88 Hence, for Baldwin, taking up arms against the king seemed to be the preferable option, especially if he considered Odo to be unreliable in his promises.

Once Baldwin started to act hostile towards Odo, the king initiated a number of measures against the count. The first half of these measures consisted in creating new channels of communication which not only served to continue to negotiate, but also to increase moral pressure on Baldwin through the threat of excommunication. This task was performed by Archbishop Fulk, who, together with a number of his suffragans, advised Bishop Dodilo of Cambrai to get in contact with the count89 and, after this attempt as well as a letter of himself on the same subject,90 had failed, organised a synod where Baldwin's excommunication was decided upon.91 Fulk's involvement was certainly due to Baldwin falling under his spiritual jurisdiction as archbishop of Reims. Yet, there was probably more to it. Fulk and Baldwin had been allied against Odo in 88892 and appear to have maintained this connection over the course of the years—when Fulk himself rebelled against Odo shortly after, this alliance was renewed. Odo might have aimed to use this connection to mediate an agreement with Baldwin while hoping that their rivalry for Saint-Bertin would prevent the archbishop from siding with Baldwin. In any case, Odo underestimated Fulk's own political agenda: the archbishop suspended Baldwin's excommunication,93 thus easing the pressure placed on the count. Odo's second measure aimed at isolating Baldwin from his allies. Laon was taken and Walker captured and executed.94 After this was done, Odo made an attempt to attack Baldwin directly in Flanders, yet was outmanoeuvred and forced to withdraw.95 The further development of the conflict was determined by the creation of new alliances and their dissolution, which strongly influenced the balance of power. Fulk's rebellion drew Odo's focus away from him and enabled Baldwin and his brother Rodulf to go on the offensive and take Saint-Quentin96 when Odo was forced back over the Seine by Zwentibold's intervention.97 However, when the alliance broke and Charles' supporters opened negotiations with Odo, the latter

87 See below.


immediately turned back against Baldwin and besieged Saint-Vaast. 98 The attack was accompanied by the renewal of negotiations, this time with Odo's brother, Robert, serving as mediator.99

These events cast some further light on the relations between Odo and Baldwin. When the king arrived at Saint-Vaast, Baldwin's men, who garrisoned the fort and the abbey, asked for peace and offered hostages while sending envoys to the count to ask for his orders. In the meantime, since the return of the envoys took more time than anticipated, Odo asked the gates to be opened and occupied both places. When Baldwin's envoys finally arrived, accompanied by Robert, the king ordered his men to leave the fort and return it to the count's men.100 Commonly this is interpreted as the conclusion of peace, with an agreement over Baldwin's possession of Saint-Vaast. 101 Yet, no such peace is actually recorded and Baldwin's submission only appears to have taken place in 897.102 In addition, Odo's next plans were to attack Baldwin's brother Rodulf at Saint-Quentin and Péronne.103 It is of course possible that Odo and Baldwin had made their peace at that moment and that Rodulf acted independently of his brother.104 Nevertheless, we may propose a different reading: Odo had occupied Saint-Vaast on the condition that he would leave it once Baldwin's will was known. It was a question of trust to open the gates to the king, who then went to pray and listened to a mass at the tomb of Saint Vedast. Considering Baldwin's mistrust of Odo's behaviour, the king's vacation of Saint-Vaast would not have been the consequence of a peace with Baldwin, but a trust-building measure, to demonstrate his integrity. Handing Saint-Vaast back to Baldwin's men would then not have meant that Odo accepted the count's claim but that he now pursued a strategy of re-establishing trust in his person.

However, making peace was complicated by Baldwin and his brother Rodulf having occupied *honores* belonging to Odo's supporters, most notably


<sup>98</sup> Annales Vedastini 895, 76.

<sup>99</sup> Annales Vedastini 895, 77. See also chapter II.4.

Saint-Quentin, *honores* they now had to give up.105 Rodulf in particular appears to have been opposed to this, which led the negotiations to fail.106 Only when Odo had taken Saint-Quentin and Péronne from Rodulf 's men and Rodulf himself had been killed by Heribert I, 107 was a final accord between the king and Baldwin mediated by Robert. 108 The solution they agreed upon appears to have been in direct response to the initial problems which had led to the outbreak of the conflict. Baldwin came to the court—something he had previously avoided doing—and was honourably received. Then he "satisfied the king in everything he ordered him to do,"109 which is undoubtedly a reference to returning Saint-Vaast to the king, thus acknowledging Odo's royal prerogative to control the distribution of the *honores*. The *Annales Vedastini* do not mention that Odo then granted the abbey to Baldwin, yet from their report for 899 it becomes clear that the count remained in possession.110 Odo's success in asserting his interests was certainly due to his military superiority after Baldwin had isolated himself from his allies and Odo had come to terms with Charles and his supporters. Furthermore, Rodulf 's death had solved the problem of another party blocking a resolution due to their own interests. Finally however, a crucial role fell to Robert's mediation. He represented a new communication channel trusted by both parties and was seemingly able to restore Baldwin's trust in the king. Trust, like the presence of interest groups at the royal court, hence played a decisive role in the capacities of a ruler to resolve a conflict by means of negotiation.

#### **VI.2 Rulers and nobles: Breaches of trust**

In all the cases presented so far, the kings do seem to have had limited direct interests in the conflicts. If any such interests are visible at all, they appear to have been focused on maintaining control over the distribution of the *honores* in question. At the root of these conflicts were the efforts of individual nobles to extend their influence which thus infringed on this royal prerogative or clashed with the interests of powerful noble groups at the royal court. If peaceful resolutions were blocked, this was either due to the influence of these groups or due to a lack of trust in the king by the opposing party. However, there are also those cases to consider in which the initiative appears to have rested mainly with the ruler him-

<sup>105</sup> See chapter I.2.5.

<sup>106</sup> Annales Vedastini 896, 77: *Odo rex placitum cum suis fidelibus habuit, volens partem regni, quam eius fideles tenuerant, Karolo concedere. Sed Rodulfus comes omne illud placitum disrupit.*

<sup>107</sup> Annales Vedastini 896, 78.

<sup>108</sup> Annales Vedastini 897, 79: *Balduinus etiam Rothberto faciente venit ad regem; quem rex honorifice suscepit, et de omnibus quae iusserat illi rex satis illi fecit, et ita remisit eum rex ad sua.*

<sup>109</sup> See chapter VI.1, n. 108.

<sup>110</sup> Annales Vedastini 899, 81: *Karolus rex obsedit castrum Sancti Vedasti, et habitatoribus excommunicatis omnibus, hi qui ipsum castrum tenebant missos dirigunt ad Balduinum et, quamvis non voluntarie, obsides regi dederunt, dato spatio ad sua exportanda.*

self, be it because he wanted to pursue his own interests or because he was under the impression that the circumstances forced him to intervene with the current order. One of these conflicts arose over the control of the county of Poitiers after the death of Ramnulf II on 5th August 890.111 Ramnulf had left an illegitimate son, Ebalus Manzer, 112 as his heir.113 The latter's claim appears to have been acknowledged by Odo since Ebalus was able to donate property to Saint-Martin of Tours for the sake of his father's, his uncles' and his own soul, immediately receiving said property back in *precarium* along with the *villa* of Doussay in the *pagus* of Poitiers and another *vicaria* on 10th October 891,114 an exchange that hardly could have taken place without the king's consent even if he does not appear in the charter issued on that occasion. The matter, however, was complicated by the claim of Count Altmar, whose father Emenon had previously held the city.115 According to Ademar of Chabannes, Odo did indeed grant Poitiers to Altmar after Ramnulf 's death.116 However, given that Ebalus Manzer and Odo were still on good terms in October 891, it seems that this did not happen immediately after Ramnulf 's death, but probably in late 891 or early 892. Since in the charter recording his donation to Saint-Martin of Tours Ebalus only describes himself as "Ebalus*,* as yet blossoming in youthful age" without giving any title,117 it would indeed appear that the countship had remained vacant from Ramnulf 's death up until Altmar's appointment as count.

However, the succession at Poitiers cut deep into the structures of Odo's rule. Altmar was related to Odo and was one of his close associates.118 Hence, the grant can be seen as a sign of royal favour resulting from this close relation. Yet, at the same time Ebalus Manzer himself was also very well connected at court, with his uncle Ebolus119 being the abbot of Saint-Denis and Odo's archchancellor.120 Thus, satisfying Altmar's claim meant that Odo estranged another one of his close supporters since Ebolus and his brother Gauzbert now entered into conflict with the king.121 This act probably damaged Odo's reputation as a ruler who could be


trusted by demonstrating that proximity to the king was not sufficient to protect one's interests. The situation became even more complicated when William the Pious, another one of Ramnulf 's relatives with whom he had entertained a close relation,122 was also drawn into the ensuing conflict. According to Ademar of Chabannes, upon his death Ramnulf had entrusted his son Ebalus to William's close friend Gerald of Aurillac, who, in turn, brought him to William. 123 It may have been due to this opposition including some of the most important Aquitanian nobles that Odo revoked his previous decision to grant Poitiers to Count Altmar and instead entrusted his brother Robert with the city's defence.124 The wording used by Abbo of Saint-Germain, who is the only source reporting about this event, may indicate that Robert was not made count of Poitiers but only temporarily installed up until a settlement between the different parties could be reached. Be it as it may, this new decision in turn led to a conflict with Altmar, who attacked Odo's forces and then rushed to take Poitiers.125 Within a short period of time, Odo had thus estranged not only one, but two of his close supporters. If Odo's previous decision to disinherit Ebalus Manzer had already been perceived as a breach of trust, this one probably deepened the mistrust against him even more.

Odo's reaction to the opposition appears to have been delayed by the evolving conflict with Baldwin II of Flanders, which kept him in Francia up until autumn 892 before he engaged in a campaign to manage the problems south of the Loire.126 The sources make no mention of negotiations, yet, it may be that, at least in William's case, Odo had hoped for a peaceful settlement. According to Abbo of Saint-Germain, the king only deprived William of his *honores* after their armies had met at the opposing shores of a river without engaging,127 possibly indicating that up until this moment he had intended to come to terms with William yet had been rejected by the count. While Ebolus and Gauzbert were quickly killed,128 the


conflict with William continued, the count, in turn, killing Odo's ally Hugh, who had been granted William's *honores*. 129

However, over the course of 893, a settlement was found not only with William but also with Altmar. This was probably due to two political developments which shifted the balance of power against Odo. First, early in 893 Fulk and his allies crowned Charles the Simple as king, thereby directly threatening Odo's right to rule. Second, in the wake of this rebellion, William and Altmar concluded an alliance, joining up with Richard the Justiciar to take advantage of the situation.130 Given that Altmar had taken control of Poitiers (which was also claimed by Ebalus Manzer), this alliance may seem surprising since it indicates that William had acknowledged Altmar's position and, hence, disregarded his protégé's interests. Undoubtedly, this had been made possible by the deaths of Abbot Ebolus and Gauzbert and might have been facilitated if Altmar, in turn, guaranteed Ebalus access to his father's property in the region. Thus, political pressure on Odo increased considerably and the king decided to appease his opponents. William gained the abbey of Saint-Julien of Brioude by a royal grant131 whereas Altmar was again confirmed as count of Poitiers132 and resumed his position within the circle around Odo shortly after,133 acts that served to strengthen the bond between these nobles and the king and that probably were also aimed at restoring trust in Odo because they demonstrated that he was willing to do his opponents justice and to make generous grants to restore peace. While Altmar regained his position in the circle around the king, William's future position is more difficult to discern. It seems that he at the very least remained neutral over the course of the next years, even if he displayed a rather self-confident attitude.134 The alliance between him and Altmar probably broke soon after. The *Vita Geraldi* reports a number of clashes between the count and Gerald of Aurillac, with William coming to the latter's aid;135 while in 902 William probably supported Ebalus Manzer's return to Poitiers which forced Altmar out of the city.136 However, for the time being Ebalus Manzer's interests appear to have once again been disregarded, undoubtedly because the deaths of his uncles had deprived him of his support. In a private charter issued in 897, Robert granted the *villa* of Doussay to the canons of Saint-Martin

129 Abbo, Bella II, 108, v. 551−561.


of Tours,137 the very same *villa* that had been given in *precarium* to Ebalus Manzer in 891. In the conflict evolving around the succession at Poitiers, Odo's early decisions appear to have been motivated firstly by wanting to protect the interests of a noble close to him and secondly by trying to reverse his mistake. Both times he violated the interests of members belonging to his inner circle and therefore betrayed the trust placed in him as ruler that being close to him meant that he would act according to their interests. These actions were probably as badly received by them as by the rest of the nobility and thus contributed to the general loss of trust that was part of the rationale of the rebellion against him.

This same failure to take into consideration how his actions were perceived by the nobles also marks the behaviour of Charles the Simple in his conflict with Count Gislebert. The first sign of this conflict is the diploma recording the second assembly of Herstal in June 919, which notes a judgement of the *scabini* of the royal palace in favour of Archbishop Roger and which returned the abbey of Saint-Servais of Maastricht, "previously unjustly taken from this church by means of violence by Count Reginar and his son Gislebert" to the church of Trier.138 At this point, the conflict around Saint-Servais already had a long prehistory, with its possession repeatedly changing from Reginar to the church of Trier and back, each time marking how the political influence of the rivalling parties had changed at the royal court.139 Like Odo's grant of Poitiers to Count Altmar, the restoration of Saint-Servais to the church of Trier also meant the interests of a close supporter by the king were being protected. The reasons why Charles turned against Gislebert, who, some years earlier, still figured prominently among his supporters,140 are difficult to determine. Following Richer's description of Gislebert, 141 scholars often assume that the conflict was the result of Gislebert's ambition to become king142 or *dux* of Lotharingia,143 or at least to increase his political room for manoeuvre in the context of the coronation of Henry the Fowler in the East Frankish realm which had taken place shortly before.144 His political ambition seems certain. As

137 Dufour, Recueil Robert et Raoul, N° 40 (27th March 897).


Flodoard reports, during the period of Charles' problems with the West Frankish nobles after the assembly of Soissons in January 920, Gislebert was elected *princeps* by Lotharingian nobles,145 seemingly after he had created a network of support by distributing *honores* among them if Richer is to be believed.146

Gislebert's ambitions to recover at least his father's position at court147 and in particular his rivalry with the church of Trier certainly played a central role in the conflict between him and the king. Yet, in regard to its development, it appears to be too limiting to attribute the outbreak and the development of the conflict solely to these ambitions, for this fails to take Charles' own interests into proper consideration. The assembly at Herstal demonstrates how much support Charles was able to muster at that moment against Gislebert: five bishops and twelve counts,148 among them Count Sigard from Gislebert's own sphere of influence149 and Otto, the son of Count Ricuin of Verdun, who had married Gislebert's halfsister Cunigunde, 150 which probably indicates that his father also did not oppose the king's course. Just how powerful this network was becomes apparent from Charles' success. Gislebert was soon also excommunicated151 and, according to Richer, chased out of the realm to Henry the Fowler's court.152

Of course, the fact that Charles was prepared at that moment does not necessitate that he had taken the initiative in the conflict. Yet, the further development of the relations between him and Gislebert nevertheless betrays that the conflict was at least as determined by Charles' interests as by the counts. Gislebert, who had probably returned from his exile during the first half of 920 when Charles had to deal with the nobles of the Western Francia, was subdued and restored to a part of his *honores*. 153 However, once peace had been restored, Charles started to advance against Gislebert's former allies. Bishop Hilduin of Liège, who had supported the count, was replaced by a Matfriding, Abbot Richer of Prüm.154 Next came Henry the Fowler, with whom Charles, after an encounter at Worms,155 concluded first an


armistice and then a peace treaty ensuring at least his future neutrality.156 Last on the list was Count Ricuin of Verdun, who appears to have switched his support over to Gislebert, but now lost a number of fortified places to Charles before he submitted and was re-accepted into the king's favour.157 With Gislebert now isolated, Charles renewed the hostilities, chasing him and Ricuin's son Otto over the course of the winter up until Lent 922.158

In either of these cases, Gislebert may have been the initiator and not Charles. Yet, in regard to Charles' activities, this seems rather improbable. Gislebert's success was tied to the crisis of Charles' reign, a crisis that had passed by September 921 and resulted in his submission. After his submission, his position had further deteriorated, which made it even more unlikely that a renewal of the conflict would result in a success for him. This only changed when the rebellion against Charles broke out, forcing the king to concentrate his forces in the West against Robert and his supporters.159 However, this was only after hostilities between the king and the count had already been renewed. Therefore, it seems indeed safe to assume that the initiative had rested with Charles. Gislebert made use of this opportunity to improve his position by allying himself with the rebels.160 However, soon afterwards, he abandoned them again when he came to an agreement with the king, as indicated by a charter he issued 25th August 922.161 The charter, confirming the exchange of property between a certain Guntbert and Gislebert's abbey of Stavelot, was signed by him as "*dux* Gislebert, count and abbot."162 A private charter does not necessarily reflect the position of the king, especially one which was issued during times of conflict. Yet, the dating line of this charter reads "in the 11th year of the reign of the most pious lord, King Charles,"163 indicating that Gislebert and the king had indeed come to an agreement. The concurrence of the count acknowledging Charles once again as his king, honouring him as "most pious" (*piissimus*) while calling himself *dux*, further gives rise to the conclusion that Charles, confronted with the rebellion, had sought refuge in appeasing Gislebert by this time granting him what he desired: the position as Lotharingia's leading noble. Gislebert, given that Charles had already broken the earlier settlement, was probably wary of the king by now, yet his doubts were probably overcome by the

156 Flodoard, Annales 921, 6; MGH Const. I, N° 1, 1−2. See chapter IV.4.


gains he could make through the new agreement. However, this peace did not hold either. For the second half of 922 Flodoard notes only a single campaign, namely Charles besieging Gislebert's *castrum* of Chièvremont and Hugh the Great hurrying to its relief and receiving hostages from Gislebert after its rescue.164 Because Gislebert's ambitions had been satisfied by Charles in the previous settlement and because the count himself was apparently considered so untrustworthy and weak by Hugh that he had to provide hostages, it again seems safe to attribute the renewal of the conflict to the king, despite the fact that he already faced the rebellion.

