The politics of being Norman in the reign of Richard the Fearless, Duke of Normandy (r. 942–996)

In 966, by the end of the reign of its third duke, Richard I, Normandy had overcome the crises that had beset it in the middle of the century. Much of this success came from the coherence of its ruling group, which expressed itself partly in terms of ‘Norman’ identity. This article uses Dudo's history of the dukes and Richard's charters to argue that ‘Norman’ as a political identity was a deliberate creation of the court of Richard I in the 960s, following the perceived failure of his and his father's policies of assimilation into Frankish culture.

Normans and the Franks from before Rollo's landing up to the death of Rollo's grandson Richard, which closes the book. 6 Searle argued that Dudo was pandering to the self-conception of his audience by telling the story of how, after a century of ethnic hostility, the dukes of the Normans had cemented their rightful position over the Norman people by forcing the Franks to recognize their power. 7 This presents historians with a problem. The evidence for structural continuity, albeit (to paraphrase David Bates) of a rather dented sort, seems basically solid. 8 However, this raises the question of why Dudo wrote the way he did: why did Dudo present the Normans as being so different from what was around them?
The answer lies in the particular circumstances of Normandy in the latter half of the tenth century. Dudo did not pander to the self-conception of an already-existing group of 'Normans' standing outside the framework of ducal authority. Rather, the idea of being Norman as it appears in Dudo was the result of the crises that beset the Norman duchy in the 940s and 960s. Before William Longsword's death in 943, every indication is that the ruler in Rouen was assimilating into the Frankish world, in terms of both political behaviour and, as far as the limited sources indicate, the language of power. However, the crises of the mid-century changed the situation. The combination of the failures of the end of William's reign and the need to appeal to newly arrived warriors from Scandinavia -the Viking warbands who provided crucial military support to the ruler in Rouen during these troubled years -seems to have involved a change in ideological strategy on the part of the Rouen court. Part of this change was the use of the language of Norman ethnicity to create a powerful if rather incoherent sense of group solidarity to galvanize the duchy's disparate elites around the duke. It is this milieu that Dudo's history reflects, but Dudo is not the only evidence for this. A unique charter, issued for the monastery of Saint-Denis by Richard I in 968, also illustrates how Richard was attempting to propagate a sense of Norman identity among his followers.
Ethnic language was not the only aspect to this change, but it is certainly the most distinctive in a West Frankish context. 9 The role ethnicity played in this area in the tenth century was relatively limited. 10 Nonetheless, Normandy was not unique in the use of this kind of 6 P. Bauduin, 'Autour d'une construction identitaire: la naissance d'une historiographie normande à la charnière des X e -XI r siècles', in P. Nagy (de.), Conquête, acculturation, identité: des Normands aux Hungrois: les traces de la conquête (Rouen, 2001), pp. 79-89, at p. 81. 7 In addition to the reference in n. 5, see Searle, Predatory Kinship, pp. 79, 95 for ethnic hostility in Dudo's work. language: looking across the Channel to England, one can see the idea of Englishness being used and transformed in a parallel manner. 11 Even in the West Frankish kingdom, one can point to ideological transformations in the tenth century. One of the most striking, if not the most successful, was Charles the Simple's dramatic redevelopment of the ideological basis of Carolingian kingship. 12 However, the distinctiveness of the ethnic language in Normandy makes the subject worth exploring.
In the earliest decades of Normandy's history, during the reign of Rollo (r. 911-28x33) and his son William Longsword, the rulers of Rouen seem to have taken up the existing political culture relatively quickly. The Norman ruling family quickly began to intermarry with the Frankish elite. William Longsword married a daughter of Heribert II, count of Vermandois. 13 His sister, Gerloc-Adele, married William Towhead, duke of Aquitaine. 14 This was not unimportant. William Longsword used his connections with his brother-in-law William Towhead and his relative Bernard of Senlis as part of his efforts within the West Frankish political scene. 15 That is to say, he was behaving like a Frankish political insider.
