To paraphrase Jacques Le Goff, did the Rollo of our documents exist? And since that is all we have, did Rollo exist? An underlying thrust of this paper is that the Rollo of the available sources is best understood primarily not as an historical figure, but rather as a literary figure created to suit the ideological needs of, and conform to the political realities of, later generations. It is important to examine the historicity of Rollo, exploring the little information about him that can be squeezed from contemporary sources.
Because of the exaggerated biography of Rollo written a century later by Dudo of Saint-Quentin, it sometimes seems that we know him fairly well. In fact, although in Dudo’s time Rollo was remembered as the great founder of the Norman dynasty, during his own lifetime he was largely invisible. Although Dudo says much about his career prior to 911, virtually every story he tells is obviously borrowed from the adventures of other Northmannic leaders told in tenth-century Frankish chronicles, and the rest are clearly legends. Dudo’s account does not contain a single verifiable fact about Rollo. The ‘Founder of Normandy’ also is never mentioned in any contemporary source before 911, and in fact the Treaty of Saint-Clair-sur-Epte, made so famous by Dudo’s story of the Northman upending Charles the Simple instead of stooping to kiss his foot, made no impact whatsoever on the writers of the time. Only three brief mentions of Rollo occur in contemporary sources, and it is upon these that we must build what little image of him we can manage.
1. The first contemporary mention of Rollo is in a charter of King Charles Simplex in 918. Here, Charles grants the lands of an abbey “… praeter partem ipsius abbati quam annuimus Normannis Sequanensibus videlicet Rolloni suisque comitibus pro tutela regni”, (except for the part that we have given to the Northmen of the Seine, namely to Rollo and his companions)
2. At roughly the same time, the Frankish historian Flodoard of Rheims wrote without mentioning Rollo: “Post bellum quod Robertus comes contra [Nortmannos] Carnotenus gessit fidem Christi suscipere receperunt concessis sibi maritimis quibusdam pagis cum Rothomagensis quam pene deleverant urbe et aliis eidem subjectis”. (After the war that Count Robert waged against the Northmen at Chartres, certain maritime pagi, along with the city of Rouen, which they had nearly destroyed and other pagi which were subjected to it, were conceded to them, and they agreed to take up the faith of Christ). 
3. At some point before 928, the archbishop of Rouen wrote to Herveus, archbishop of Rheims, asking for advice on how to handle lapsed pagan converts. Herveus in turn wrote to Pope John X, asking “What should be done when they have been baptised and rebaptised, and after their baptisms continue to live in pagan fashion, and in the manner of pagans kill Christians, massacre priests, and, offering sacrifices to idols, eat what has been offered?” It is also perhaps significant that in a poem mourning the death of Rollo’s son, William Longsword, William is called the Christian son of a pagan father, although this might have been a rhetorical reference to Rollo’s earlier life.
It seems that after the Battle of Chartres, which most historians date to 911 and at which a large Northman force was soundly defeated, King Charles granted some land around Rouen and to the sea to Rollo and his companions, who converted to Christianity but, at least in some cases, quickly reverted to paganism. A number of important points can be made. First, there is no indication that Rollo was involved with the Battle of Chartres. Dudo later says that he in fact led the Northmannic army there. However, given that Dudo says that Rollo led every major Northman force in France, and some in England, it seems to be his way of showing the importance of a man who left virtually no trace in the historical record. Charles’ motivation seems to have been to cut off future attacks on the Seine and its tributaries by giving those Northmen who already controlled Rouen, the first major city on the Seine, royal recognition in exchange for their blocking access to other Northmannic forces. If Rollo in fact were not at Chartres, it would eliminate the contortions that historians have traditionally gone through to explain why Charles rewarded a man who had just suffered a great defeat.
Second, it is not clear exactly which lands Rollo and his companions received, but it would seem to have been roughly the Roumois and the Pays de Caux. Some historians, believing that the concession of Charles Simplex and two further royal “grants” in 924 and 933 comprised a formal concession of the future Normandy, have drawn neat maps dividing Normandy into three parts, and thus made the first concession cover all of Normandy east of the Risle, plus the entire Pays d’Ouche west of Évreux, but there is no evidence that Charles had such a great extent in mind. Third, the territory controlled by Rollo by no means contained the only Northmannic settlements in the future Normandy. Place-names show that the Northmannic presence was strong not simply in the Pays de Caux and Roumois, but also in the Bessin and the entire Cotentin peninsula. These Northmen of western Northmanland, at least to begin with, had no connection with the Northmen of Rouen. Finally, such arrangements as Saint-Clair-sur-Ept” were probably not intended to be permanent. Previous Frankish grants to Northmen had always proven temporary, either because the Franks managed to recover their losses or because the Northmen themselves could not hold it together. By 911, extensive Frankish experience told Charles that the loss of Rouen would only be a temporary setback; Saint-Clair-sur-Epte is unique not because it happened, but because against all odds and precedent the Rollonids managed to make their new principality stick.