Hence, the conflict between Charles and Gislebert appears to have been, possibly initiated, but certainly twice renewed by the king. Who was the driving force behind these actions? Charles' allies, most of all the church of Trier and the Matfrid family and their rivalry with Reginar's son, certainly played an important role. Yet, it seems safe to assume that neither of them opposed the first settlement with Gislebert. Archbishop Roger and Count Matfrid both accompanied Charles at Bonn when he concluded the treaty with Henry the Fowler165 and the former continued to serve as royal archchancellor at least up until June 922.166 Furthermore, Roger and Gislebert cooperated in 923 when they invited Henry the Fowler to come to Lotharingia.167 As the rivalries between them do not seem to have constituted an insurmountable obstacle to a settlement of the conflict, the source for its constant renewal appears to have been with Charles himself. The support Charles was able to gain against Gislebert indicates that his reasons were initially widely accepted by the nobles of the realm. Therefore, it seems improbable that the outbreak of the conflict would have been badly perceived by the majority of the nobles supporting Charles' rule. However, it appears that Charles considered the subsequent settlements with Gislebert mere temporary arrangements which would gain him time to strengthen his position before attacking again. As trust is a matter of being predictable and reliable, each renewal of the conflict probably undermined the trust Charles enjoyed among the nobles, or at least among those already harbouring doubts against him. Faced with the rebellion, Charles, like Odo, tried to improve his position by coming to terms with his opponent. Unlike Odo however, he then broke these settlements and thus further damaged his reputation instead of trying to restore trust in himself by keeping his side of the bargain and displaying himself as a reliable ruler.

<sup>164</sup> Flodoard, Annales 922, 11: *Rotbertus filium suum, Hugonem, in regnum Lotharii mittit cum aliquo Francorum agmine, protper Capraemontem, Gisleberti castrum, obsidione liberandum, quod Karolus premebat obsessum. Quo comperto, Karolus obsidionem relinquit, et Hugo, acceptis obsidibus a quibusdam Lothariensibus, ad patrem remeavit.*

<sup>165</sup> MGH Const. I, N° 1, 1−2.

<sup>166</sup> Roger makes his last appearance in DChS 121 (15th June 922), which is also the last diploma issued by Charles.

<sup>167</sup> Flodoard, Annales 923, 18.

#### **VI.3 Speaking justice: Symbolic acts**

While royal actions could undermine the relations between the king and the nobles when they were perceived as violations of the existing norms, rulers could also aim to strengthen trust by performing acts of justice. The most striking records of kings actively trying to do justice are certainly the capitularies, "legislative texts that recorded the decisions reached by kings in discussion with their leading men, and communicated them to a wider political community."168 Among the West Frankish kings ruling after the death of Charles the Bald, only Carloman II appears to have issued any capitularies. The first was issued 22nd February 883 at Compiègne,169 comprising only three very short chapters on robbery. The second capitulary followed about a year later in March 884 at the assembly of Ver170 and this time not only added a lengthy introduction but also extended the chapters, again mostly on robbery, to fourteen. Most of these chapters contain nothing new but only repeat old capitularies on how to pursue and judge robbers,171 quick excommunication being the most important measure. The lack of capitularies during the late Carolingian era has often been interpreted as a sign of institutional weakness.172 This view has recently been challenged, not at least by Steffen Patzold, who argues that capitularies did not serve to uphold order by recording texts of normative significance but that instead they were meant to create order by instructing people to behave in a moral way and thus prevent arising conflict.173 Even if this new reading might be considered as going too far in negating the legal character of the capitularies, it, nevertheless, becomes apparent that there seems to have been more to these texts than mere collections of newly drafted laws.

As François Bougard has observed, the capitularies issued in Italy towards the end of the 9th century became more and more formulaic and repetitive in nature; he argues that their function was not only pragmatic, but also symbolic. In regard to new kings, capitularies seem to have been drafted less to ensure justice was done in practice, but to restate the moral foundations of traditional kinship and to demonstrate an intention of good government.174 Hence, it is not surprising that Charles the Bald, for example, issued two capitularies shortly after the had taken over the Italian realm175 for both served to strengthen his rule there. Carloman's capitularies were not issued at the beginning of his reign in 880 but three years


<sup>168</sup> MacLean, Legislation, 395. See also Ganshof, Recherches; Geiselhart, Kapitulariengesetzgebung; Pössel, Authors.

later. However, the first one was issued about half a year after Carloman had taken over the northern part of the realm in succession to his brother Louis, at a moment when the Northern nobles had already turned away from him and withheld their support against the Northmen.176 The second, more impressive capitulary was issued at exactly the same moment, when an accord between Carloman and the nobles had finally been reached. Thus, both were issued at moments when Carloman needed to present himself in the best light, confirm his willingness to rule as his predecessors had done—in fact, the second capitulary is full of references to them177—and to establish order and justice in the realm. Interpreted in this context, the main intent of these capitularies was therefore much less to ensure justice was done on the ground.178 Instead they served to create the image of a king who was aiming to bring order to the realm in cooperation with the nobles around him at a moment when his own rule was or had been challenged by parts of the nobility. In short, they were issued to build trust in Carloman.

Where capitularies dealt with general problems, kings also delivered justice in person. Judicial assemblies are the most obvious example, although records of them are quite rare.179 From Louis the Stammerer, Louis III and Charles the Fat in the Western realm none are preserved, of Carloman II we have one,180 from Odo three181 and from Charles the Simple four.182 From Charles the Bald's reign, by way of comparison, only five have been preserved,183 meaning that, compared to the length of the reigns, the number of such *judicia* remained more or less unchanged.

176 See chapters II.2 and V.1.1.


However, if these numbers are compared to those of diplomas issued by the different rulers, a trend towards an increasing number of judicial assemblies per number of diplomas is noticeable for Charles the Simple: 4 out of 120 diplomas, or 3.3%, make note of them compared to 5 out of 468, or 1%, for Charles the Bald. 184 This increase in regard to the overall number of issued diplomas points towards the importance Charles attributed to these assemblies. This is an interesting observation, given that, according to Richer's description of Charles' character, one of the king's two faults was to neglect his jurisdictional duties.185 Of course Richer is a rather problematic source, yet this same reproach also surfaces in a chapter of the acts of the synod held at Trosly in 909 by Archbishop Heriveus and his suffragans.186 In this chapter, phrased as a mirror for princes and probably meant for Charles to read,187 the importance of jurisdiction is especially emphasised.188 Being close to Charles, Heriveus, in particular, would have been very well aware of the king's flaws and shortcomings and the chapter can hence be read as a direct response to these.

In this context, it is worth making some further comments. Especially during Odo's reign, judicial acts are often only referred to by sources not originating at the royal court, but by regional ones. A very interesting example of this is the notice of a local *placitum* held in April 890 at Nîmes.189 The notice narrates that Bishop Gibert of Nîmes had approached Odo, asking him to intervene in a conflict between his church and a certain Genesius, who, according to the bishop, had usurped the *villa* of Bizac, which had been given to the church by a woman called Bligardis about twelve years earlier; the notice further adds that Genesius, in order to obtain a confirmation of his claim from the count, had forged letters by the king. Odo had then tasked Count Raymond of Nîmes with investing the matter, which, in turn, led to said *placitum*. For the reigns following the death of Charles the Bald, this is the only time that such a case is reported. Does this mean that Odo was the only king to be approached by nobles from distant regions who sought royal justice in such cases? This seems rather unlikely and there are indeed some traces of, if not identical, at least similar cases. For example, in 916, Charles


186 Schmitz, Konzil, 187; Sassier, Royauté, 200. For the synod, see, next to the works of Schmitz, also Schröder, Synoden, N° 29, 189−196.


the Simple was approached by Abbot Acfred of San-Esteban of Bagnoles at whose request he issued a diploma confirming the possessions of his abbey.190 However, Bagnoles was in dispute with the abbey of Saint-Polycarpe over some property that was now confirmed by Charles' diploma as belonging to the former.191 Like in the case of Bishop Gibert, an ecclesiastical institution distant from the royal court hence sought out the king in a judicial dispute. Furthermore, we may also point out the conflict between Count Erlebald of *Castriciensis* and the church of Reims, which resulted in the count trying to seek out Charles yet being killed before he was able to meet him.192 Another noble, albeit this time one close to Charles, 193 approached the king to obtain justice. These examples indicate that Odo's case was in fact not special in its nature, but only due to the circumstance that sources have been preserved making note of it. This circumstance leads to the cases when royal diplomas did make note of royal judicial decisions. The loss of diplomas certainly contributes to the overall low number of such references, yet cannot solely explain them. Instead, it would seem that the royal chancellery only went through the effort of drafting a diploma under very particular circumstances, which would signal the special importance of these cases and their recordings for the rulers.

Carloman II's *judicium* is referred to by one of his earliest diplomas, issued 12th January 881 at a *villa Caciacum*. 194 The diploma reports that Carloman, at the request of Bishop Abbo of Nevers, restored the *villa* of Cours to the bishop's church after having held a "general council" (*commune consilium*) with the "great and first men of our palace," which is most probably a reference to a judgement made at the royal court by a judicial assembly.195 The first point to note is the moment the diploma was issued. In January 881, about nine months had passed since the assembly of Amiens where the West Frankish realm had been divided between Carloman and his brother Louis, 196 so the assembly took place already well into Carloman's reign. However, the diploma recording the judgement is only the second one preserved from Carloman's reign, the first one having been issued in 30th November 880.197 A rather long time had passed until Carloman actually started delivering diplomas. The timing, however, appears to be tied to an actual change in politics. Despite the partition of the realm, from Amiens onwards, the two brothers had always remained together, meeting with the East Frankish rulers at Gondreville and then campaigning against Lothar II's son Hugh as well as

192 Flodoard, Annales 920, 2−3.


<sup>190</sup> DChS 85.

<sup>191</sup> For this dispute, see Bautier, Recueil Eudes, 34−36.

<sup>193</sup> See chapter III.2.1.7.

Boso. 198 Louis left Carloman in the South to celebrate the holiday at Compiègne only a short time before Christmas.199 The deliverance of diplomas begins at about the same moment when Carloman actually started to be the only king around,200 marking the starting moment of his sole reign. In this context, the judicial assembly gains a new weight. Very much like capitularies, such assemblies served to strengthen the king's legitimacy and his bond with the nobility,201 they were, following François Bougard, a recurring "ritual of auto-affirmation."202 Like his later capitularies in the North, the assembly can be seen as a demonstration of Carloman's willingness to actively bring justice to the realm in cooperation with his leading nobles at the beginning of his reign—a means to strengthen trust in his rule. In addition to such thoughts, there also might have been a second reason behind this assembly. As Robert-Henri Bautier has proposed, the property restored to the church of Nevers might have previously been in the hands of Boso. 203 Hence, the restoration was not only a symbol of justice, but also a sign that Carloman was willing to distribute Boso's *honores* among those who supported him. In this context, another of Carloman's diplomas is of interest. In May 881, he again restored property, this time to the church of Orléans.204 This property, the *villae* of Chautay, Germigny-sur-Loire and Marzy, formed a block with the *villa* of Cours restored at the judicial assembly and probably belonged to Boso as well,205 thus emphasising Carloman's message to the nobles. These are of course only assumptions since, apart from the location of the *villae* in the Berry and the Nivernais (within the vicinity of Boso's power base), there is no actual evidence that they had been held by Boso before.

Like Carloman, Odo also held a judicial assembly early in his reign, at some point before 22nd March 890.206 It dealt with the appeal of a certain Ricbert, who up to this point had held property belonging to Saint-Martin of Tours. This property had originally been rented out by the monks to the archbishop of Tours, who had bequeathed it to his relative Ricbert. Ricbert, in turn, appears to have maltreated his *coloni*, who took legal action against him. In a local judicial assembly under the



<sup>204</sup> DCmII 53.

provost of Suèvres, the priest and archtreasurer Berno and the *scholarum magister* Odalricus of Saint-Martin, at the request of these *coloni*, Ricbert had been ordered to restore the property to the abbey. Ricbert's appeal to Odo failed, instead the king confirmed the previous judgement in accord with his brother Robert and "other nobles residing in the vicinity"207 and furthermore condemned Ricbert to compensate the monks by handing over some property (which, in turn, he was allowed to keep until his death). Hence, this judicial assembly appears to have had a very local character, as it not only revolved around a centre of Robertian power, but also invoked his brother and local nobles.208 The impression that this assembly was indeed only of minor political importance for Odo is furthermore emphasised by the circumstance that it appears not to have been recorded by a royal diploma, but was only afterwards referred to by a private charter noting the execution of the judgement.209

Another of Odo's judicial assemblies is also referred to only later on. A diploma issued for the abbey of Montiéramey210 notes a royal *placitum* held at Verberie where the three archbishops Fulk of Reims, Walter of Sens and Wido of Rouen, together with the bishops Dido of Laon, Honoratus of Beauvais and Riculf of Soissons, had investigated the case of the monk Aginus who claims, against Montiéramey, the abbey of Saint-Sauveur of Alfa for himself. While this assembly was certainly more important than the previous one for Saint-Martin of Tours, the question remains why Odo only referred to it in a diploma issued 30th September 892, in which he confirmed the judgement. At this moment, Odo's reign had entered a critical stage, as the king was in conflict with a number of powerful nobles: Baldwin II of Flanders, the Aquitanian nobles supporting Ebalus Manzer and probably at this point Count Altmar as well. As we have argued, at this moment Odo was facing a severe crisis of trust, which directly led to directly to the rebellion around Fulk. Odo was in dire need of rebuilding lost trust and the diploma confirming the judgement in favour of Montiéramey may have been his intended means for this, as it would show him as king who acted in concord with his nobles, enforcing justice within his realm.

This image depicted in Odo's diploma can also be found in some of Charles the Simple's. His first recorded judicial assembly dates to 19th January 916 and revolves around the abbey of Süsteren.211 The abbey had originally been given by Arnulf to the priest and artist Siginand, 212 who, in turn, appears to have asked to pass on Süsteren to the abbey of Prüm while keeping it as a *precarium*. <sup>213</sup> Zwentibold


<sup>207</sup> DOdo 19, 85: …*caeteri pernobiles in circuitu residentes…*

<sup>208</sup> On parties in judicial conflict participating in such assemblies as judges, see Nelson, Dispute Settlement, 53−54.

had then confirmed this donation, and left Süsteren with Siginand, but it was to revert to Prüm upon his death.214 However, after that, according to Charles' diploma, Süsteren had "in unjust and violent efforts of the powerful been taken from him."215 Charles, "after many and almost countless reclamations without effect"216 now used the occasion to submit the case to an assembly at Herstal with the participation "of the entire realm, of bishops, counts and *proceres* as well as judges of varying power and a *conventum* of all the nobles, all of them our *fideles,*"217 subsequently naming them—four bishops, fourteen counts and seventeen others.218 Following their judgement, he then restored Süsteren to Prüm. Unfortunately, the sources do not reveal the identity of those powerful men who had managed to bring the abbey into their possession.219 However, the message of the diploma was not only a demonstration of how Charles was willing to do justice in accord with the nobles around him, but also, and probably more importantly, a display of the powers backing him in this case. Thus, the diploma became a very concrete political statement, showing how close he had become to the Matfrid family, whose member Richer had become abbot of Prüm. Both, the display of power as well as the alliance with the Matfrid family, indicate that the opponent was someone with considerable political weight. And indeed, Charles' intervention appears not to have been the end of the conflict between this unknown party and Prüm, although it has to remain open when it resurfaced. In 949, Otto the Great again dealt with the problem, copying Charles' diploma in large parts, including the phrasing about the general assembly and adjusted only the name of the place and the names of participants.220

Three years later, this display of power and cooperation was repeated, again at Herstal.221 This time it was the church of Trier to which, by judgement of the "*scabini* of our palace and by attestation of all of our *fideles,*"222 the abbey of Saint-Servais, previously given to the church by Arnulf, was restored. This time the adversary is noted: Reginar Longneck and his son Gislebert are said to have unjustly and violently taken the abbey from the church.223 The naming of Reginar and Gislebert is not the only difference from the previous diploma however. Mark

214 DZ 2.

219 On the hypothesis of Wisplinghoff, Untersuchungen, 464, that these had been men close to Gislebert, see chapter III.2.1.7. Also Koziol, Politics, 515, sees the judgement in the context of Gislebert.