Moreover, as far as we can tell from what little evidence survives, William used the language of Carolingian authority. There survives a Planctus written shortly after his death by a monk associated with the monastery of Jumièges. It is possible that this monk was specifically Anno, later abbot of Micy. The work addresses William's son Richard, as 'count of Rouen'. 16 This is the only work before the end of the century written for a patron from Rollo's family and for a Norman audience. The poem was almost certainly commissioned by Richard's paternal aunt Gerloc-Adele, wife of William Towhead count of Poitiers, and, as Elisabeth van Houts has recently argued, reflected the concerns of her specific 11 P. Wormald, 'Engla Lond: The Making of an Allegiance', Journal of Historical Sociology 7 (1994), pp. situation. 17 Moreover, the poem addresses the young Richard I directly. 18 Given the links between Poitiers and the Rouen court, it is probable that the poem was known in Rouen. It is also likely that the Planctus, commissioned by William Longsword's sister and addressed to his son, uses language tailored towards the concerns of the Rollonid family and the court at Rouen.
Thus, this poem intended to please high-ranking Normans describes the Norman ruler as 'count of Rouen', implying that, at the turn of the 940s, the Rouen rulers saw their position as that of a Carolingian count. This is particularly interesting because there does not seem to have been a count at Rouen at the end of the ninth century, when Rollo arrived in Neustria. 19 Count Odilard, the man usually cited as a possible count of Rouen, was more probably a count from around Laon or the area of modern Picardy. In addition to appearing in a charter of King Charles the Simple of 905, requesting a grant of land at Pîtres along with Bishop Ralph of Laon, Odilard also appears as a witness to a 906 charter of Robert of Neustria, count of Tours, for Saint-Amand, making his area of operation more likely around there than on the Upper Seine. 20 This is significant. Rather than simply taking over an already-existing institutional position, it seems that the early tenth-century rulers of Normandy tried to claim the language of Frankish power for themselves. Taking the title of 'count' was not necessarily the default choice: it was a specific claim made by the Rouen rulers about the nature of their power, a claim to be on equal terms with all the other great West Frankish magnates, the kind of equal terms seen in the marriages they contracted. It must be said that no writers other than the Planctus author seem to have paid attention to this. With the exception of the Planctus, all of the limited number of sources referring to Normandy between 911 and 942 are hostile or indifferent, and as such refer to the Norman ruler as just that -princeps Normannorum. 21 It is an open question whether or not this strategy would eventually have borne fruit outside Normandy, for William Longsword's assassina- case of the Évrecin -as we shall see, Rouen appears to have completely lost control of the area, and the loss seems to have rankled. Even in upper Normandy, Richard did not exercise power until the deposition of the Robertian lieutenant Ralph Torta, father of the bishop of Paris, in the 950s; and even then, he does not seem to have been a major force, and remained subordinate to Hugh the Great. 30 The second major crisis of the mid-tenth century was the so-called Norman War, a conflict fought between Richard and a group of other nobles, most prominently Theobald the Trickster, count of Blois, Chartres and Tours. This conflict was fought between about 960 and 966, but its roots seem to have been a long time in the making. The chief beneficiary of the curtailment of Richard's authority we noted previously was Theobald, Hugh's chief lieutenant. 31 The limited indications we have from the Chronicle of Nantes and some scattered references in Dudo's work indicate that Theobald was attempting to assert control over Brittany, a place where William Longsword had had some claim to hegemony. 32 More concretely, when Rouen lost the Évrecin, Theobald gained it. In charters from Chartres, we find the bishop of Évreux as a regular witness, indicating that the area was under Theobald's control. 33 It has been argued by a number of historians that Hugh the Great's death in 956 and the subsequent minority of his son Hugh Capet led to a general crisis in the Robertian 'state'. 34 The point is well made, and the Norman War must be seen as part of this. Where Theobald took advantage of the period between Hugh the Great's death and Hugh Capet's coming into his inheritance in 960 to place Châteaudun, Blois and Chinon under his own control, Richard took advantage of the removal of Hugh the Great's heavy hand to try and reclaim Évreux. 