So far we have a group of Northmen, led by Rollo, occupying Rouen and its surrounding area with the permission of King Charles, and at least nominally Christianised but subject to spectacular reversions. In the following years, while the sources are silent on events within the newborn Rollonid Principality, the Carolingian political landscape was changing dramatically. During the 910s, Charles managed to alienate many of his nobles through various actions, especially in Lotharingia. These events inspired Henry I, king of the East Franks, to renew East Frankish claims to Lotharingia, and Robert of Neustria, brother of the former King Odo, to lead a revolt against Charles. For several years the struggle continued, culminating in 922 when Charles fled his kingdom and Robert was crowned king. In the following year, Charles returned with an army; in the ensuing battle Robert was killed, but his son Hugh the Great and Herbert II of Vermandois defeated Charles. The events that followed are confused, but apparently neither Hugh nor Herbert would allow the other to become king, so they settled on Ralph, the son-in-law of King Robert and the duke of Burgundy. Possibly they believed that as a relative outsider to the West Frankish world, he would be easier to control. Herbert then arranged a meeting with Charles, arrested him, and threw him into captivity almost until the end of his life in 929. His possession of the Carolingian claimant to the throne only enhanced Herbert’s power, as did his arrangement in 925 for his five-year-old son Hugh to be made archbishop of Rheims, at the time the spiritual capital of the West Frankish realm. One may assume that the new archbishop was somewhat subject to Herbert’s influence.
To pick up the scanty narrative of events in the Rollonid Principality from contemporary sources, in 924 Flodoard reports: “The Northmen entered peace with the Franks through the oaths of Counts Hugh [the Great] and Herbert [of Vermandois] and also Archbishop Seulf [of Rheims], in the absence of King Ralph; but with Ralph’s consent the lands of Maine and the Bessin were conceded to them in the peace-treaty.” The neat, three-part maps of Normandy make this concession cover all the lands between the Vire and the Risle, and usually claim that Maine was a mistake, since it does not lie within Normandy and was never claimed by the Rollonids until well into the eleventh century. But seen in the light of previous “grants” by Frankish kings to Northmen, this should be seen not as a transfer of clearly-defined territory from one party to the other, but rather as permission by the king for Rollo and his companions to take whatever control they can over lands that have slipped completely out of the king’s power; in other words, trying to replace “bad” Northmen (i.e, ones with whom the king has no relationship) with “good” ones (with whom he does). We can tell from the very existence of this treaty that the Northmen of Rouen had fallen out with the king since the initial concession, and now were being reconciled. But the reconciliation did not last. Flodoard informs us in 925 that an army of Northmen of Rouen moved east, plundering Beauvais, Amiens and Noyons. At the same time, the Frankish natives of the Bessin rose against the Northmen there, and a Frankish army led by Hugh the Great’s men ravaged the Roumois. The Rouennais army quickly returned home, just in time to face a new invasion of Herbert of Vermandois and Arnulf of Flanders, along with the count of Ponthieu. They besieged the Northman stronghold of Eu, and despite a large relief force from Rouen led by Rollo (this is the second time he is mentioned by a contemporary source), they succeeded in capturing and destroying it. But the hostilities seemed to end there, for the moment.
The final years of Rollo are very shadowy, although he seems to have played some role in Frankish politics. In 927, a war broke out between King Ralph and Herbert of Vermandois. When Ralph returned to Burgundy to see to his duties there, Herbert apparently began to float the idea of a restoration of Charles Simplex. He brought Charles to meet with the Northmen (presumably led by Rollo) at Eu, where “the son of Rollo [William Longsword] committed himself to Charles and confirmed friendship with Herbert.” Apparently at this time, Herbert’s son Odo was left with Rollo as a hostage. This was a normal component of peace treaties during this period. The hostages were treated honourably, and in addition to serving as incentive for the parties to behave as in extreme cases, hostages could be executed for bad behaviour on the other side’s, they also served to create a closer relationship between the sides. Flodoard does not explain why this meeting took place, or what the participants expected to accomplish. In the following year, however, Herbert and Ralph were reconciled; but Rollo did not return Odo to his father until Herbert committed himself to Charles Simplex. Flodoard’s account of this is the third and final time Rollo is mentioned in contemporary accounts. It would seem that Rollo, once he had allied himself with Charles, refused to accept Herbert’s change of heart, and forced Herbert to renew his own alliance with Charles. It should also be noted that Rollo owed his original entré into Frankish politics to Charles, and that he had never met Ralph. The “grant” of 924 was made on Ralph’s behalf, but in his absence. In the event, however, nothing came of this uneasy alliance among Rollo, Herbert, and Charles, as Charles died in 929.