<sup>215</sup> DChS 84, 189: …*pontentiorum injusto violentoque conamine ei ablata est.*

<sup>216</sup> DChS 84, 189: *…post multas et pene innumeras reclamationes sine effectu…*

<sup>217</sup> DChS 84, 189: *…habito generali placito apud Haristallium, in conventu totius regni, tam episcoporum quam comitum et procerum ac judicum diversarum potestatum omniumque conventu nobilium, cunctorum fidelium nostrorum…*

<sup>218</sup> See chapter III.2.1.7.

<sup>220</sup> DOI 111.

Mersiowsky may have gone too far by denying the charter the character of a royal diploma and calling it a *Gerichtsurkunde*; 224 however, the diploma is certainly different from a normal royal diploma. The list of witnesses, featuring five bishops, twelve counts and eighteen *scabini*, is not included in the *narratio* this time, but added to the end of the charter. Furthermore, the diploma is sealed with a special seal225 and lacks the royal monogram.226 In fact, about a month later, Charles issued another diploma confirming the judgement of Herstal,227 indicating that the first diploma had not been drawn up with the same intention as the one from 916. Nevertheless, the assembly as such and its record certainly were symbols of Charles' power at this moment and a signal to his supporters that he had their interests at heart—a means to improve trust in his rule.

The detailed naming and the number of the nobles partaking in these two judicial assemblies is what makes these two different from the two others recorded by Charles' diplomas. The first of these other diplomas228 records the petition of Robert of Neustria for his abbey Saint-Denis 28th May 917, requesting that Charles returned the *villa* of Lagny-le-Sec, which had been granted to the abbey by King Dagobert229 but then had been "unjustly taken from them and for a long time been possessed by others."230 Charles, acknowledging the justness of his reclamation231 and following the judgement of bishops, counts and others of his *optimates*232 then restored the *villa* to Saint-Denis. However, the importance of the diploma in fact does not lie in the judgement, but in the way it is constructed around the relation between Charles and Robert. The latter is the "dearest *demarcus* and our most loyal [*fidelis*] Robert," later furthermore noted as "Abbot Robert, our kinsman."233 Finally, the diploma also adds the installation of a prayer service, not only for Charles, but also for Robert. 234 The judgement in favour of Saint-Denis is only an addition to this demonstration of proximity between the king and the *marchio*, which shows how Charles was correcting very old injustice for his closest supporter. Read in this way, the diploma recording the judgement was a political statement similar to the two assemblies of Herstal: it was a demonstration of the support the king enjoyed at this very moment, yet one focused on the relation between Charles and Robert.



<sup>228</sup> DChS 89.

The last *judicium* noted by one of Charles' diplomas took place in September 921.235 This time, it was the brother of his deceased wife Frederuna, Bishop Bovo of Châlons, who informed him that some property had been estranged from his church. At the request and following the judgement of a number of his *fideles*, his counts and bishops,236 Charles then restored said property. Like the diplomas recording the *judicia* of Herstal, this one adds the names of the *fideles* participating in the judgement: the counts Hagano, Helgaud and Raoul de Gouy, as well as the bishops Abbo of Soissons and Stephen of Cambrai—most of them belonging to the group of nobles which had formed around Charles in the last years before the final rebellion of 922.237 If the judgement recorded by the diploma was meant as a demonstration of power, it was a rather inept one. The number of participants was so small that it would undoubtedly have been reasonable to do without naming them and leave it to the general terms of *fideles*, counts and bishops that are also to be found in the diploma recording the judgement for Robert. Given that it would have been easily possible to do so, it seems as if the reference to them was indeed intentional. What mattered was not their number, but their names: they were meant to be seen. As we have argued, from September 920 onwards, seven months after the assembly of Soissons, Hagano's appearance in royal diplomas changed significantly, now acting together with other nobles, seemingly in an effort to integrate him in the network of the nobles around Charles. 238 Hagano appearing in this diploma was another step in this direction, showing him acting together with other nobles and taking part in the judicial procedures at court. Like the previous ones, also this last diploma recording a *judicium* was hence an important political statement by Charles.

However, judicial assemblies and their recordings were only one part of the king doing justice. As Philippe Depreux has pointed out, royal diplomas dealt with a number of judicial problems very similar to the decisions taken at these assemblies, such as settling disputes and restoring unjustly withdrawn freedom or, most importantly, property.239 Up until Odo's reign, a surprising number of these latter went to the church of Autun.240 Louis the Stammerer, at the request of Bishop Adalgarius and Count Theoderic the Chamberlain, restored the *villa*  of Bligny-sur-Ouche, which previously had been passed over to the count of Autun.241 Louis' diploma was copied by Carloman II, replacing only the property restored, the *villa* of Teigny.242 Again, it saw Adalgarius and Theoderic acting

237 See chapters III.2.1.7, III.2.1.8 and III.2.2.

241 DLS 29 (23rd January 879).

<sup>235</sup> DChS 112 (20th September 921).

<sup>236</sup> DChS 112, 268: *Cujus petitioni libentissime faventes deprecantibus et adjudicantibus fidelibus nostris, comitibus nostris atque episcopis…*

<sup>238</sup> See chapter III.2.2.

<sup>239</sup> Depreux, Absence, 10.

<sup>240</sup> DLS 29, DDCmII 49 and 68, DDOdo 21 and 35. For Boso, see Charmasse, Cartulaire Autun II, N° 1, 85−86. DProv 17 is a forgery, see Bautier, Receuil Eudes, CXLIV−CXLVI.

<sup>242</sup> DCmII 49 (30th November 880).

together, this time supported by the new count of Autun, Richard. The first two also appeared together, assisted by Count Ansgar of Oscheret, in Carloman's second restoration, this time of the *villa* Marigny.243 The diplomas reveal how well connected Adalgarius was within the elite of the nobility, as he not only acted with one of the central figures of Louis the Stammerer's and Carloman II's reign— Theoderic the Chamberlain244—but was also able to obtain restitutions from the *honores* belonging to the count of Autun with the support of that very same count. Moreover, after Louis' death Adalgarius had belonged to Boso's supporters, and even became his archchancellor.245 Seemingly, this had not damaged his position within the West Frankish elite. Under Odo, Adalgarius continued to use his influence to obtain restitutions for his church, albeit with differing results. At his request, Odo returned Congy, Servigny and Savigny, held by the noble lady Emma, to the church of Autun,246 yet when the lady presented royal diplomas confirming her claim to the domains, Odo, after he had consulted his *fideles*, returned the property to her, and decided that she was to hold them until her death when they should finally fall back to the church.247 Adalgarius, who pursued an aggressive course by using his influence at court to increase the wealth of his church even if his claim was not entirely justified, finally is also revealed by Odo's other restoration to him. With the support of Odo's brother Robert and by presenting a number of charters as evidence for his claim, he obtained the *villa* of Tillenay.248 While the diploma itself is genuine, apparently the evidence brought forward on this occasion by Adalgarius appears to have been forged.249 Geoffrey Koziol argues that this circumstance was actually known by Odo who decided to ignore it to tie Adalgarius, one of the key players in Burgundy, closer to him250—the bishop later even became his new archchancellor. Yet, even if this was not the case, the restorations obtained by Adalgarius reveal how a key figure of politics could use his connections to influence the royal judgement in his favour.

Similar to Adalgarius, Bishop Geilo of Langres also seems to have profited from his connection to Charles the Fat to increase the wealth of his church by receiving not only grants, but also by having property restored. On three occasions he managed to have *honores* restored, the abbeys of Moutier Saint-Jean251 and Saint-Seine, 252 as well as the *villa* of L'Ormeau.253 However, powerful nobles using their connections to influence royal justice was only one half of the story.


<sup>243</sup> DCmII 68 (6th March 883).

The first diploma restoring an abbey to the church of Langres—the second one being an exact word-by-word copy—uses some exceptional language to describe the situation. According to the diploma, the abbey had been "taken away [from the church of Langres] in tyrannical and sacrilegical temerity of the *principes* as well as in known presumption without any respect for the allmighty God" and later adds that this had been done by violent means.254 This was far from the other wording usually used in diplomas restoring property which, if at all, normally only pointed out that said property had been unjustly taken away.255 While the diploma was written by a West Frankish scribe, probably from Langres,256 it nevertheless seems that it was drawn up in the royal chancellery since there is a broad consistency between its arenga as well as parts of the *formulae* with two other diplomas issued for the church of Dijon, which would indicate that the scribe must have either known them or used a common template.257 Hence, the diploma transmits a very clear message originating at the royal court on the willingness of the emperor to protect the interests of those who supported him. However, the question is to whom this message was addressed. Both diplomas for the church of Langres using these phrasings were not issued in the West Frankish realm, but in the East, at Lorsch and Schlettstadt. On both occasions, nothing in the sources indicates the presence of a larger assembly of nobles other than those who usually were with the king. Hence, were these diplomas meant especially for Geilo, as a sign of the strength of the relations between him and Charles? Given the number of restitutions made by Charles to West Frankish institutions—apart from the three for the church of Langres he also restored property to those of Lyon,258 Châlons259 and the abbey of Saint-Vincent of Mâcon260—it seems reasonable to assume that the restitutions were indeed aimed at a wider audience also in the Western realm and supposed to build trust in the emperor by demonstrating his efforts to further the interests of those loyal to him.

In using this strategy, he certainly was not alone. The last restitution made by Carloman II, also for the church of Châlons, transported a very similar message. Delivered 13th March 884 at Compiègne, therefore at the same time he had issued his second capitulary, he restored the abbey of Saint-Sulpice to the church.261 The


way the diploma depicts Bishop Berno, its petitioner and recipient, is most striking. He is not only the one "who from the beginning of our reign excelled, being most useful and loyal to us in council and aid," but also the one who "decided to remain within this same loyalty."262 This loyalty was then rewarded by what appears to have been a longstanding problem for the church of Châlons, to have the abbey of Saint-Sulpice returned to it. Already back in 859 and again in 875, Charles the Bald had restored the same abbey to the church, then held by a certain Bernard and a certain Geoffrey respectively, yet both times under the reservation that Saint-Sulpice remained with its current holders up until their death.263 This same reservation was now also renewed by Carloman, this time for a certain Rothard. While Charles the Bald's diplomas betray nothing about Bernard and Geoffrey, Carloman's reveals a bit more about Rothard, who appears as "our *vasallus* and judge."264 Was it possible that on all three occasions the kings or their local representatives had lain hands on Saint-Sulpice to grant it to their own followers? If this was indeed the case, Carloman's diploma, despite the reservation, would be a symbol of how far for his closest followers, those who remained loyal to him, he was willing to correct injustice even caused by himself—undoubtedly the best way build trust while starting into the new era of his reign.

Similarly, Odo's first restoration265 also coincides with one of the crucial moments of his reign, the great assembly at Orléans in June 889 where nobles from Septimania and Aquitaine were present. The *villa* of Doussay, apparently a large estate to which twenty other *villae* were dependent, was restored to the monks of Saint-Martin of Tours.266 Unfortunately, no diploma confirming the restoration has been preserved, the only trace of Odo's act being a private charter issued by Ebalus Manzer 10th October 891.267 According to this charter, it had been Charles the Bald who had originally restored the *villa* to the monks, yet "after a long time, [it was] maliciously usurped by Magenarius, after him unjustly held by his son Osbert, but then again, during the reign of Lord King Odo, at the city of Orléans, in the presence of his *proceres*, restored to us."268 This same language was also applied in Odo's other preserved restoration, this time for the church of Narbonne,269 noting that fiscal property that had been bestowed to the church for a long time and

264 DCmII 76, 198: *…vasallus et judex noster Rothardus…*

266 Bautier, Recueil Eudes, 228.

<sup>262</sup> DCmII 76, 198: … *qui nobis in principio regni nostri et consilio et auxilio satis proficuus et fidelis extitit atque in eadem fidelitate perseverare decrevit…*

<sup>263</sup> DDChB 212 and 381.

<sup>265</sup> DOdo 9.

<sup>267</sup> Bautier, Recueil Eudes, Appendice III, 227−233.

<sup>268</sup> Bautier, Recueil Eudes, Appendice III, 231−232: *…post multum tempus, Magenarius dolo usurpavit et, post eum, filius ejus Osbertus, injuste tenuit, sed rursus, domni Odonis regis imperio, Aurelianis civitate, nobis restitutam in praesentia suorum procerum…*

<sup>269</sup> DOdo 24.

unjustly usurped by the comital power was now being returned by the king.270 Hence, it seems safe to assume that Odo's restoration for Saint-Martin also emphasised him as a ruler who corrected ancient injustice and protected the interests of those close to him.

Like his predecessors, Charles the Simple used restitutional acts to portray himself as a king who brought justice to his supporters in crucial moments of his reign. One of his first diplomas, issued at the request of Bishop Lambert of Liège for his church on 26th September 894, spoke of parts of the fisc of Arches in the Porcien which had been taken from said church by "force of arms and violence,"271 but were now restored by Charles, who had just been acknowledged by Arnulf as king and had returned with Lotharingian forces to the West. The two assemblies of Tours-sur-Marne in 899 and 922 also saw restitutions. In a diploma for the abbey Notre-Dame of La Grasse, Charles not only confirmed a large number of specific properties, but also those which had been estranged from the abbey.272 The diploma was in large parts a copy of one of Charles the Bald; 273 but Philippe Lauer remarked that this particular phrase had been significantly changed from *reliqua quae ibi collatae fuerunt* ("the remaining [property] which then had been *bestowed* [to them]") to *reliqua quae ibi ablatae fuerunt* ("the remaining [property] which then had been *taken away* [from them]").274 This change certainly had been intentional, demonstrating to those present at the assembly how Charles was willing to protect the interests of those who supported him. However, the actual impact of this phrase was certainly minimal. La Grasse could use Charles' confirmation in their favour, yet if they did, they still had to present evidence that the disputed property had indeed belonged to them. In contrast to this diploma where the restitution of property was only one phrase at the end of a long list of confirmations and hence hardly the centre of attention, the diploma issued in 922 was drafted around this very act. This time the monks of Saint-Thierry of Reims approached Charles and presented the king a diploma of Charles the Bald confirming them in the possession of the domain of Marzelle, protesting that "their living had been unjustly withdrawn from them."275 Charles, who acknowledged their protest as justified, by his power then returned the *villa* to them.276 With Robert's army encamped only a few miles away, Charles once again tried to build trust by portraying himself as a just king who protected the church in its need and corrected injustice.277



<sup>270</sup> DOdo 24, 112: *Fiscos vero qui sunt in Biterrensi comitatu sancto Paulo confessori a longo tempore collatos et a potestate comitali injuste usurpatos plenissime reddidimius…*

<sup>273</sup> DChB 340.

However, this highly symbolic potential of restitutional acts may not only have served to strengthen royal rule by building trust with the receiving party, but at times also have been considered a breach of trust by a third one. Among the first diplomas issued by Charles the Simple at Richard the Justiciar's request was one for Bishop Herfrid of Auxerre and his church, which restored the *villa* of Cravant sur l'Yonne.278 As we have argued, Richard tried to channel the contact between the local Burgundian nobility and the king, installing himself as the only mediator.279 In doing so, he also presented himself as the only means to obtain justice from the king, as demonstrated by this restoration, which was obtained at his intervention. However, in 902, Charles did not only interact directly with Herfrid—that is to say without Richard being present—he also again restored property to Herfrid's church, this time the *villa* of Mailly-sur-l'Yonne which had been "once unjustly taken away."280 By doing so, Charles hence not only circumvented Richard, but also demonstrated to the Burgundian nobles that he himself was in fact still the source of justice and not Richard, thus further adding to the attack on Richard's position. In front of this background, it is probably not surprising that, once Richard had returned to court a few years later, Charles again restored property at Richard's request, this time to Saint-Martin of Autun,281 and thus again confirmed Richard as the authority who was to be addressed if royal justice was needed.