35 The general course of the war can be quickly summed up: Theobald attempted to capture Rouen, was defeated by Richard in battle, and subsequently had his lands ravaged by bands of pagan Vikings, newly arrived in Gaul. 36 Dudo is the main narrative source for what happened, and, as might be expected, he emphasizes Richard's own role, but the role of Richard's Viking allies seems to have been significant. They killed Theobald's son, also called Theobald; and burned down the city of Chartres itself. 37 What we can see, therefore, is that Richard's chief allies, in this crucial twenty-year period, were pagan Vikings. It is very probable that they were really pagans. 38 Dudo goes to some lengths to have Richard convert them or expel the recalcitrant, a narrative strategy that would have been unnecessary had they not been known to be pagans to begin with. 39 This also probably means they were new arrivals, as assimilation to at least some form of nominal Christianity seems to have happened fairly quickly within Normandy, as in the Eastern Danelaw. 40 It is therefore not surprising that the limited evidence seems to indicate a change in ideological strategy at about this time. Richard and his court deliberately attempted to galvanize a notion of Norman ethnicity as a way of appealing to a newly arrived elite who did not have an investment in the language of high Carolingian politics. The recasting of 'Norman' as an endonym is a new development as of the mid-tenth century. 41 This can be seen in the change in titulature that occurs at this time. The first evidence of Norman rulers using titles such as comes, marchio, or dux Normannorum dates from this period, the mid-to-late 960s. This represents a shift in conception of the authority of the Rollonid rulers: no longer counts of Rouen, but rulers of the Normans. The first, albeit slightly weak, piece of evidence for use of this sort of title is a diploma of King Lothar dating from 966 in favour of Mont-Saint-Michel, which     45 Although only preserved in a later copy, this charter has no diplomatic irregularities, and it fits, as we shall see, much better into the context of the late 960s than into any other time. 46 Moreover, Dudo knew of it: his history records a grant of the estate of Berneval to Saint-Denis by Rollo, as does the charter, and it is much more probable that Dudo got his information from a charter recording a genuine grant than that Dudo selected an estate in the Pays de Talou at random and some later forger based a charter on it. 47 This charter is the record of an actual event. This is important, because the event in question, the actual ceremony of the grant of Berneval to Saint-Denis, acted as a method of concretizing differences between Richard's followers -his Normans -and his neighbours. In the intitulation, as in the 966 diploma, Richard uses the title of marduo [sic for marchio] Normannorum. More than this, however, Franks and Normans are consistently juxtaposed throughout the charter. It opens with an address to 'both gentes, to wit, the Franks and the Normans', and refers to a meeting at Gisors as being a placitum of, again, the Franks and the Normans. 48 The immunity clause prohibits a long list of comital officials (all of whom, interestingly enough, have Carolingian titles such as vicarius) 42  from infringing the donation, but also prohibits 'any Frank'. 49 It is possible that the word francus here means 'free man', but this would be a very unusual usage for an immunity clause, and in context it seems likely both that Richard is specifically addressing 'the Franks'. Perhaps more interestingly, this also implies that none of the officials named as being under Richard's power was envisaged as being identified as a Frank -that is, these 'Franks' were distinct from Richard's officials, who were presumably 'Normans'.
Finally, the witness list juxtaposes Hugh Capet, dux Francorum with Richard, princeps Normannorum, and also notes that Theobald the Trickster (who is not otherwise known to have had dealings with Saint-Denis) was there as well. 50 The dichotomy must have been clear. This language of differentiation, reflecting an event which actualized that difference, comes in the context of this recently fought war: when Richard's men and Theobald's men looked at one another, it cannot have been far from their minds that these were the men who had burned down Chartres, who had tried to burn down Rouen, who had killed the count's son, who had stolen Évreux, and so on. In short, this charter is a relic of an instance in which difference was performed. The split between Frank and Norman, so present in the charter, would have been reflected in the physical space, dividing into groups of 'us' and 'them'. How 'Norman' was propagated as a political brand with which people could identify is mostly lost to history. Much of it must have been based on unrecorded personal interactions. 51 Nonetheless, this kind of performance of difference was one way of doing it, and this charter is the best surviving evidence.