We do not know when Rollo died, but it must have been sometime between 927, when Flodoard last mentions him, and 933, when William Longsword makes his first recorded appearance as the Rollonid ruler. We may suspect, however, that Rollo played a greater part in the Frankish world than this bare narrative of his career has shown. For instance, a later source calls him a friend of William of Aquitaine, and the fact that William married Rollo’s daughter lends credence to this story. But overall, Rollo died in much the same obscurity in which he lived. Although in retrospect his achievement as founder of Normandy seems considerable, in his own day he was simply a Northman leader who got some territorial concessions from the Frankish king. The Rollonid Principality on the death of its founder was a small area centered upon Rouen, surrounded by neighbours hungry to reclaim what had been lost to the foreigners, and allied with the king who was losing the Frankish civil war. Its future was still very much in doubt, and in fact it barely survived its founder’s death.
That is, more or less, what we know about Rollo and his career. In conclusion, it is important to note that there are some things that we do not know about Rollo. We do not know with any certainty what his name was. We call him Rollo, because that’s what the sources generally call him, although some more distant writers referred to him as Ruinus, Roso, and possibly Rotlo. It is generally assumed that his “real” name was Hrólfr; this is his name in the later Norse stories, and if those stories were based on the historical Rollo, then some weight can be put on this theory. If, however, the historical Rollo was simply grafted on to pre-existing Norse stories, then Hrólfr may simply have been considered a reasonably good fit. Although most historians seem to have accepted the identification of Rollo as Hrólfr, some have dissented, suggesting that Hrólfr is not a logical origin for the Latinisation Rollo. Without considering at all the implications, it could be suggested that Göngu-Hrólfr’s brother in the Heimskringla, a historical figure who settled in Iceland, is named Hrollaugr, a name which much more easily lends itself to latinisation as Rollo.
We do not know his age, where he came from, or when he arrived at Rouen. Dudo places his arrival in 872, but that seems to be in order for him to be in place to lead the siege of Paris, which of course he did not. It is often suggested that he arrived shortly before his agreement with King Charles in 911, although it is likely that he would have to have been there longer in order to have the level of control that would have made him worth dealing with. The earliest report of his origin, Dudo, makes him a Dane but 12th century Norse texts make him Norwegian. The Danes themselves never seem to have claimed him, and Dudo knows nothing about Denmark that he did not read in Strabo. Furthermore, his daughter had an unambiguously Norwegian name, Gerloc, although this could reflect the origin of her mother, not Rollo. The evidence points generally, but not conclusively, to Rollo being Norwegian. There were Norse settlements on the lower Seine as early as the 840s, so it is not impossible, whatever his ethnicity that he was in fact born in Normandy, but the reference to William Longsword in the Planctus as born overseas implies otherwise, unless Rollo travelled in his youth and returned to make his fortune at Rouen. Most likely, he was born either in Norway or in a Norwegian colony elsewhere, perhaps in the British Isles, where some later traditions have him in his pre-Norman career.
As for his age, all we can say is that he was probably old enough in 911 to be a force to be reckoned with, perhaps thirty but young enough to remain active until the late 920s, say in his early seventies, the age of Henry I of England when he died. That would place his birth between around 855 and 880. The age of his son, William Longsword, does not help, because we do not know when he was born either. He was old enough to rule by as early as 927, maybe twenty-five and given his lack of concern for the succession at the time of his murder in 942, he must have expected he would still have children, so probably less than fifty. William does not seem to have considered Richard I, whom William probably never even met, to have been his heir. It was only upon William’s sudden and unexpected death that Richard was put forward as William’s successor, and since William was married at the time to Leutgarde of Vermandois, he probably expected that a legitimate heir would be forthcoming. This would place William’s birth between roughly 880 and 905, which does not narrow the range for Rollo.
This largely expends our knowledge of the historical Rollo. In his own day, he was a shadowy figure who does not seem to have made much of an impression on his contemporaries, or had much of an impact on his time. But his descendents over the course of the tenth century transformed his pirate chiefdom into the duchy of Normandy, and when Dudo in the 11th century, and the Norse saga-tellers of the 12th and 13th looked back at Rollo. They could not accept the birth of such a great nation from such an insignificant figure. So they transformed the historical Rollo into what they felt was a more fitting founder for Normandy, a literary Rollo and this is the Rollo we know.