Hence, doing justice was a highly symbolic tool used by the rulers to strengthen their position by building trust in moments of political change and turmoil. Carloman II's capitularies are the most prominent examples, with the second one being issued at the heights of the difficulties between him and the nobility from the northern part of the realm. Similarly, the only reference to a judicial assembly in one of Odo's diplomas dates to September 892 when Odo's reign was shaken by the conflicts in Aquitaine and with Baldwin II of Flanders. As it seems, these rulers made use of judicial acts to portray themselves as just rulers, acting in accord with the nobles around them to bring order to the realm and thus to build trust in their rule. This image changes to some extent under Charles the Simple. He used judicial acts as well—most importantly the late diploma for the monks of Saint-Thierry of Reims in the context of the second assembly of Tours-sur-Marne—to

is shown complaining about incursions of lay officials within his diocese (DChS 24, 49: …*quod in sua parrocia seu, in cunctis episcopiis qui in suffragio ipsius positi sunt gravissima, quedam contra jura canonum atque instituta legum increverit consuetudo, adeo ut comites atque judices seu ministri illorum sacerdotes Domini sive reliquos ecclesie ministros ad placitum suum violenter ducere et fidejussores tollere atque eos more laicorum distringere presumant…*). Charles reacted to this complaint, among other measures, by protecting the clerics of the archdiocese from coercive measures as well as the jurisdiction of lay officials. While not a restorational act, its message concords with the later one for the monks of Saint-Thierry of Reims.

<sup>278</sup> DChS 31 (after 30th June 900). DChS 33 (30th June 900) restores the mint to the church of Autun. However, this passage appears to be an interpolation. Bautier, Recueil Eudes, CLII, n. 1.

<sup>279</sup> See chapters III.1.2 and III.2.3.

<sup>280</sup> DChS 42 (14th March 902), 92: *…holim injuste a pręfata ecclesia abstractum…*

<sup>281</sup> DChS 59 (8th August 908).

demonstrate that he was a ruler who corrected old injustice for those loyal to him. Yet, the most prominent examples for such acts, the judicial assemblies of Herstal, which predate these events, emphasise not so much the justice of the king than the wide support Charles enjoyed. The diplomas issued on these occasions were undoubtedly symbols of royal justice, yet, more importantly, they were symbols of royal power directed against the king's opponents. While the number of diplomas recording judicial assemblies is rather small, those making note of restitutions is far bigger. Much more than in the former, in the latter the influence of the recipients becomes visible, with a number of nobles enjoying a particularly close position to the ruler which benefitted them numerous times. Charles the Fat especially appears to have used restitutional acts to tie West Frankish nobles to himself. Yet, restitutional acts were at times also used for the symbolic value inherent to them, issued at moments of particular political importance to transport the image of a ruler who protected the interests of and brought justice to those loyal to him.

#### **VI.4 Against the king: Questioning the right to rule**

Rebellions against the ruler were nothing uncommon in the Frankish world, even if they differed in motivation and scope. The first rebellion to be examined here is the one which developed over the succession of Louis the Stammerer and led to the coronation of Boso of Vienne. According to Hincmar, the rebellion was due to the ambitions of Boso's wife Ermengard: "She declared that, as the daughter of the emperor of Italy and the one-time fiancée of the emperor of Greece, she had no wish to go on living unless she could make her husband a king."282 Regino of Prüm, on the other hand, notes that "[Boso] disregarded the adolescent sons of Louis [the Stammerer], disdaining them as if they were of inferior birth because on Charles' order their mother had been spurned and repudiated."283 Despite the differences in the accounts, an important observation can be drawn from them: both describe Boso as the central figure of the rebellion. This becomes even more evident when we also take the rest of their accounts in consideration:284 <sup>285</sup>

284 Annales Bertiniani 879, 239: *[Boso] […] partim comminatione constrictis, partim cupiditate illectis pro abbatiis et uillis eis promissis et postea datis, episcopis illarum partium peruasit ut eum in regem ungerent et coronarent.* Translation by Nelson, Annals, 219.

285 Regino, Chronicon 879, 114: *Denique nonnullos episcopos partim minis partim suasionibus in societatis foedera colligit et Lugdunum ingressus ab Aureliano eiusdem urbis metropolita et aliis pontificibus in regem super prefatum Burgundiae regnum inunguitur…*

<sup>282</sup> Annales Bertiniani 879, 239: *Interea Boso, persuadente uxore sua, quae nolle uiuere se dicebat, si, filia imperatoris Italiae et desponsata imperatori Greciae, maritum suum regem non faceret…* Translation by Nelson, Annals, 219.

<sup>283</sup> Regino, Chronicon 879, 114: *[Boso] pro nihilo ducens adulescentes filios Ludowici et velut degeneres despiciens, eo quod iussu Caroli eorum genitrix spreta atque repudiata fuerit.* Translation by MacLean, History, 180.

So Boso persuaded the bishops of those parts to anoint and crown him king. He had partly browbeaten them by threats and partly won them over because they were greedy for the abbeys and estates he promised them, and which he gave them afterwards.284

#### *Annales Bertiniani 879 Regino, Chronicon 879*

Partly by persuasion and partly through threats [Boso] rallied several bishops to make alliance with him and after he had entered Lyon he was anointed as king of Burgundy by Aurelian, metropolitan of that same city, and some other bishops.285

According to both accounts, not only was Boso the driving force behind the rebellion, but there also was no stronger opposition against Louis III and Carloman II. Indeed, it was Boso himself who had to rally support to be crowned king. From these accounts, it would appear that the rebellion was not due to the young kings infringing on what Boso had perceived to be his privileges, but the result of his (or, according to Hincmar, his wife's) ambitions. However, Regino's report corresponds with Boso's *electio* in which Boso claimed that after Louis the Stammerer's death the throne had become vacant due to the lack of suitable persons.286 Boso disputed the existence of legitimate kings *per se*. This leads us to the second important observation: the rebellion did not break out as the result of a crisis of trust in the brothers and their rule (as did the one against Odo287), but because Boso wanted to become king himself.288 This was not a rebellion against Louis and Carloman, it was "Boso's rebellion."

This impression is further strengthened when analysing the support Boso was able to rally. As Robert-Henri Bautier, based on the witness list of Boso's *electio*, has demonstrated, this support originated in the east and south-east of the realm, including the archbishops of Aix, Arles, Lyon, Vienne, Tarentaise and Besançon along with a number of their suffragans as well as Abbot Geilo of Tournus.289 Of these men, only four can be traced at Louis the Stammerer's court: Archbishop Aurelian of Lyon, Bishop Adalgarius of Autun, Bishop Lambert of Mâcon and Abbot Geilo. 290 Of these, Aurelian and Geilo were connected to Boso who had previously intervened on their behalf. Hence, at least among the higher clergy, Boso's support rested mostly on men who do not appear to have been taking part in royal politics. It is probable that some of them hoped to increase their influence on the affairs of the realm by being close to the new king. After the coronation, the archbishops of Vienne and Arles as well as the bishops of Valence and Grenoble can be seen at Boso's court.291 However, key positions at Boso's new court were occupied by men who had already been influential under Louis the Stammerer.

<sup>286</sup> MGH Capit. II, N° 284 C, 368: *…idoneas personas […] non invenientes…* See chapter I.1.1.

<sup>287</sup> See chapter I.2.3.

<sup>288</sup> See also Bautier, Origines, 49−50, who points out that Boso's declaration of Mantaille shows no signs of any territorial restrictions but was aimed to demonstrate his claim for the entire realm.

<sup>289</sup> Bautier, Origines, 50−51.

<sup>290</sup> DDLS 19, 20, 26, 27, 29 and 33. See chapter II.1.

<sup>291</sup> DDProv 20 (Vienne), 21 (Arles), 22 and 24 (Valence) as well as 25 (Grenoble).

Aurelian crowned Boso,292 Adalgarius became the new archchancellor and Geilo was promoted to bishop of Langres.293 This may not be much of a surprise since Boso could hardly ignore important nobles such as these when it came to distributing honours and *honores*. Yet, at the same time, this limited the possibilities of advancement for his other supporters which makes it rather doubtful whether this was indeed the dominating motivation behind their support. In fact, it seems that Boso's network was extremely fragile, with several of these men soon defecting from his side and turning to Carloman, most notably Adalgarius294 and Geilo, 295 but also the bishops of Marseille296 and Valence.297 From the small support Boso enjoyed among those previously influential at Louis the Stammerer's court as well as the circumstance that their reintegration under Carloman's rule was apparently unproblematic, it would appear that their decision to take part in Boso's rebellion was less motivated by a rejection of Louis the Stammerer's sons or their political ambition, but by their personal relation to Boso and the promises and threats he made, as stated by Hincmar and Regino.

This underlines the importance of Boso's motives for the rebellion even more. On these motives, Stuart Airlie has brought forward an intriguing explanation. Boso was related to the Carolingian family in multiple ways. He was married to the daughter of Emperor Louis II, Charles the Bald had married his sister Richilde, his aunt Theutberga had married Lothar II. According to Airlie, Boso, also being showered with offices by Charles the Bald, had been turned "into what we might call an honorary Carolingian,"298 going even as far as to speculate that Charles might have considered naming Boso as his successor.299 However, with Charles' death, Boso's extraordinary position changed. Boso remained one of the central men of the realm, yet his position was nowhere near as exalted as before.300 With Louis the Stammerer's succession further deepening the division among the rivalling noble groups, the pressure on Boso's position, which was based on his relations to the king and the other nobles, increased even further. According to Airlie, it was this threat that caused Boso's rebellion: "He had risen spectacularly high. He had nowhere to go but down."301

The character of Boso's rebellion and of his support for it had a significant impact on Louis III's and Carloman II's capacity to handle it. First, if Airlie's


explanation is accepted, this means that their rule was not challenged by a larger group of nobles who questioned their right to rule, but by a single man who feared the loss of his political position and activated his personal connections to tackle the issue. This entails that his supporters were not principally opposed to the brothers' rule and that their turning away from Boso and rehabilitation under the brothers' regime was possible. This is a very strong contrast to the opposition Odo encountered, which even endured considerable military setbacks. Second, in terms of opposition against Louis and Carloman, Boso's rebellion, despite his ability to muster a large number of nobles in his support, was rather limited and fragmented.302 The groups of nobles who had dominated Louis the Stammerer's politics remained largely intact and now supported Louis III and Carloman II in suppressing Boso's revolt, providing them with sufficient resources to overcome the rebels. While it still took two years of siege to capture Boso's stronghold of Vienne,303 nothing indicates that the rebellion posed as much of a challenge to Louis and Carloman as Fulk and his allies posed for Odo. Finally, Boso's rebellion occurred at a moment when the Carolingian monopoly on the throne was still unchallenged304 and stirred up an alliance of the Carolingian rulers as a response.305 Boso, in contrast to Charles the Simple against Odo, could not hope to win further support from rulers outside the West Frankish realm. His position remained isolated while Louis and Carloman could rely on the help of their peers to take out the rebellion.

While the rebellion Louis III and Carloman II faced differed considerably from the one Odo had to deal with, the one against Charles the Simple appears to have been the result of a development rather similar to the one that had brought him to the throne. In chapter III we have emphasised the cooperation and support Charles enjoyed from important members of the nobility even during the crisis of 920 and up into the rebellion. However, this emphasis should not distract from the fact that his relations to an increasing number of nobles were deteriorating, swelling the ranks of an opposition forming against him. As argued, the aim of this opposition was at first not to question his right to rule, but to press him for a change of policy, as parts of the nobility had already attempted to do in early 920 at the assembly of Soissons. This only changed over the course of 922 when Charles evaded negotiations with the opposition led by Robert and his son Hugh. From this moment on, key members of the circle around him, like Count Heribert II of Vermandois and Bishop Abbo of Soissons, turned away from him, which indicates how much he had lost the trust of even the nobles closest to him. The conflict with Gislebert may serve to help to explain the development of this crisis of trust, yet this reading is not unproblematic. The reasons for the outbreak of the conflict and

<sup>302</sup> On the fragmentation of his support see, Bautier, Origines, 49−50 and Castelnuovo, Élites, 383−392.

<sup>303</sup> Annales Bertiniani 880, 243 and 882, 247.

<sup>304</sup> Airlie, Semper fideles, 139−140.

<sup>305</sup> See chapter IV.2 and MacLean, Response.

Charles' role in it are far from certain. Charles renewed the conflict twice, yet the second time the rebellion had already taken its course while the first time also only occurred in the period after the assembly of Soissons. Hence, while the renewals of the conflict may serve to explain the deterioration of the relations with the other nobles, they cannot explain its cause. This only changes if we assume that the behaviour demonstrated by Charles over the course of the conflict was not an isolated occurrence but to a certain degree symptomatic in situations when he felt that his interests were at stake.

As demonstrated, favouring Hagano certainly did not mean that Charles did not listen to other nobles either. Yet, Hagano's special proximity to the king combined with his origin outside the circles of the West Frankish nobility were perceived as infringement on the privilege of the leading nobles to counsel Charles, based not on their personal relation to him, but on their prestige and rank. Elevating men of his own choice to advisors may have been rooted in Carolingian tradition306, yet at least in the late Carolingian era it posed considerable problems307, as a number of similar cases demonstrate. Geneviève Bührer-Thierry has analysed the reproaches in the Mainz continuation of the *Annales Fuldenses*308 against Charles the Fat's closest councillor, his archchancellor Bishop Liutward of Vercelli. 309 As with Hagano, the Mainz author attacked Liutward because of his (supposedly) humble origins, being elevated by the ruler, and only by the ruler instead of by right of rank, to his side.310 Again, similar to Hagano, Liutward was furthermore accused of being extremely familiar with Charles the Fat311 and even, according to Regino of Prüm, of being the emperor's only councillor.312 To this example we might add the reproaches against Zwentibold, who, again according to Regino, was blamed for listening to advisors who were not noble and even to women.313 Reproaches of


such violations of the nobles' prerogative to counsel the king are thus revealed as a central argument when it came to defining "bad kingship."314

After Soissons, Charles adjusted his policy by attempting to publicly integrate Hagano into the noble hierarchy. From this point of view, Soissons was a compromise that suited the interests of the king, who was to keep his intimate at court. However, in his efforts to provide Hagano with a position that would allow him to gain the acknowledgement of the other nobles and to occupy the position at the king's side, Charles probably went too far. Flodoard reports that Robert's son Hugh started to advance on Laon because Charles had granted the abbey of Chelles, belonging to Charles' aunt Rothild who also happened to be Hugh's mother-in-law, to Hagano. 315 Charles choosing Chelles may have been motivated by his relation to Rothild: Taking the abbey from his aunt, he may have hoped not to inspire any more discontent among the nobles. Seemingly, however, he had underestimated the weight of Rothild's connection to the Robertians, as well as the amount of dissatisfaction that had already accumulated at this point.