In its language, the document reads like a compromise. On the one hand, Richard is made to acknowledge Hugh Capet as his senior, and given that Hugh was at the actual ceremony, this subordination was also presumably made physical. 52 The use of the word senior is interesting. There are a number of late 960s charters where Hugh Capet is described as a senior -it is, for instance, how he appears in Geoffrey Grisegonelle's charter of 966 reforming the monastery of Saint-Aubin -and it seems to be part of a definite strategy to obtain formal submission from his father's vassals. Moreover, the content of the charter itself makes a restoration of Berneval to the monastery of Saint-Denis, a Robertian house. 54 The practical impact of the charter, therefore, favours the Robertians, both materially and ideologically: Richard shows himself to be the man of Hugh Capet, verbally acknowledging his seniority and granting property to a Parisian monastery to demonstrate it. In particular, a grant to Saint-Denis probably represents a sign of favouring a specifically Robertian monastery. Mid-tenth-century Saint-Denis was still important to the Robertians as the Robertian necropolis. Hugh the Great was buried there, and Hugh Capet would be as well. 55 There were already links between Normandy and Paris. Hugh, archbishop of Rouen, had been a monk of Saint-Denis before being appointed to the archiepiscopal throne. 56 The monastery of Saint-Germain-des-Prés had been given large estates in southern Normandy. 57 Richard himself was married to Hugh Capet's sister Emma, who appears in this charter. 58 However, this affair is not a simple renewal of a long-standing alliance. The monks of Saint-Denis were made to come to Richard. This is more than simply standard supplicatory language. The space the monks cross to reach Richard is not abstract. They ask him once at Gisors, and he makes them come to him at Rouen, where they ask him again. 59 Then, and only then, does he grant their request: in his city, surrounded by his men, in a way presented very emphatically as his decision, influenced only by God and his own people. This indicates the relative strength of Richard's bargaining position: the monks are able to influence him to restore the estate, but Richard is able to do so in the full possession of his majesty. Indeed, Richard twice refers to his territory as a regnum. 60 Thus, Richard presents himself as the autonomous princeps Normannorum, ruling over his Norman subjects from Rouen. The ruler at Rouen was still claiming equality with the other West Frankish magnates, but was now doing so by stressing difference instead of similarity.
This leads into Dudo's milieu. Dudo primorum Normanniae ducum or the Historia Normannorum, was commissioned by Richard I in around 994, re-commissioned on Richard's death, and finished at around the turn of the millennium. The work probably had no immediate political point. There have been several attempts by historians to find such a point. 61 Few are convincing, and none has found widespread acceptance. 62 Like his contemporary Richer of Rheims, Dudo's chief goals seem to have been literary rather than narrowly political. 63 Dudo presented his patrons as worthy lords because they were wise and godly princes who behaved in a properly lord-like manner, and he does not seem to have had any more specific political aim. 64 As Christiansen points out, Dudo's history is a 'long-winded prosimetrical treatise'. 65 While some parts might be aimed at a more general audience, including laymen -an interesting and plausible, although of necessity unprovable, suggestion -it is hard to imagine how the work would accomplish even a general goal such as instilling a sense of group identity, let alone a more immediate political aim. 66 Richard I's purpose in commissioning it must be distinguished from the purpose of the re-commissioners, chiefly Richard's half-brother Ralph, count of Ivry, a major figure in the early years of Richard II's reign. This must in turn be distinguished from Dudo's purposes in writing it. The only outright evidence for Richard's purpose is Dudo's claim that Richard wanted him to write about 'the customs and acts of the Norman land, and moreover the laws of his forefather Rollo, which he established in the realm'. 67 This is, clearly, not very helpful. As such, Richard's motives must of necessity be surmised. Probably the most convincing suggestion is that Dudo's work was intended simply to be a monument to the Norman rulers, rather than to do anything specific. 68 61 By contrast, some, at least, of Dudo's goals are relatively straightforward. Both in terms of the form of the work and in terms of what Dudo tells us about its intended audience, we can safely follow Boje Mortensen in arguing that it was meant primarily as a school textbook for the Norman scholae, intended for teaching trainee authors about all the different kinds of prose and poetry they could possibly use. 69 Let us finally ask what the goals of Ralph of Ivry may have been. 70 In the first place, Dudo and Ralph seem to have had a close relationship. Apart from the De moribus, we possess two original charters handwritten by Dudo. 71 Both of these display much the same characteristics as the De moribus: an emphasis on terms of logic (the invocatio uses words such as rationicatio); a rather convoluted prose style; an emphasis on the importance of Richard I; and the presence of Ralph of Ivry, in whose name the 1011 charter is written, and who appears as an intercessor in the 1015 charter. 72 It therefore seems likely that Dudo's presence in the Norman court at the beginning of the second millennium was largely to do with his association with Ralph.