Yet, Chelles probably was only the tip of the iceberg. After Hugh's advance, Flodoard's account is full of hints about the position Hagano now occupied. After the outbreak of the conflict, Hagano's "accomplices" took the fortress of Epernay.316 Richard the Justiciar's son Hugh the Black, when just about to join Robert, encountered "200 men who were with Hagano"317 who were about to plunder possessions of the church of Reims. Flodoard's wording is rather ambiguous: were these Hagano's own men or were these forces Hagano had been able to raise from contacts he had made over the past years? In any case, the report seems to indicate that Hagano had now risen to a position from where he was able to contribute to the war effort. Furthermore, at Laon Hagano appears to have amassed some treasures that were distributed by Robert after he taken the city.318 Combined with the grant of Chelles, these passages indicate that Charles had indeed tried to furnish Hagano with the resources necessary to enter the highest ranks of the nobility. Yet, even if this was not the case, it seems that Charles' relation to Hagano was still perceived as especially close. Flodoard uses strong indicators to describe this relation. "Because of his love for Hagano,"319 he tells us, Charles departed Laon,


<sup>314</sup> Nelson, Bad Kingship, 19.

evading the encounter with Hugh. After Laon had fallen to Robert, "Charles together with Hagano" hurried to retake the city.320 Soon after, again "with Hagano", he retreated over the Meuse.321 For Flodoard, Hagano was always with Charles, he accompanied the king on all occasions, he was his support and his downfall. The same image can also be found in Folcwin's report, although much more compressed: "Confirmed in his reign, [Charles] elevated a certain Hagano, whose lineage and nobility were unknown to the Franks, esteeming him over everyone else and calling upon him in council in a more familiar way than the others. The Franks, seeing this, not lightly endured this unknown man's friendship; and they were incensed against the king about this and they strove to attack him at the royal city of Laon."322 Folcwin simplifies the events by omitting the assembly of Soissons, yet the key message is the same as Flodoard's: Charles' efforts to integrate Hagano in the noble hierarchy had not been enough to pacify the nobles because at the same time he had failed to resolve the other issue at stake, namely to normalise the relations with his intimate.323

Charles' adherence to his relation to Hagano reveals a distinctive obstinacy in pursuing his own interests despite the resistance demonstrated by the nobles and the threat this resistance constituted for his rule. This appears to be the same obstinacy as revealed by his renewals of the conflict with Gislebert. 'Obstinacy' is perhaps too strong a word, but certainly the years of fighting against Odo had already been marked by a characteristic persistence, as Charles faced constant setbacks yet also made constant efforts to gain new allies.324 During these early years the circle of nobles supporting Charles and leading the rebellion against Odo played an important role. Charles' final act however—turning to Hundeus and his Northmen—took place after these nobles had submitted to Odo. Thus, the alliance with the Northmen was probably Charles' own decision in his efforts to prolong the fight against his opponent. Although this alliance had triggered Archbishop Fulk to harshly criticise Charles, threatening him to end his support and to

<sup>320</sup> Flodoard, Annales 922, 10: *Karolus cum Haganone Laudunum contendit.*

<sup>321</sup> Flodoard, Annales 922, 10: *Et cum cotidie, copiis Rotberti crescentibus, decrescerent Karoli, clam tandem secendes cum Haganone trans Mosam proficiscitur.*

<sup>322</sup> Folcwin, Gesta Abbatum S. Bertini Sithensium, c. 99, 625: *Qui confirmatus in regno, quendam Haganum, cuius genus et nobilitas ignorabatur a Francis, super omnes diligendo extulit et hunc familiarius ceteris sibi consiliarium ascivit. Quod videntes Franci, non leviter ignoti hominis amicitias tulerunt; regique ob hoc infensi, insidias ei in Lugduno regia civitate tetenderunt.*

<sup>323</sup> In this context, it is worth mentioning an anomaly within Charles' diplomas. From 31st May 922 (DChS 115) onwards, a certain Hagano appears as royal notary, replacing Gauzlin, who had become bishop of Toul (Lauer, Recueil Charles III, XXV). It has been suggested that this Hagano is identical with Charles' intimate (Koziol, Politics, 468). Given that we know nothing about Hagano's past and education, this is certainly a possibility. However, it seems rather improbable that there was no other man within the royal chancellery (still led by Archbishop Roger of Trier) to be up to the task and we therefore argue that the matching names are a mere coincidence.

<sup>324</sup> See chapters I.2.5 and I.2.6.

excommunicate him,325 Charles, in a very similar situation in 923, reverted back to this strategy. When after the battle of Soissons he was not only abandoned by his Lotharingian forces but his negotiations with Heribert II of Vermandois, Archbishop Seulf of Reims and other important nobles of the Western realm also failed, he again appears to have sought support among the Northmen, both from the Loire and the Seine.326 The decision to turn to the Northmen certainly reveals how desperate his situation was at that point, yet it again also demonstrates Charles' determination to pursue his interests to the last, with total disregard for any consequences this might have on his relations with the nobles. This persistence in pursuing his interests may also be what Richer alluded to when, in his characterisation of Charles, he attributed to the king "a lack of self-control over his lusts."327 Gerhard Schmitz and Yves Sassier have also noted this as a strong theme in the already mentioned second chapter of the acts of the synod of Trosly328 where it is emphasised that only someone who was able to control himself could legitimately rule over others.329 It is of course impossible to tell whether this reproach for a lack of self-control and the persistence shown by Charles had the same roots, yet it is tempting to assume that both were connected.

If this persistence was indeed symptomatic of Charles' behaviour, it may serve to explain the crisis of trust against him leading to the rebellion. This crisis certainly evolved over time and culminated further throughout the rebellion up until Charles was captured by Heribert II of Vermandois. Its full extent is revealed in the aftermath of the battle of Soissons. Robert had been killed, yet the *primates* of the realm, among them Heribert II and Archbishop Seulf of Reims, continued to deny Charles' right to rule, rejecting his request to be reacknowledged by them, turning instead to Robert's son-in-law Raoul, son of Richard the Justiciar, on whom they


called to become their new king.330 While at the outbreak of the rebellion Charles still had the support of some parts of the nobility, at its end he was so deeply mistrusted that even the death of his opponent could not convince the nobles to treat with him again.

Yet, there is a problem with this explanation. Given that Charles' persistence can be already made out at the beginning of Charles' reign, it would seem likely that this behaviour would have marked his entire rule. However, early on, in the wake of the turmoil caused by Archbishop Fulk's murder and a couple of years later when Richard the Justiciar was replaced by Robert of Neustria at court, Charles' actions instead betray a great degree of political flexibility. A settlement with Baldwin II of Flanders was sought and Robert was finally successfully integrated into the rule. After this, the diplomatic evidence indicates strong support for Charles' rule among the nobles throughout the reign up until the beginning of the rebellion. If it was indeed Charles' persistence that caused the deterioration of his relations with the nobles, during the years preceding Soissons a development must have taken place that either strengthened this trait or lifted restrictions that so far had ensured that its effects on his relations to the nobles remained rather limited. This change would most likely have taken place in Charles' direct vicinity, in the circle of men and women in direct contact with him and with influence on his actions. Frederuna's death in 917 and Hagano's rise are both conveniently close to explain such a change,331 yet given the lack of evidence they—like any other possible explanations—are mere speculation.

Be that as it may, Charles, like Odo in 892, appears to have lost the trust of the nobles. However, unlike Charles, Odo had managed to suppress the revolt against him. Superiority in resources certainly played a crucial role here, enabling him step by step to bring his opponents' possessions under his control and finally force them to surrender to him.332 At first, Charles' forces may have been sufficient to reach a stalemate with Robert's army, as can be seen from the development of the first campaign where neither side tried to attack the other.333 Yet, while Charles' and Robert's armies were encamped in close proximity to each other, Robert's allies were able to deal Charles' position a serious blow by taking Laon while his

333 Flodoard, Annales 922, 8−10: *Quo comperto, Karolus, Mosa retransmissa, cum nonnullis qui ad se venerant Lothariensibus, villas Remensis aecclesiae depraedari necnon incendere coepit… Rotbertus igitur, super Maternam fluvium […] procedit obviam: quem subsequens Karolus cum Lothariensibus Maternam transiit… Rotbertus etiam, praefato Rodulfo cum Burgundionibus veniente, fluvium transiens subter Sparnacum, non longius tribus leugis ab exercitu Karoli castra metatus est. […] Hinc Rotbertus apud Calmiciacum, et Karolus circa Remis castra metati sunt.*

<sup>330</sup> Flodoard, Annales 923, 14.

<sup>331</sup> Hagano's influence on Charles is similarly depicted by Koziol, Politics, 469, who assumes "that Hagano was telling Charles what Charles already deeply believed…", thus strengthening Charles' belief in his heritage and his enmity towards Robert of Neustria.

<sup>332</sup> Annales Vedastini 897, 77. See chapter I.2.7.

own army appears to have suffered from frequent defections.334 This constant loss of support seems to have shifted the balance of power more and more against Charles, which explains why in 923 he attempted to take Robert by surprise before his opponent was able to fully muster his forces.335 Yet, military superiority was only one of the reasons for Odo's success. As we have argued, Odo also managed to restore trust in him when he returned Saint-Vaast to Baldwin II of Flanders in 895 and used a seemingly highly regarded mediator, his brother Robert. Charles also made such efforts yet these efforts failed—probably because of their innate character.

The aim of the second assembly of Tours-sur-Marne was to demonstrate that Charles was still king and willing to act as one, protecting the Church and rewarding loyalty.336 His diploma for the monks of Saint-Thierry issued at the same occasion showed him correcting old injustices.337 However, these measures probably failed to address a central part of the problem. Based on Charles' diplomas, Geoffrey Koziol has attempted to describe Charles' understanding of his position as a king. "He creates an image of monarchy […] in which a ruler stands 'gloriously' apart from and above margraves, listening graciously as they come before the royal majesty to present their entreaties, 'craving on bended knee'."338 This analysis is not unproblematic. First, the formula describing petitioners approaching Charles on their knees is used only on very rare occasions.339 Second, this language is also used by Charles' predecessors and hence nothing particularly special.340 And third, Koziol at the same time dismisses the more frequent references to consensus in Charles' diplomas341 as mere "elements in the display of royal majesty," arguing that for Charles they were simply "components of ceremonial of kingship that displayed a kingdom's proper order under a king's protecting, beneficent majesty."342 Accordingly, no one would have been able to claim the


right to counsel the king because of his birth or office—these positions were to be given by Charles and him only.343 Yet, as we have shown, Charles held contact with a large number of nobles throughout his reign, carefully treating them according to their ranks and integrating them into his rule.344 Hence, references to consensus need to be taken as serious expressions of the way Charles ruled. Nevertheless, Koziol's description of the image created in Charles' diplomas cannot simply be dismissed. His description of the image of distance between the ruler and the nobles depicted in the diplomas is accurate.

As Laurent Feller has observed, in 919 a new feature appears in Charles' diplomas which tied royal grants to the loyalty of the recipient.345 This is a feature that also appears in the diplomas issued at Tours-sur-Marne. The diploma issued for Bishop Guigo's church of Girona referred to the great loyalty shown by the bishop before granting him the remnants of the royal fisc.346 On the other hand, the Tours-sur-Marne diplomas lack any references to consensus. They only show Bishop Guigo approaching the king, to make requests to Charles' royal majesty.347 Even the restoration of property to the monks of Saint-Thierry saw only Charles deciding about the justness of their cause: "Acknowledging their claim to be true, through our power we return them said *villa* with everything belonging to it."348 Hence, Charles' message in these diplomas was clear: loyal service was to be rewarded, but providing this reward was only his own choice.

Hereby, he differed from Odo and his brother Carloman, who issued diplomas under similar circumstances.349 Odo's diploma from September 892, which noted the previous judgement against the monk Aginus, shows the king acting in front of and in accordance with his *fideles*. Abbot Erkanger of Montiéramey is said to have approached Odo "in the presence of our *fideles.*" Odo further confirms the judgement made by the bishops at the royal *placitum* of Verberie, again noting to have acted "with the consens of our *fideles.*"350 This is similar to Carloman's capitulary from 884, which was delivered at the moment of the reintegration of the nobles from the northern part of the realm and, while emphasising his position


as king,351 also shows him acting together with those around him.352 This point is further stressed by the diploma issued only a few days later, which restored the abbey of Saint-Sulpice to the church of Châlons at the request of its bishop, Berno. While the diploma emphasises Berno's loyalty in council and aid,353 it also informs that the king acted at the "suggestion and request" of two others of his *fideles*, Bishop Angelwinus of Paris and Count Theoderic of Vermandois, two of the leading nobles who had previously distanced themselves from his rule.354 Like Odo, Carloman too, facing a severe crisis of his rule, stressed his cooperation with the nobles. Compared to this, the message Charles sent with the diplomas issued at Tours-sur-Marne was most likely the wrong one.355 If he wanted to restore trust in him, he needed to demonstrate not only that he was willing to act as a king, but to act in accordance with those around him, by listening to their counsel. Just how central this point had become is further demonstrated by the only diploma preserved by Robert, issued in January 923. As Geoffrey Koziol has noted,356 this diploma addresses this exact same subject by replacing the common address "to the faithful of the church of God and of ourself " (*fideles eccelsiae Dei et nostri* ) with "Let it remain known to the sageness of all our faithful, that is, the leading men of the Franks…"357 In contrast to Charles, Robert thus portrayed himself as someone who cooperated with the nobles, listening to their sage counsel. This was undoubtedly a direct response to the former's failed policy.

Yet, despite Charles' shortcomings in addressing the reproaches brought against him by the nobles, questioning his right to rule appears to have been a difficult decision for the nobles. Robert's diploma, although it makes references to the rebellion by claiming that he had become king "because of indespensable reasons,"358 fails to go into any further detail and carefully avoids to name Charles at all, thus becoming "an embarrassed reflection on the circumstances that had brought Robert to the throne."359 Furthermore, some time after Charles

355 See also Görich, Misstrauen, who demonstrates that symbolic actions could fail in building trust when relations were dominated by mistrust and one of the parties acted under certain convictions which blocked the right choice of symbolic actions.

357 DRoI 1, 8: *Notum autem manet sagacitati omnium fidelium nostrorum, procerum scilicet Francorum, qualiter per divinam clementiam, causis necessariis existentibus, omnium favore principum ad regni gubernacula moderanda regie majestatis sceptra suscepimus.* Translation by Koziol, Robert, 250.

<sup>351</sup> See chapters II.2 and VI.3.

<sup>352</sup> MGH Capit. II, N° 287, 371: *Karlomannus gratia Dei rex omnibus venerabilibus episcopis, abbatibus, comitibus, iudicibus omnibusque sanctae Dei ecclesiae et nostris fidelibus. Cum ad palatium Vernis anno dominicae in carnationis DCCCLXXXIV, anno autem regni nostri quinto, indictione secunda, mense Martio convenissemus et pars fidelium nostrorum nobiscum, placuit, ut quaedam statuta sacrorum canonum necnon quaedam capitula antecessorum nostrorum renovarentur…*

<sup>353</sup> DCmII 76, 198: *…qui nobis in principio regni nostri et consilio et auxilio satis proficuus et fidelis extitit…*

<sup>354</sup> *…hoc idem quoque cum eo suggerentibus et deprecantibus fidelibus nostris, venerabili scilicet Angelwino Parisiorum episcopo et Teuderico comite valde nobis dilecto.*

<sup>356</sup> Koziol, Robert, 250−251.

<sup>358</sup> See chapter VI.4, n. 357. Own translation.

<sup>359</sup> Koziol, Robert, 249.

had been captured, a synod under the new archbishop Seulf took place at Reims, condemning the participants of the battle of Soissons to a harsh penitence of three years.360 Besides Seulf, the bishops Abbo of Soissons, Adelelm of Laon, Stephen of Cambrai, Adelelm of Senlis and the newly ordinated Airard of Noyon were present, thus a mixture of Robert's and Charles' former supporters,361 as well as legates from the other bishops of the archdiocese of Reims. Interestingly, the synod makes no differentiation between the men who had followed Charles and those who had followed Robert, treating them exactly the same way. Worries among the clergy about a new war among the nobility may indeed have played a role in the decision,362 yet, nevertheless, one might wonder why Robert's supporters—the very same who after his death still refused Charles and called for Raoul—would have agreed to be placed on the same level with their opponents. Like Robert's diploma, the notice reporting the synod refrains from bringing forward any reproaches against Charles. This silence points towards an underlying problem for the nobles opposing Charles. Whatever the accusations against him might have been, questioning his right to rule meant to rebel against the legitimate king. As Wojciech Falkowski's study shows, at least south of the Loire, the division among the nobility whether Charles' deposition had been legitimate or not continued to exist at least up until his death in 929.363 But even in the North the question of Charles' kingship remained to be a problem. When his captor Heribert II of Vermandois entered into conflict with King Raoul in 927, he used this kingship as a political lever, threatening Raoul's position by reinstalling Charles and letting the Northmen pay their allegiance to him.364 During the following year, although according to Flodoard under the pressure of the Northmen, other counts and bishops followed their example.365 The episode was only short-lived as Charles was returned to his prison by Heribert as soon as his dispute with Raoul had been settled,366 yet it reveals that the idea of Charles being king was still vivid enough to be exploited politically. What was the nature of this power behind Charles' claim to the throne? As we have argued, Charles' Carolingian descent had been a powerful argument when used against Odo who was lacking this kind of legitimacy. After Raoul's death in 936, it was Robert's son Hugh the Great who, instead of

360 On the penitence, see Schröder, Synoden, 215.

361 See chapters III.2.1.7, III.2.3 and III.3.


364 Flodoard, Annales 927, 39−40: *At Heribertus Karolum de custodia ejecit secumque in pagum Veromandinsem, scilicet ad Sanctum Quintinum, deduxit. […] Karolus igitur cum Heriberto colloquium petit Nordmannorum, ad castellum quod Auga vocatur, ibique se filius Rollonis Karolo committit et amicitiam firmat cum Heriberto.*

365 Flodoard, Annales 928, 41: *Filius tamen Heriberti, Odo, quem Rollo habebat obsidem, non redditur illi, donec se committit Karolo pater cum aliis quibusdam Franciae comitibus et episcopis.*

366 Flodoard, Annales 928, 43: *Hugo et Heribertus ad Heinricum, colloquii causa, proficiscuntur; indeque revertentes, obviam pergunt regi Rodulfo. Rursusque Heribertus se illi committit, iterum redacto sub custodia Karolo.*

taking the crown for himself, recalled Charles' own son Louis IV from England to succeed on his father's throne.367 The only argument speaking in favour of Louis was, very much like in his father's case, his Carolingian heritage. This was probably what Charles could rely on most—his undeniable claim, for the legitimacy of his kingship was based on his descent. The fact that the nobles nevertheless decided to oppose him is less of a sign that this idea had faded, but points towards the degree of the loss of trust in Charles.