Ralph evidently liked Dudo's style. The 1011 charter, issued in his name, granting goods to Saint-Ouen of Rouen, did not have to be written by Dudo. Other charters to Saint-Ouen dating from the 1010s do not have the same style, including two originals from the period Dudo was active. 73 Dudo was not the regular scribe of Saint-Ouen, which means that he must have been deliberately chosen by Ralph to write his 1011 charter, which implies that Dudo's emphases -maybe on learning logic, more probably on the importance of Richard I -were such that Ralph actively sought him out. Given the points made above about how Dudo's work was probably chiefly -although not exclusively -intended to be something rather than to do something, it is likely that Ralph specifically commissioned Dudo to write a memorial to his half-brother Richard. The close connection between Dudo and Ralph, however, did not end with simply Ralph's choice of author. Dudo probably did not invent the content of his history. This does not mean that it is true, but it does mean that much of its content was a reflection of the ideas of others. Chiefly, in fact, one other: Ralph of Ivry, whom Dudo names as his relator, a word which can only mean that Ralph was his chief informant. 74 Given Ralph's close relationship with Richard I, therefore, it makes sense that the story Ralph told Dudo followed the agenda of the late tenth-century ducal court fairly closely. This is not to say that Dudo was simply Ralph's mouthpiece. However, it is likely that the plot and some of the themes of his work (including, for our purposes, what it means to be Norman; as we shall see, there are strong parallels between the charter and the De moribus) were taken from what Ralph of Ivry told him.
There is some circumstantial evidence to support this. Emily Albu once pointed out that everyone believed Dudo for 800 years. 75 However, the implications of this have rarely been fully thought through. Everyone told the same story as Dudo. William of Jumièges' account of early Norman history, written in the 1070s, is just Dudo's, abridged. 76 The Inventio et Miracula Sancti Vulfranni from Saint-Wandrille in the 1050s also retells Dudo's story, as does the Introductio Monachorum of Mont-Saint-Michel written at around the same time. 77 Moreover, parts of Dudo's work were known very early on. Arnoux has demonstrated that the Fécamp Chronicle found in Rouen manuscript 528, written between 996 and 1001, not only retells Dudo's story, but part of it copies Dudo's very words. 78 Arnoux then argues that Dudo probably wrote the rest of the Chronicle. 79 However, given that the rest of the Chronicle does not employ Dudo's prose style, and uses sources that Dudo did not use in his work and probably did not know, most notably the Vitae of Sts Audoënus, Leodgar and Wandregisl, it is likely that these portions were written by a different author, compiling a history out of others he had access to. This in turn means that this anonymous author thought that Dudo's history was an authoritative version of the Norman past. Thus, Dudo's history was found 74 Dudo, De moribus, p. 125; Christiansen, 'Introduction', p. xxv.

75
convincing by others around him within a very short space of time. 80 This implies that there was no other version of Norman history in common circulation at the end of the tenth century. Either Dudo was receiving and adapting a story that had already been propagated, or he was given major help in becoming so influential so quickly. The former is perhaps more likely, given the influence of Ralph of Ivry on the content of the work. Either way, Dudo was most likely reflecting a pre-existing story rather than creating one from scratch. What, then, can we say about the ideas that Dudo is reflecting?