#### **VI.5. Conclusion**

In this chapter, we have emphasised the role of two factors in the relations between the king and the nobles: the existence of powerful groups of influence and the necessity for the king to be trusted by the nobles who supported his rule. In conflict situations, either factor could limit the options of the king when deciding upon a strategy. Thus, Charles the Simple, when dealing with Count Baldwin II of Flanders, faced the same problem as his father Louis the Stammerer against Bernard of Gothia, for his closest allies prevented him from finding a peaceful arrangement with his opponent. In both of these cases, the nobles in question were rivals of those in conflict with the king, either for influence at the royal court or for the contested *honores*. Thus, it became impossible for the rulers to find a settlement with their opponents that did not at the same time violate the interests of their supporters. While in Louis the Stammerer's case the conflict was finally resolved by the superiority in resources the king and his allies were able to muster, leading to Bernard's deposition, in Charles' case it was the removal of the nobles blocking a solution that opened the way for the final settlement.

However, such a settlement required a certain degree of trust between all the parties involved. As the case of Odo and Baldwin shows, the lack of trust could have a similar effect as the presence of rivalling groups dominating the circle around the king. Under these circumstances, it was impossible to come to terms with each other, although this time not from the king's perspective, but from the perspective of the noble who appears to have deeply mistrusted Odo. There were ways to overcome such obstacles: for example, Charles the Simple chose to recompense Count Gislebert by finally elevating him to a position similar to his father Reginar's, something he had refrained from doing up until that moment. This offer must have been too tempting for Gislebert to resist, although he, after Charles had already broken an earlier agreement, had no reason to trust the king. Odo, on the other hand, finally resolved his conflict with Baldwin by a combination of military superiority and trust-building measures, which made the count finally admit to the royal prerogative to control the distribution of *honores*.

A loss of trust, such as it is visible in the case of Odo and Baldwin, was not *per se* necessarily problematic, at least not as long as it remained an isolated case. However, it could become an issue if the mistrust against the king spread within the nobility, especially if it reached the inner circle around the king with the most powerful nobles and his closest advisors, that is to say, when it developed into a severe crisis of trust. The decades around the turn from the 9th to the 10th century saw a number of such crises: when Carloman II suffered a disastrous defeat against the Northmen in 883, the nobles appear to have mistrusted his ability to deal with the Viking threat to a such degree that they took the matter into their own hands, although without denying Carloman's right to rule, still acknowledging him as king. On the other hand, in the cases of Odo and Charles the Simple, the crises of trust appear not to have evolved due to a lack of trust in their capacities to deal with a specific problem, but because of what the nobles perceived as transgressions against the norms of kingship. The conflicts we have analysed in this chapter either betray already existing mistrust of singular nobles against their king or were cases of royal actions that were probably perceived as "bad kingship" by wider circles of the nobility, undermining their trust in the rulers and all occurring around the same time the rebellions against the rulers were developing.

While this proximity in time is unproblematic in Odo's case since his conflict with Baldwin as well as the one arising over the succession at Poitiers368 predate the rebellion against him, in Charles' case the connection between his actions against Gislebert and the loss of trust in him causes some problems. While the conflict started before the deterioration of his relations with ever larger parts of the nobility, at least until the outbreak of the rebellion he appears to have had considerable support against the count. In any case, his last renewal of the conflict only took place when he was already fighting the rebels and thus cannot have contributed to the crisis. However, these actions are likely to have been symptomatic for Charles' behaviour also in other situations, betraying an obstinacy when it came to pursuing his interests against those of the nobles around him. This obstinacy can, for example, also be seen in his adherence to his intimate Hagano. After Soissons he did come to terms with the nobles and made efforts to publicly integrate Hagano into the noble hierarchy. The agreement they had found was a compromise, protecting the royal interest of keeping Hagano at court as well as the interest of the nobles, who would not accept an outsider close to the king. Yet, while Charles provided Hagano with the *honores* linked to an elevated position at court and thus aimed to integrate him further into the highest ranks of the nobility, he failed to address the second main issue the nobles had had with his intimate: his overly close relation with the king. In his persistence to keep Hagano close to him, Charles did not meet the expectations of the nobles arising from the previous agreement. Whether it were his actions against Gislebert or his policy in the case of Hagano, Charles

<sup>368</sup> For more examples of Odo's actions that probably were probably perceived as breaches of the norms, see chapter I.2.3.

failed to present himself as a reliable ruler who had the interests of the nobles at heart and thus undermined their trust in him.

Trust could be built by symbolic measures as in particular ways of doing justice. For example, Carloman II's second capitulary was issued after his defeat against the Northmen, just at the very moment when an agreement with nobles from the North had been reached. This is hardly a coincidence since at the same time he also issued a diploma recording a royal judgement in favour of one of his supporters—one example of how references to judicial assemblies were used by the various rulers to demonstrate the royal will to do justice in cooperation with the nobles. Finally restorational acts also fall within this category. While less frequent, if the occasion presented itself, like the references to the judicial assemblies, they were issued to portray how a ruler was bringing justice to those supporting him, remedying old injustice and thus strengthening trust in him. However, while the protection of the interests of those close to the ruler by royal justice was a characteristic common to the acts of all rulers, there was also a second characteristic in which Charles the Simple's acts differed from those of his predecessors. While Carloman and Odo emphasised their cooperation with the nobles, in his later acts Charles failed to so. The diplomas recording the judgements of Herstal, with their long lists of participants, are certainly reminiscences to the royal cooperation with the nobility. Yet, especially Charles' last restorational act issued at Tours-sur-Marne for the monks of Saint-Thierry of Reims, at a moment when the rebellion was already in full progress, is of a different nature. While still stressing the power of the king to bring justice, it focuses on the loyalty of the recipients to the king and not on his cooperation with them. In an atmosphere of deep mistrust against Charles, this was undoubtedly the wrong message. Trust could not be built by calling upon unconditional loyalty as a king's prerogative, but only by demonstrating the willingness to integrate the nobles into the rule and therefore to allow them a certain degree of control over the royal politics by their participation in the decision-making process, especially since this participation was perceived as one of their most important privileges by the leading nobles of the realm.

## **Conclusion**

Charles the Simple became king of the West Frankish realm because a strong opposition had formed against his predecessor Odo. The Robertian had, despite his efforts, failed to fully integrate his early opponents into his rule. Once Odo's reign started to suffer from a severe crisis of trust, these men acted as a rallying point for others. For this opposition, Charles possessed an advantage over Odo that made him an ideal candidate to replace the Robertian: his Carolingian blood. In a world in which the idea of Carolingian kingship was still prevalent, Charles' descent could be used as a powerful argument in the political discourse. This did not entail that Odo was to be disposed as a king. It merely meant that with Charles there now existed an alternative for all those discontent with the Robertian's rule. Both of them, Charles and Odo, were now kings, acknowledged by their peers as well as by the nobles, and even acknowledging each other. Thus, Charles' succession to Odo presents itself as a pragmatic solution to the conflict between the different parties, which took the fragile position of Odo and the weakness of a claim his brother Robert might have had to follow him on the throne into consideration.

Royal power being the result of the interaction between the king and the nobles, the king's ability to manipulate the network of nobles around him can be used as a means to measure the possibilities and limits of said power. Charles' early years reveal certain limitations to this ability: close to him were those who had already supported him during the years of the fight with Odo. However, at the same time, the presence of these nobles hindered the full integration of powerful men like Robert of Neustria or Richard the Justiciar. They did acknowledge Charles as king and cooperated with him initially, yet did not belong to his inner circle, which was dominated by those early supporters. This led to a fragile balance which ended only after the murder of Archbishop Fulk, and thereby the creation of a vacuum of power at the royal court that was filled first by Richard and then by Robert. A ruler not being able (or willing) to act against the interests of a powerful group of nobles was nothing new however. When his father Louis the Stammerer attempted to create his own basis of power after the death of Charles the Bald, the leading nobles of the realm forced him to abandon his plans and to rule with them instead. These men then created networks which dominated royal politics over the next few years, up until the reign of Charles the Fat.

This does not mean that the rulers were unable to promote nobles of their choosing. The presence of men like Count Robert of Troyes under Carloman II or the *fidelis* Robert under Charles the Simple indicates that there was room at court for men who drew their importance from their personal relation to the king. However, these men needed to be accepted by the other nobles already present in the circle around the king. In fact, in most cases nobles whose influence increased under the various rulers either appear to already have had connections to the circle dominating the court or soon created such contacts. This pattern is also revealed during the rule of Charles the Fat. The emperor certainly did make efforts to install men of his own choosing in the West Frankish realm, yet in most cases his choice fell on men like Odo and Bishop Geilo of Langres, both of whom had strong connections to the networks that had dominated politics under Louis the Stammerer and his older sons. This kind of policy, while it limited the pool of choices, also had distinctive advantages. These men were not only well connected among the local nobility and thus ideal candidates to serve as power brokers for the rulers, they were also well connected among those nobles already present at court, thus facilitating their integration into the hierarchy there.

This latter point was especially problematic. Not only Charles the Simple early in his reign, but also Odo had to cope with the problem of having to integrate different groups of nobles into his rule. Like Charles, Odo also was supported by a distinctive group of nobles, nobles who, after he had become king, occupied the innermost layer of the circle around him. Accepting new nobles into this circle meant that adjustments to the existing hierarchy had to be made, necessarily leading to conflicts between the interests of those already present and those newly joining. For both Odo and Charles, this posed a problem almost impossible to solve since their closest supporters already occupied the positions the men who were to be newly integrated into the rule claimed for themselves. Odo's efforts finally failed, contributing the rebellion breaking out against him. Charles was more successful, winning room for manoeuvre although only through the death of one of his most important supporters, Archbishop Fulk.

Indeed, for large parts of his reign, Charles the Simple demonstrated a remarkable capacity to mediate between the interests of the individual nobles and his own. The integration of the Lotharingian nobility meant that a completely new group of nobles—or better, several new groups—had to be integrated into the circle around the king. Charles not only managed to find a balance between these different groups, to deal with their rivalries and use them to his advantage when pursuing his goals within Lotharingia, but he also managed to do this without neglecting the interests of the West Frankish nobles at his court. Concerning the latter, he increased his efforts and kept especially Robert of Neustria close at his side, thus making sure that the necessary adjustments to the hierarchy at court were accepted by the West Frankish nobles and found the general consent. Similarly, when a group of new nobles rose to influence at the royal court around 920, Charles appears to have managed this transition without creating major discontent among the other nobles. Again, however, it is worth noting that most of the nobles added to the circle around the king on these occasions were already well established and often connected to those already present at the royal court. Nevertheless, adjustments to the existing hierarchy had to be made and Charles appears to have succeeded in implementing them in cooperation with the circle of nobles around him. Charles' capacity in these matters becomes especially evident when comparing his success with the problems his brother Carloman encountered once he had taken over the rule in the northern part of the realm. Early on, Carloman

failed to take the interests of these nobles into consideration and to make the necessary adjustments to the circle around him, which led them to withdraw from court and left Carloman with only very limited means to deal with the Viking threat.

If necessary, it was certainly possible to rule without the support or even against important nobles and their networks—at least for a certain time. In the early 900s, when Charles the Simple was faced with Robert's opposition, he managed to create a counterweight against the *marchio* by forging new alliances and extending his own basis of support. However, situations like this one were extremely unstable and needed a certain amount of political flexibility if new developments occurred that shifted the balance of power. In Carloman's case, his defeat against the Northmen brought the final change. In Charles', it was his conflict with Richard the Justiciar that made it necessary to finally satisfy Robert's claim to occupy a position at court adequate to his position within the noble hierarchy of the realm.

Situations like this were also extremely dangerous because they provided other rulers with the opportunity to intervene on the behalf of the ruler's opponents. Thus, Odo's own vulnerability in the conflict with Charles was exploited by Arnulf and Zwentibold while Charles' conflict with Count Gislebert was followed by activities of Henry the Fowler in Lotharingia. Of course, this also worked the other way round. Charles the Simple intervened twice in Lotharingia when parts of the local nobility were looking for allies against their ruler (or, in 911, against the possibility of a king from the rivalling Konradiner family) outside the borders of the *regnum*—the second time with success, due to the support he could muster among the local nobles. Louis the Stammerer had attempted to avert such situations by concluding a treaty with Louis the Younger which prohibited interventions of this kind. This worked rather well when soon after his death Boso had himself crowned king, which led to the creation of a system of alliances among the Carolingian rulers aimed at crushing the rebellion. However, at the same time, this same system of alliances made it impossible for Louis III and Carloman II to effectively pursue their interests in Lotharingia. Charles the Simple appears to have been well aware of the advantages and disadvantages of this system. When his rule started to become unstable towards the early 920s, he attempted to revive the old cooperation by initiating a new one with Henry the Fowler. While much less effective than the earlier alliances, this cooperation at least secured Henry's non-intervention after the rebellion had broken out against Charles.

However, while such situations of instability existed during certain periods, for most of the time Charles the Simple appears to have enjoyed the support of at least large parts of the nobility. This is especially evident from his policy towards the Northmen. Charles was, very much like his predecessors, able to muster large armies consisting of contingents from the various parts of his realm. However, this ability should not be confused with military success on the battlefield: large armies were as little guarantee for victory as small ones were for defeat. Nor should it be confused with the capacity to secure the realm solely by military means, as this was a task at which none of Charles' predecessors had succeeded. However, it does demonstrate that the nobles supported Charles' strategy of dealing with the Viking threat. This had not always been the case under his predecessors: after Carloman's disastrous campaign against the Northmen, the nobles around him enforced a diplomatic solution. Similarly, the local nobles allowed the Northmen to move freely in exchange for their own safety, thus undermining Odo's defensively oriented strategy which was aimed at containing the raiders rather than defeating them. The support of the nobles in Viking matters not only becomes evident in the military measures, but also in diplomacy. Negotiating deals with the Northmen was known as an effective means to remove them from the realm and it was frequently used by the West Frankish rulers. As such, Charles' treaty of Saint-Clair-sur-Epte followed their example, containing elements already used by his predecessors. Nevertheless, being concluded in the wake of major military victory, it marked a distinctive change in strategy which could only have been implemented in accordance with the leading nobles.

Charles was able to raise forces from different parts of the realm, yet this ability very much depended on the cooperation of a small number of nobles. Forces from Neustria and Burgundy, led by Robert of Neustria and Richard the Justiciar, the two men who controlled large parts of these regions, won the battle of Chartres. The existence of powerful nobles like these two was nothing new. Men like them had risen before through their connections with the rulers and by royal will and consent. Boso, to name but one example, had been one of them. Under Louis the Stammerer and his older sons, Hugh the Abbot and Gauzlin had dominated the politics of the realm together with their networks. As long as the rulers were able to secure the cooperation of these men, this was even an advantage since these powerful nobles served as perfect conduits of royal power. Thus, the decreased range of royal travels that can be noted for Charles the Simple's reign in comparison to his predecessors did not mean that the regions now not visited anymore were outside the range of royal power. Especially in Burgundy, previously a region with manifold royal contacts to the local nobility and repeated royal presence, Richard the Justiciar now served as an intermediary and power-broker for the king.

Yet, while Robert and Richard undoubtedly played a key role in Charles' reign, his dependency on them appears to have been much smaller than those of Louis the Stammerer and his sons on the alliances around Hugh the Abbot and Gauzlin. The circles around Charles' father and brothers appear to have consisted almost exclusively of members of these alliances, with a strong tendency to also absorb those who the kings themselves had elevated into their networks. This changed over the course of Charles the Fat's reign, leading to a politically fragmented nobility which supported different claimants to the throne after his death. This condition persisted over the course of Odo's reign into the one of Charles the Simple, contributing to the rebellion against the former and causing Charles' initial problems with Robert. Yet, at the same time, at least for Charles this fragmentation also provided him with more room for manoeuvre after the death of Archbishop Fulk. His power was not solely dependent on the cooperation of Robert or Richard (though their cooperation greatly increased his options), it was also and mainly founded on a network of less powerful nobles, first from the Francia, then also from Lotharingia. Here, Charles forged links by marriage and patronage. Wigeric certainly owed his promotion to count of the palace to his marriage to Charles' niece, as did Count Erlebald, who married one of Charles' daughters. Richer of Prüm and Charles' former notary Gauzlin became bishops because Charles aimed at promoting men of his choice to important positions. Bishop Stephen of Cambrai and Count Ricuin of Verdun regained influence at the royal court which they had lost under Louis the Child. These ties helped Charles to keep the balance of power within the realm.

In the end, Charles was deposed because he failed to maintain the good relations not only with the most powerful men of the realm, but also with those other nobles. At the core of this development was a crisis of trust in Charles, whose actions in the late 910s started to transgress the boundaries of what was politically possible and opportune. Charles, like Odo in 892, was not trusted anymore because he violated the interests of the nobles while pursuing his own. This loss of trust went deep, estranging even those nobles from him who had previously belonged to his most important supporters. This situation was exacerbated by another development. In contrast to the early years of his reign, the fragmentation of the leading ranks of the nobility had been overcome. While this did not mean that another system of alliances had been created that dominated royal politics, it did mean that the old rivalries that had shaped politics within the realm around 900 had come to an end. Over the course of Charles' reign, Robert, who served as rallying point for the rebels, had forged links to other leading nobles, including the family of Richard the Justiciar and Count Heribert II. These links now became active, drawing more and more of Charles' supporters over to the rebels.

At first, the rebellion was certainly not aimed at Charles' deposition and Robert's elevation to the throne. Charles' claim to the throne was still a strong one and the question of whether to deny his right to rule deeply divided the nobility. Indeed, it seems much more likely that the goal of the opposition was to take control of the royal politics, with Charles remaining king. It was Charles' own refusal to negotiate that initiated the escalation of the conflict. This refusal was in line with his persistence to pursue his own interests to the detriment of those of the nobles during the last years of his reign, even if it meant bending or breaking earlier agreements—a persistence that led to the creation and intensification of the crisis of trust in him. The possibilities and limits of royal power in the late Carolingian age were defined by the cooperation of the nobles and Charles, by neglecting this need to create consensus while overemphasising his royal will and majesty, deprived himself of the foundations of his rule.

## **Bibliography**

#### **Primary Sources**


#### **Secondary Sources**

Abadal i de Vinyals, Ramón d', Els primers comtes Catalans, Barcelona 1958.


Des Mérovingiens aux premiers Capétiens (Variorum Collected Studies Series, 351), Aldershot 1991, part V, 41−68.


Bougard, François, Laurent Feller and Régine Le Jan (eds.), Les élites au haut Moyen Âge. Crises et renouvellements (Collection Haut Moyen Âge, 1), Turnhout 2006.


Deutschland, Frankreich und Italien vom 6. bis zur Mitte des 14. Jahrhunderts (Kölner historische Abhandlungen, 14), Cologne 1968.


Ehlers, Joachim, Geschichte Frankreichs im Mittelalter, Stuttgart 1987.


Ergebnisse des gleichnamigen Kolloquiums auf der Moritzburg bei Dresden, 27. bis 29. September 2007 (Vita curialis, 2), Berlin 2009, 193−220.


www.perspectivia.net/publikationen/bdf/atsma\_neustrie\_1/kaiser\_royaute (accessed 25.09.2018).


supérieures de Civilisation médiévale, 18–20 novembre 2004 (Culture et société médiévales, 18), Turnhout 2009, 301−322.


Bührer-Thierry, Régine Le Jan and Vito Loré (eds.), Acquérir, prélever, contrôler. Les ressources en compétition (400−1100), Turnhout 2017, 207−226.


1986, 45−64. Reprinted in: Janet L. Nelson, The Frankish World 750–900, London 1996, 51−74.


(Münchner Historische Studien. Abteilung Mittelalterliche Geschichte, 6), Kallmünz i. d. Oberpfalz 1994.


Beiträge zur Kultur- und Identitätsumbildung in der Völkerwanderungszeit, Brno 2007, 29−38.


Oktober 1994) anläßlich der 1200-Jahrfeier der Stadt Frankfurt am Main (Quellen und Abhandlungen zur mittelrheinischen Kirchengeschichte, 80), vol. 2, Mainz 1997, 453−518.


## **Index**

#### A

Abbo, bishop of Nevers 89, 94, 319 Abbo, bishop of Soissons 161, 177, 179, 185, 186, 187, 188, 200, 203, 220, 224, 226, 227, 228, 230, 324, 333, 342 Abbo of Saint-Germain 102, 276, 310 Acfred, abbot of San-Esteban of Bagnoles 176, 319 Actard, bishop of Tours 85 Adalard, archbishop of Tours 109, 136 Adalard, count of the palace 22, 23, 79 Adalard the Seneschal 23 Adalbero, bishop of Augsburg 167 Adalbero, bishop of Laon 282 Adalelm, uncle of King Odo 306 Adalgar, bishop of Autun 46, 120, 210 Adalgard, abbess of Sainte-Croix of Poitiers 101 Adalgarius, bishop of Autun 59, 60, 82, 85, 86, 89, 90, 94, 95, 100, 115, 117, 118, 120, 123, 210, 324, 325, 331, 332 Adalung, count 49, 50, 65 Adelaide, countess 151, 152, 195, 220 Adelaide, queen, wife of Louis the Stammerer 2, 22, 23, 24, 25, 26, 27, 29, 34, 37, 45, 48, 50, 60, 64, 70, 71, 83, 86, 88, 94, 147, 156, 186, 188, 191, 192, 193, 196, 197, 213, 222, 223, 224 Adelaide, wife of Richard the Justiciar 119, 120 Adelelm, bishop of Laon 228, 342 Adelelm, bishop of Senlis 342 Adelelm, count of Arras 177 Adelelm, count of Troyes 105 Ademar of Chabannes 116, 309, 310 Aduvirus, abbot of Notre-Dame of Amer 117 Ælfthryd, daughter of Alfred the Great 197 Aeneas 284 Æthelwulf, king of Wessex 198 Agambert, bishop 138 Agelmar, bishop of Chalon-sur-Saône 106 Aginus, monk 42, 321, 340 Agio, archbishop of Narbonne 138

Agnes, empress 26 Ahasuerus/Xerxes, Persian king 195 Aimo, abbot 213 Airard, bishop of Noyon 342 Alan the Great 216 Alardus, count 101 Alberich, count 163, 166 Albert of Vermandois 282 Albrada 49 Albuin, count 87 Aledramnus, count, brother of Theoderic of Vermandois 41, 48, 50, 51, 86, 92, 104, 106, 108, 117, 123, 153, 170, 198, 271, 273 Alfred the Great, king of Wessex 197, 261, 263 Almanus, count 215 Alpais, daughter of Charles the Simple 199, 200 Alpais, daughter of Louis the Pious 22, 23 Altmar, count of Arras 112, 152, 153, 157, 159, 195, 215, 303, 304 Altmar, count of Poitiers 52, 58, 59, 60, 112, 113, 115, 116, 117, 118, 120, 305, 309, 310, 311, 312, 321 Angelrannus, bishop of Meaux 160 Angelwinus, bishop of Paris 46, 97, 98, 99, 341 Anne, wife of the fidelis Stephen 192 Ansbald, abbot of Prüm 88 Ansegisus, archbishop of Sens 26, 85, 86, 301 Ansgar, count of Oscheret 85, 94, 99, 100, 101, 105, 106, 115, 123 Ansgarde, queen, wife of Louis the Stammerer 22, 27, 28, 29, 70, 100, 147 Anskeric, bishop of Paris, archchancellor 46, 47, 50, 51, 53, 68, 69, 71, 107, 109, 113, 115, 117, 151, 152, 155, 156, 159, 161, 179, 187, 195, 204, 206, 213, 223, 245, 309 Aquinus, provost of Flavigny 61 Argrim, bishop of Langres 42, 47, 113, 115, 121, 162, 209, 219 Arnaldus, bishop of Toul 88

Arnulf, duke of Bavaria 257 Arnulf I, count of Flanders 145, 197 Arnulf of Carinthia, emperor 16, 21, 29, 32, 34, 35, 36, 37, 38, 40, 45, 48, 49, 50, 51, 53, 54, 55, 56, 62, 63, 64, 65, 66, 67, 68, 69, 71, 72, 73, 114, 146, 147, 148, 151, 163, 166, 167, 172, 240, 241, 242, 243, 244, 245, 246, 248, 258, 259, 260, 274, 321, 322, 328, 349 Arnulf of Metz 252 Arnulf, son of Charles the Simple 252 Arnust, archbishop of Narbonne 142, 189, 192 Arveus, royal chaplain 93, 94 Atto, viscount 136 Audacher, royal notary 92 Aurelian, archbishop of Lyon 41, 42, 85, 86, 89, 90, 331, 332 Ava, abbess of Sainte-Croix 88 Ava, abbess, sister of William the Pious 60

#### B

Baldric, bishop of Utrecht 182 Baldwin II, count of Flanders 35, 36, 40, 42, 43, 44, 48, 49, 51, 52, 53, 54, 55, 56, 57, 58, 60, 62, 63, 65, 69, 71, 73, 75, 114, 118, 121, 122, 150, 151, 153, 156, 157, 160, 197, 240, 241, 247, 277, 281, 297, 302, 303, 304, 305, 306, 307, 308, 310, 321, 329, 338, 339, 343, 344 Baltram, bishop of Strasbourg 248 Bathilde, mother of Clothar III 156 Beatrix, daughter of Heribert I of Vermandois 213 Bego, count of Paris 22, 23 Bera, abbot of Saint-Chinian 138 Berengar, archbishop of Narbonne 140 Berengar, bishop of Cambrai 193, 194 Berengar, bishop of Verdun 193 Berengar, count 170, 171, 176, 180 Berengar, count of Le Mans 73 Berengar, father or brother of Poppa 287 Berengar I, emperor 5, 32, 256, 257 Bernard, deacon of Saint-Julien of Brioude 62 Bernard, king of Italy 50 Bernard of Auvergne, count 79, 80, 83, 84, 85, 90, 94, 95, 100, 104, 106, 121, 137, 206

Bernard of Gothia, count 80, 82, 83, 84, 85, 89, 121, 235, 264, 299, 300, 301, 302, 343 Bernard, otherwise unknown 327 Bernard, son of Gauzlin 301 Berno, bishop of Châlons 87, 91, 97, 98, 105, 123, 327, 341 Berno, bishop of Orléans 213 Berno, priest 321 Bernuin, bishop of Vienne 42 Berta, daughter of Charlemagne 146 Betto, bishop of Auxerre 135 Bilechildis, sister of Gauzlin 301 Biltgerius, abbot of Saint-Philibert of Tournus 115 Björn, Viking leader 282 Bligardis, otherwise unknown 318 Bonifatius, abbot of Sainte-Seine 110 Boso of Vienne, count, dux, king of Provence 16, 24, 25, 27, 33, 36, 37, 38, 66, 70, 79, 80, 82, 83, 84, 85, 89, 90, 91, 94, 95, 100, 115, 119, 121, 133, 158, 163, 171, 234, 238, 239, 240, 260, 264, 266, 298, 320, 324, 325, 330, 331, 332, 333, 349, 350 Boso, son of Richard the Justiciar 176, 179, 208 Bovo, bishop of Châlons, brother of Frederuna 161, 177, 185, 189, 190, 193, 194, 196, 197, 198, 200, 202, 324 Bovo I, abbot of Corvey 35 Bovo II, abbot of Corvey 194 Bovo, monk at Corvey 194 Burchard, count 248 Burchard, duke of Swabia 257

## C

Carloman, king 144 Carloman of Bavaria, king 16, 26, 32, 146, 235, 252, 264 Carloman II, king of the West Frankish realm 2, 5, 6, 19, 22, 24, 25, 26, 27, 28, 30, 31, 32, 36, 38, 39, 40, 47, 70, 82, 83, 85, 87, 90, 91, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 104, 105, 106, 115, 116, 121, 123, 128, 132, 133, 134, 135, 138, 139, 140, 143, 144, 147, 149, 153, 191, 234, 236, 237, 238, 239, 240, 253, 260, 263, 264, 265, 266, 267, 268, 269, 270, 271, 290, 291,

292, 293, 297, 298, 299, 316, 317, 319, 320, 324, 325, 326, 327, 329, 331, 332, 333, 340, 341, 344, 345, 347, 348, 349, 350 Charlemagne 6, 26, 130, 143, 144, 145, 146, 174, 200, 201, 248, 252, 254, 295, 334 Charles Martel 146 Charles, son of Charlemagne 146 Charles II the Bald, king of the West Frankish realm, emperor 1, 4, 6, 11, 12, 15, 19, 22, 23, 26, 33, 38, 39, 48, 70, 77, 79, 80, 81, 85, 86, 87, 88, 92, 93, 96, 109, 114, 119, 121, 122, 128, 130, 132, 134, 135, 138, 142, 143, 144, 146, 147, 153, 156, 158, 176, 188, 192, 195, 198, 210, 234, 236, 249, 252, 253, 259, 262, 263, 279, 289,


Conrad I, king of the East Frankish realm 3, 142, 168, 171, 172, 173, 249, 250, 251, 254

Cunigunde, daughter of Ermentrude and Reginar Longneck 49, 171, 180, 181, 191, 199, 313

#### D


#### E


Egfred, bishop of Poitiers 112, 311 Eldefred, provost of Saint-Julien of Brioude 62 Eldegarde, daughter of Count Ermenfrid 199 Elduin, count of the palace 47, 50, 62, 207 Elefons, bishop 138 Emeno, count 81, 82, 84, 264, 299, 301, 309 Emma, abbess of San-Juan-de-las-Abadesas 139 Emma, daughter of Robert of Neustria 210, 219, 227 Emma, noble woman 325 Engelberga, wife of Emperor Louis II 238 Erbernus, archbishop of Tours 213 Erifons, fidelis 100, 101 Erkanger, abbot of Montiéramey 340 Erkanger, count of Boulogne 46, 47, 48, 50, 54, 104, 152, 159, 177, 179, 192, 195, 204 Erlebald, count of Castriciensis 176, 185, 200, 215, 319, 351 Ermemirus, bishop of Girona 116 Ermenfrid, count of Amiens 156, 170, 177, 179, 192, 199, 203 Ermengard, daughter of Emperor Louis II, wife of Boso of Vienne, mother of Louis the Blind 38, 119, 120, 330 Ermentrude, daughter of Charles the Simple 200, 224 Ermentrude, daughter of Louis the Stammerer and Adelaide 49, 147, 191, 199, 201 Ermentrude, wife of Charles the Bald 23, 191, 210 Ernustus, royal notary 154, 159, 215 Esther, Jewish queen 195 Etbert, fidelis 196, 204

#### F

Flodoard of Reims 2, 3, 43, 110, 120, 139, 157, 160, 161, 179, 194, 200, 201, 202, 203, 205, 213, 217, 218, 219, 221, 222, 225, 226, 227, 229, 230, 241, 280, 282, 285, 287, 288, 290, 292, 313, 315, 335, 336, 342

Folcwin, abbot of Lobbes 336

Everbert 48, 304, 305

Formosus, pope 43, 45, 46, 61, 66, 67, 68, 72, 241, 243 Franco, abbot of Corbie 156 Franco, bishop of Liège 64 Franco, bishop of Nevers 118 Frederuna, queen, wife of Charles the Simple 129, 151, 183, 189, 191, 192, 193, 194, 195, 196, 197, 198, 199, 201, 202, 203, 204, 214, 224, 225, 324, 338 Frodoynus, bishop of Barcelona 89 Frotar, archbishop of Bourges 82, 84, 85, 86, 93, 94, 99, 101, 114, 116, 123, 299, 300, 301 Fulbert, bishop of Cambrai 187 Fulcher, bishop of Nantes 213 Fulk, archbishop of Reims, archchancellor 2, 21, 29, 31, 32, 34, 35, 36, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 74, 75, 96, 110, 114, 120, 121, 122, 150, 151, 153, 154, 155, 157, 159, 160, 163, 173, 187, 188, 190, 193, 206, 213, 215, 222, 224, 233, 240, 241, 243, 244, 247, 250, 280, 291, 292, 298, 303, 304, 305, 306, 311, 321, 333, 336, 338, 347, 348, 351 Fulk the Red, viscount 136, 213, 217