The idea of being Norman in Dudo's work is of a fairly generic order, as in the 968 charter. For Dudo, the Normans are above all unified by their loyalty to the ducal house and their distinction from the Franks, although what makes them distinct from the Franks is never specified in great detail. While Dudo does use 'Danish-born' as a synonym for 'Norman', the people so designated are still defined by their loyalty to the duke. 81 This comes through most clearly in Book IV, where King Louis is made to say that 'the land of the Normans will never be under tutelage except to the government of one lord . . . the Danish-born people do not know how to be servants except to one master alone'. 82 Indeed, one might almost call Dudo's concept of what makes someone a Norman incoherent. His switches between explicitly removing a necessarily biological element from Norman identity and implicitly equating 'Norman' and 'Scandinavian'. On the one hand, we have the famous story of Rollo forging one people out of many, and Dudo's tracing of Norman, or 'Dacian', descent back to Antenor the Trojan, which gives Scandinavian and Frank a common origin. 83 On the other hand, however, we have an emphasis on the spoken language of Old Norse, specifically contrasted with 'the Roman tongue'. 84  of his way to emphasize a century-long continuity in both the ability of the Norman rulers to speak Old Norse and the fact that they actually did so. 85 It is for this reason that Dudo's famous story about Richard going to Bayeux to learn Old Norse was included: if there is any truth behind it, it probably reflects a hostage situation more than a language school, but in Dudo's work it explains and draws attention to Richard's abilities in this field. 86 Equally, there are several occasions where Dudo uses Dacigena, that is, Scandinavian-born, placed in specific opposition to Francigena, Frankish-born, as a synonym for 'Norman'. 87 The only real constant in all the evidence is that a Norman obeys the duke. Dudo in particular emphasizes this, making much of the bonds of fidelity linking the Normans and their ruler. 88 The climax of his narrative is when the new Vikings of the 960s make a deal with Richard: either convert and obey, or leave. 89 There was no ideological space for Normans who did not obey the Norman leader, and one is reminded of the seeming assumption in the 968 charter that all of the officials under Richard's power, despite their Carolingian titles of vicarius and centenarius, were Normans. 90 Friendly outside observers, such as Ralph Glaber, writing in the 1030s, also emphasize the role of the dukes in Norman identity, saying that the whole province lived 'as if one bloodline or family' under the dukes. 91 As we have already noted, Dudo was probably not making this up. It is likely that the incoherence came from his sources, and particularly from Ralph of Ivry. Ralph would be a good person to have a duke-focused idea of what being Norman actually entailed. As already noted, he was Richard I's half-brother, and therefore closely tied to a family that emphasized its Scandinavian blood. But he was not biologically Scandinavian at all: his mother, Sprota, was Breton, and his father, the 85 aforementioned Esperleng, was Frankish. 92 Hence, when he appears in the 968 Berneval charter as a Norman -for it is he who is one of Richard's fideles supporting the petition of the monks of Saint-Denisthis is an implicit acknowledgement that, despite a background in which he did not necessarily have anything in common with the new arrivals, he could share in the same identity owing to his obedience to Richard. 93 What, in the light of this, can we say about being Norman in the late tenth century? We can say three main things. First, that a Norman identity was being propagated by the ducal court several decades before Dudo wrote. Second, that the key part of the idea of the Normans is that they are a unified gens under one ruler, and while this was given most sophisticated expression in Dudo's history, it was not his invention. Third, that because Dudo's portrayal of the distinctiveness of being Norman is based on an ideological project emanating from the ducal court, it cannot be taken either as his own invention or, necessarily, as a product of the turn of the millennium rather than the reflection of older debates.
What was the appeal of promoting a unified ethnic identity for Richard's government? The Norman aristocracy in the latter half of the tenth century was disparate, and there was no necessary reason that a Norman identity need appeal to any of them. A number of important Norman families were of West Frankish origin. Of all the great families of eleventh-century Normandy, the easiest to trace to a West Frankish background are the Tosny family, who were probably introduced to the duchy via their kinship with Archbishop Hugh of Rouen, and whose first traceable member, Ralph, was important enough to be one of Richard's men to sign the 991 peace treaty with AEthelred the Unready. 94 However, there was also no pre-existing sense among Scandinavians in Gaul that their common geographical and/or linguistic heritage needed to be something that unified them. 95 We have already seen how, in practice, Christian 'Normans' teamed up with Christian 'Franks' against pagan 'Normans'.
Nonetheless, in the 960s, the Rouen court needed to galvanize a disparate group of people into a whole, and to do it quickly, lest it be unable to secure and maintain their support. In the face of the Norman War, there was an urgent need to appeal to incoming Viking warriors who could provide much-needed military support. Moreover, the kind of language of authority used by William Longsword's court would probably have had little effect on the newcomers. As new arrivals into Gaul, there is no reason why they would have had any particular investment in a Carolingian vocabulary of legitimacy, particularly given how intimately tied in to Christianity it was. In less troubled times, there might perhaps have been time to let them adapt. This is what seems to have happened in the 920s and 930s, when the Norman heartland in Rouen was not under direct threat. However, in the 960s, the pressure of war meant that results were needed sooner rather than later.
It may have helped that the Normans were already united as a group -by the Franks. Throughout the tenth century, both in narrative sources and in charter evidence, persons of Scandinavian descent are referred to by Frankish sources as undifferentiated 'Normans'. Flodoard, for instance, makes no difference between Richard -a third-generation immigrant born of Christian parents, with blood ties to the highest levels of the nobility -and Harald, a pagan war-leader newly arrived at Bayeux. 96 If Richard wanted to look for a common theme uniting himself and a group of Scandinavian arrivals, therefore, it made sense to accept the word 'Norman' and recast it as an endonym. However, the incoherence that attached to the concept leant it flexibility, allowing it to also appeal to followers of the duke whose background was not Scandinavian. 97 They too could be Normans, so long as they obeyed the duke, for obedience to the duke was the only real constant of what being Norman involved.
As noted above, appeal to ethnic identity was simply the most distinctive strategy of legitimation available to the Norman ruler. It was not the only one, nor, in many contexts, even the most important. The other surviving charter written in Richard's voice, issued for Fécamp in 990, uses no ethnic terminology at all. 98 Instead, it is a charter replete with the language of high ecclesiastical ceremony and is explicitly liturgical, particularly in the sanctio clause threatening violators of the grant with 96  damnation. 99 This document announces to anyone encountering it the importance of the church and of Christianity to the life of the Norman realm. Richard was making a public statement -for it seems to have been issued at a large, formal gathering -that he was a godly prince, ruling a godly people and a pious church. Of course, this charter was issued for the dedication of Richard's college of canons at Fécamp, with the full complement of bishops from the ecclesiastical province of Rouen attending, so it is to be expected that its wording might be more overtly ecclesiastical than that of the 968 charter, issued in a more explicitly secular gathering. Moreover, in the ducal heartland, there was no need to define 'Norman' against an outside Other -everyone present could reasonably expected to be Norman. 100 Nonetheless, this charter demonstrates that in the ducal heartland around Fécamp, towards the end of the millennium a language of legitimation couched entirely in terms of Carolingian Christianity was thought to be able to succeed, at least in some contexts, without needing to be bolstered by ethnic identification.
There is, unfortunately, limited evidence for the way in which the ideological strategies of the Norman ruler were received on the ground, and all of it is later than the turn of the millennium. In charters issued by members of the great Norman aristocratic families between around the year 1000 and the year 1040, few make much reference to political ideas in their arengae, focusing instead on the spiritual benefits of donating to the church. The chief exception to this is a charter issued around the turn of the 1030s by Roger de Montgomery, which talks of the 'monarchia of the Normans' in much the same way as found in ducal rhetoric. 101 The Montgomery family, of whom one, Roger II, was famously ex Northmannis Northmannus, seem to have had a particular connection to a Norman identity, although such explicitness on the matter was atypical among their peers. 102 However, these acts, as van Houts argues, have striking similarities of vocabulary with the work of the Norman historian William of Jumièges, who was likely the author of these acts. Given that William, as a scribe, is linked not to Roger but to the monastery of Jumièges (similar language is used in a 1038 act in the name of Richard, count of Évreux, to the same monastery), we cannot necessarily postulate the kind of seemingly close relationship between Roger and William as