#### G

Garnegaud, viscount of Blois 111 Gausfred, possibly father of the counts Gauzbert and Gauzlin 137 Gauzbert, brother of Abbot Ebolus 52, 56, 309, 310, 311 Gauzbert, count 137 Gauzfrid 51 Gauzfrid, bishop of Strasbourg 189, 198, 250 Gauzfrid, count, brother of Gauzlin 49, 50, 81, 84, 264, 299, 300, 301, 305 Gauzlin, abbot of Saint-Germain, bishop of Paris, archchancellor 25, 26, 27, 28, 30, 31, 33, 36, 37, 39, 46, 49, 71, 80, 81, 82, 83, 84, 85, 86, 88, 89, 90, 91, 92, 93, 96,

97, 98, 99, 100, 101, 102, 103, 104, 106, 107, 108, 109, 110, 113, 121, 122, 123, 215, 236, 237, 238, 264, 266, 269, 271, 272, 279, 293, 299, 301, 350

Gauzlin, count of Maine 73, 137, 152 Gauzlin, royal notary, bishop of Toul 189, 196, 336, 351 Gebhard, count 249 Gebhard, count, dux of Lotharingia 165, 167, 168, 169, 249 Geilo, abbot of Saint-Philibert of Tournus, bishop of Langres 35, 83, 85, 89, 90, 94, 95, 105, 106, 113, 115, 122, 123, 325, 326, 331, 332, 348 Genesius, otherwise unknown 318 Geoffrey, otherwise unknown 327 Gerald of Aurillac, count 116, 137, 170, 206, 207, 310, 311 Geran, bishop of Auxerre 135, 209, 278 Gerard, count of Vienne 23 Gerberga, queen, sister of Otto I, wife of Louis d'Outremer 195 Gerhard, count 164, 166, 167, 168, 200, 201, 224 Gerulf, abbot of Beaulieu 116 Gerulf, count 171 Geruntius, archbishop of Bourges 137 Gibert, bishop of Nîmes 318, 319 Girart de Roussillon 23 Girbald, count 119 Gisela, abbess of Nivelles, daughter of Lothar II 151, 195, 289 Gisela, daughter of Charles the Simple 200, 284, 287, 289 Gisela, daughter of Louis the Pious 256 Gisela, daughter of Louis the Stammerer and Ansgarde 100, 147, 191 Gislebert, father of Reginar Longneck 49 Gislebert, otherwise unknown 73 Gislebert, son of Reginar Longneck, dux of Lotharingia 3, 49, 129, 139, 175, 176, 177, 180, 181, 182, 183, 184, 188, 190, 221, 223, 229, 230, 231, 255, 257, 260, 312, 313, 314, 315, 322, 333, 343, 344, 349 Goiram, count 85 Gottfrid, count of the palace 200, 201, 224 Gottfrid, Viking leader 283, 289, 290 Gregory, abbot of Saint-Martin of Autun 106 Grimoard, viscount of Paris 156 Gualdricus, bishop of Auxerre 135 Guarnegaud, viscount 136, 213

Guarner, count 170, 171, 187 Guigo, bishop of Girona 141, 189, 229, 340 Gunnor, duchess of Normandy 282 Guntbert, otherwise unknown 314 Gunter, bishop of Le Mans 73 Guntram, count 258 Guthrun, Viking leader 263

#### H


Heilwich, daughter of Eberhard of Friuli 43, 177

	- archchancellor 3, 11, 43, 154, 155, 157, 159, 160, 161, 162, 174, 178, 183, 185, 187, 188, 203, 219, 221, 223, 224, 225, 226, 227, 280, 285, 292, 304, 318

Heriveus, count 137


## I

Imo, son of Wulfard and Susanna 23 Isaac, bishop of Langres 41 Isaac, count of Cambrai 169, 171, 186, 187 Isaias, bishop of Nantes 217

## J

John, archbishop of Rouen 35 John VIII, pope 23, 70, 82, 233, 234, 264, 300 John X, pope 190, 256, 285 Judith, daughter of Charles the Bald 198 Judith, empress, wife of Louis the Pious 93 Judith, queen, wife of Charles the Bald 218

## L

Lambert, bishop of Le Mans 107 Lambert, bishop of Liège 328 Lambert, bishop of Mâcon 89, 331 Lambert of Spoleto, emperor 43, 45, 61, 66, 67, 68, 69, 245, 256 Letard, bishop of Mâcon 110 Letard, count 170, 171, 187 Leuthard, count of Paris 23 Liedgardis, possible name of Heribert II's wife 220 Liutgard, queen, wife of Louis the Younger 236 Liutpert, archbishop of Mainz 35 Liutward, bishop of Vercelli 104, 105, 334 Lothar I 49, 138 Lothar II 23, 151, 195, 238, 239, 249, 266, 289, 319, 332 Louis the Pious, emperor 6, 93, 146, 174, 200, 248, 252, 256, 334 Louis II, emperor 80, 233, 238, 332 Louis II the German, king of the East Frankish realm 29, 33, 200, 252 Louis II the Stammerer, king of the West Frankish realm 2, 4, 6, 19, 21, 22, 23, 24, 25, 26, 27, 28, 29, 34, 37, 38, 39, 48, 70, 71, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 96, 99, 100, 102, 111, 115, 116, 121, 122, 123, 128, 132, 134, 135, 138, 139, 143, 144, 146, 147, 149, 152, 198, 233, 234, 235, 236, 237, 240, 260, 263, 264, 265, 267, 278, 281, 299, 300, 301, 305, 317, 324, 325, 330, 331, 332, 333, 343, 347, 348, 349, 350


## M

	- Auvergne 26, 95, 97

#### O

	- Saxony, wife of Zwentibold 164, 167

## P

Patericus, vassus 211 Philippe-Auguste, king of France 199 Pippin, brother of Heribert I of Vermandois 41, 45, 48, 50, 51, 68 Pippin, father of Heribert I of Vermandois 46 Pippin, son of count Rudolf 105, 107, 123 Pippin the Hunchback, son of Charlemagne 146 Pippin the Short, king of the Franks 143, 144 Poppa, mother of William Longsword 287

## R

Radulf, count 119 Ragenar, viscount of Auxerre, nephew of Richard the Justiciar 120, 121 Ragenold, dux of Maine 103, 106, 107, 122, 271, 283 Ragenold, Viking leader 288 Raginelm, bishop of Noyon 91 Ragnar, Viking leader 282 Rainald, viscount 136, 213 Rainardus, fidelis 94 Raino, abbot of Saint-Serge of Angers 136 Raino, bishop of Angers 111, 213 Ramnulf I, count of Poitiers 106 Ramnulf II, count of Poitiers 34, 36, 51, 52, 58, 112, 113, 114, 116, 118, 123, 274, 309, 310 Rampo, nephew of Richard the Justiciar 120 Rampo, relative of Fulk of Reims 61 Raoul, count of Ivry 282 Raoul de Gouy, count of Ostrevent 176, 177, 185, 187, 196, 199, 200, 203, 204, 220, 224, 324 Raoul, king of the West Frankish realm 3, 186, 208, 209, 210, 219, 224, 227, 228, 273, 288, 337, 342 Ratbert, bishop of Valence 94, 95 Ratbod, archbishop of Trier, archchancellor 145, 164, 165, 166, 169, 173, 174, 178, 183, 248 Ratbod, bishop of Utrecht 172 Rather, count of Nevers 137 Raymond, count of Nîmes and Albi 137, 138, 207, 223, 318

Reginar Longneck, count, marchio 48, 49, 50, 51, 164, 165, 166, 167, 168, 169, 170, 171, 173, 174, 175, 176, 177, 180, 181, 182, 183, 187, 188, 190, 199, 204, 216, 223, 229, 233, 245, 246, 247, 248, 249, 250, 255, 283, 312, 322, 343 Regino, abbot of Prüm 2, 22, 24, 26, 30, 32, 44, 45, 166, 245, 246, 247, 255, 265, 270, 283, 330, 331, 332, 334 Ricbert, otherwise unknown 320, 321 Richard I, duke of Normandy 282 Richard II, duke of Normandy 282 Richard the Justiciar, count, marchio 47, 49, 54, 58, 59, 60, 61, 62, 63, 75, 94, 95, 112, 115, 119, 120, 121, 124, 130, 134, 135, 137, 150, 151, 152, 154, 155, 157, 158, 162, 169, 170, 171, 176, 188, 192, 193, 204, 205, 206, 207, 208, 209, 210, 212, 213, 214, 216, 219, 220, 222, 223, 227, 258, 273, 277, 278, 279, 311, 325, 329, 337, 338, 347, 349, 350, 351 Richer, abbot of Prüm, bishop of Liège 167, 180, 182, 190, 201, 256, 313, 322, 351 Richer, count 215 Richer of Saint-Remi 1, 3, 160, 181, 202, 255, 312, 313, 318, 337 Richgard, empress, wife of Charles the Fat 105 Richilde, empress, sister of Boso and Richard the Justiciar, wife of Charles the Bald 80, 119, 158, 176, 191, 252, 332 Ricuin, bishop of Strasbourg 189, 250 Ricuin, count of Verdun 170, 172, 180, 181, 185, 199, 223, 229, 257, 313, 314, 351 Riculf, bishop of Soissons 41, 42, 49, 160, 321 Robert, archbishop of Rouen 282 Robert, bishop of Metz 173, 248 Robert, bishop of Noyon 177, 215 Robert, count of Troyes 100, 101, 105, 123, 147, 204, 347 Robert, fidelis 152, 173, 195, 204, 208, 347 Robert of Neustria, count, marchio, king of the West Frankish realm 2, 3, 5, 33, 50, 52, 64, 73, 74, 102, 108, 111, 112,

Regembald, abbot of Psalmody 207

113, 117, 118, 121, 123, 130, 135, 136, 137, 149, 150, 151, 152, 153, 154, 155, 156, 157,

158, 159, 160, 161, 162, 163, 169, 170, 171, 173, 176, 177, 178, 179, 180, 182, 183, 184, 186, 187, 188, 190, 192, 193, 195, 197, 199, 202, 204, 205, 206, 207, 208, 209, 210, 211, 212, 213, 214, 215, 216, 217, 218, 219, 220, 221, 222, 223, 224, 225, 226, 227, 228, 229, 230, 231, 256, 258, 259, 273, 277, 278, 279, 286, 287, 291, 292, 293, 298, 304, 307, 308, 310, 311, 314, 321, 323, 324, 325, 328, 333, 335, 336, 337, 338, 339, 341, 342, 347, 348, 349, 350, 351 Robert the Strong, count 32, 33, 106, 107, 108, 109, 262, 279 Rodulf, abbot of Conques 137 Rodulf, abbot of Saint-Bertin and Saint-Vaast 35, 36, 40, 43, 47, 48, 52, 56, 57, 114, 122, 215, 240, 275, 292, 297, 304, 305 Rodulf, brother of Baldwin II of Flanders 51, 54, 56, 57, 60, 62, 65, 69, 73, 302, 306, 307, 308 Roger, archbishop of Trier, archchancellor 161, 178, 183, 184, 196, 224, 231, 312, 315 Roger, candidate for the episcopal see of Beauvais 92 Roger, count of Laon 228 Roger of Maine 73, 118, 136, 152, 220 Rollo, Viking leader, princeps 3, 5, 200, 283, 284, 285, 286, 287, 288, 289, 291 Rorico, father of Gauzlin 103 Rorico, son of Charles the Simple 252 Rothaid, daughter of Charlemagne 146 Rothard, vasallus et judex 327 Rothild, sister of Louis the Stammerer, abbess of Chelles 73, 152, 176, 205, 218, 220, 221, 335 Rotrude, daughter of Charlemagne 146 Rotrude, daughter of Charles the Bald, abbess of Andlau 172 Rudolf, bishop of Laon 46, 153, 154, 159, 160, 186, 192, 215 Rudolf, count 105, 107, 123 Rudolf, count, marchio, king of Upper Burgundy 61, 63, 66, 68, 69, 119, 163, 245, 256 Rudolf II, king of Upper Burgundy 258 Rumald, abbot of Fossés 186, 203

#### S


#### T


Theodosius, otherwise unknown 139 Theodrada, queen, wife of Odo 33, 40, 48, 92, 108, 114 Theodulf, bishop of Paris 159, 179, 187, 223 Theodulf, scribe 145 Theotarius, bishop of Girona 94, 110 Theuderic III, king of the Franks 323 Theutberga, empress, aunt of Boso and Richard the Justiciar, wife of Lothar II 332 Thietmar of Merseburg 258 Throannus, royal notary, bishop of Orléans 111 Thurcetel, Viking leader 197, 280

#### U

Uurmaelon, Breton count 216

#### W

Wala, bishop of Auxerre 87 Waldrada, concubine of Lothar II 289 Walker, count 184 Walker, nephew of Odo 41, 52, 53, 115, 306, 317 Walo, bishop of Autun, nephew of Richard the Justiciar 61, 120, 121 Walter, archbishop of Sens 40, 41, 42, 51, 61, 91, 109, 112, 113, 118, 120, 121, 321 Walter, bishop of Orléans 85, 94, 99, 109, 112, 123 Waltger, count 171 Warnulf, bishop 104 Welf, abbot of Saint-Riquier 93 Wibald, bishop of Auxerre 87 Wicheramnus, abbot of Saint-Philibert of Tournus 176, 209 Wido, archbishop of Rouen 41, 42, 160, 285, 321

Wido, count 94, 99, 119, 120, 123, 124

Wido, count of Senlis 287 Wido, noble from Burgundy 94 Wido of Spoleto, emperor 29, 35, 36, 37, 40, 45, 55, 66, 67, 68, 71, 94, 105, 113, 114, 115, 119, 122, 124, 305 Widukind of Corvey 258 Wifred, count 139 Wifred of Cerdagne, archbishop of Narbonne 140 Wigeric, bishop of Metz 231 Wigeric, count of the Bidgau, count of the palace 163, 168, 171, 180, 181, 198, 199, 224, 351 Willebertus, bishop of Châlons 87 William, abbot of Croix-Saint-Ouen 100 William Longsword, son of Rollo 285, 287, 288 William the Pious, count, marchio 47, 58, 59, 60, 62, 75, 100, 112, 115, 116, 118, 119, 120, 137, 138, 150, 159, 205, 206, 207, 209, 210, 219, 222, 223, 224, 310, 311 William the Younger, nephew of William the Pious, count 138, 207, 208, 210, 219 Wulfard 23 Wulfard, abbot of Flavigny, archchancellor 23, 26, 27, 83, 86, 94, 97, 123 Wulfard, count of Angoulême 23 Wulfard, father of Wulfard and Adalard 23 Wulfgrim, son of Wulfard and Susanna 23

#### Z

Zwentibold, king of Lotharingia 41, 48, 49, 50, 54, 55, 56, 62, 63, 65, 66, 68, 69, 72, 146, 148, 151, 153, 164, 165, 166, 167, 168, 172, 173, 174, 175, 180, 223, 233, 242, 244, 245, 246, 247, 248, 250, 252, 260, 303, 306, 307, 321, 334, 349

**The prevalent image of the late Carolingian age is one of decline and fall. Charles III the Simple's (893/898−923) rule, which has hardly received any scholarly attention since the late 19th century, is perceived to have been the classic example of this development. Enthroned by rebels as well as cast down by a rebellion he is said to have been a weak ruler, powerless in the face of the ambitions of the nobles of the West Frankish realm. Yet, what do "weak" and "powerless" mean? In modern scholarship, early medieval rule is understood not as a question of command and obedience but as the result of cooperation between rulers and nobles. Thus, royal actions, such as the defence of the realm against the Northmen, interactions with other rulers or in regard to conflicts with or between the nobles, are reflections of the relations between the ruler and the circle of nobles around him. A ruler's power therefore depended on his ability to integrate the most powerful nobles into his rule, to mediate between their interests and to create consensus over the course of action. Based on this view, a new assessment of Charles the Simple's rule, the circle of nobles around him, the actions taken by him and thus his royal power is provided in this study, with the rules of his predecessors since the death of Charles the Bald in 877 serving as a basis for comparison.**

*Horst Lößlein* **prepared his PhD in a cotutelle at the Université de Limoges and the Ludwig-Maximilians-Universität München. His research interests include early medieval kingship, conflicts and norms, diplomatics, ideas and the meaning of legitimacy as well as literacy in the early and high middle ages. He is currently working as a research assistant at the University of Hamburg.**

WWW. HUMANITIES-MAP. NET