 The middle section of Jacques Le Goff, Saint Louis, (Paris: Gallimard, 1996) is entitled ‘Le production de la mémoire royale: Saint Louis a-t-il existé?’ see page 314: “Le Saint Louis de nos documents a-t-il existé? Et comme c’est le seul qui s’offre à nous, Saint Louis a-t-il existé?” Steven L. McKenzie, King David: A Biography, (Oxford: Oxford University Press, 2000) also faces the challenge of recovering a ‘biography’ from a literary source fraught with legend.
 Dudo of Saint-Quentin, De moribus et actis primorum Normanniæ ducum, edited by Jules Lair, Mémoires de la Société des Antiquaires de Normandie 23 (Caen: F. Le Blanc-Hardel, 1865), now translated as Dudo of Saint-Quentin, History of the Normans, translated by Eric Christiansen (Woodbridge: Boydell, 1998). A new edition and translation into French by Pierre Bouet is promised. Dudo’s biography of Rollo can be found in Dudo, Book 2. Dudo’s historicity was savaged in Henry Howorth, “A Criticism of the Life of Rollo as Told by Dudo of St Quentin,” Archaeologia 45 (1880): pages 235-50, and Henri Prentout, Étude critique sur Dudon de Saint-Quentin et son histoire des premiers ducs normands (Paris: Picard, 1916). Despite defences such as Lair’s introduction to his edition of Dudo and Johannes Steenstrup, Normandiets Historie under de syv første Hertuger, 911-1066, Mémoires de l’Académie royale des sciences et des lettres de Danemark, 7me série, Section des Lettres 5.1 (Copenhagen: Andr. Fred. Høst & Søn, 1925), Dudo’s critics have largely held the field as even his harshest critics seem to hold to a largely Dudoesque early Normandy. In recent years, however, Dudo has enjoyed a significant resurgence. At Caen, a “neo-Dudonist” school is emerging, seeking to rehabilitate Dudo as historian, led by Pierre Bouet and François Neveux; see François Neveux, La Normandie des ducs aux rois (Xe-XIIe siècle) (Rennes: Ouest-France, 1998). Further, some historians have come to appreciate Dudo as a source not for the history of the 10th century, but for the intellectual climate of Normandy and the Carolingian world in the 11th century. See, Eleanor Searle, “Fact and Pattern in Heroic History: Dudo of Saint-Quentin,” Viator 15 (1984): pages 119-37; Leah Shopkow, “The Carolingian World of Dudo of Saint-Quentin,” Journal of Medieval History 15 (1989): pages 19-37; Pierre Bouet, “Dudon de Saint-Quentin et Virgile: L’Enéide au service de la cause normande,” in Recueil d’études en hommage à Lucien Musset, Cahier des Annales de Normandie 23 (Caen: Musée de Normandie, 1990), pages 215-36; Victoria B. Jordan, “The Role of Kingship in Tenth-Century Normandy: Hagiography of Dudo of Saint-Quentin,” Haskins Society Journal 3 (1991): pages 53-62; Emily Albu (Hanawalt), “Dudo of Saint-Quentin: The Heroic Past Imagined,” Haskins Society Journal 6 (1994): pages 111-18; Felice Lifshitz, “Dudo’s Historical Narrative and the Norman Succession of 996,” Journal of Medieval History 20 (1994): pages 101-20; and the articles in Dudone di San Quintino: Sono qui raccolte le relazioni tenute dagli intervenuti al Convegno su Dudone di San Quintino, organizzato a Trento dal Dipartimento di scienze filologiche e storiche dell’Universita atesina il 5 e 6 maggio 1994, edited by Paolo Gatti and Antonella Degl’Innocenti, Labirinti 16 (Trent: Universita degli studi di Trento, 1995).
 Northmannus and its variants were initially the Latin words for Viking. Over time, Northmannus came to mean Norman. During the tenth century, the word still clearly had its Viking connotation. Since, during this period, the words Viking and Norman create a false dichotomy between groups who were, to the medieval mind, identical, I will use Northman and Northmannic to refer to both Vikings and Normans.
 Recueil des actes de Charles III le Simple, edited by Ferdinand Lot and Philippe Lauer (Paris: Imprimerie Nationale, 1949), no. 92. The settlement of Rollo and the Northmen of Rouen is discussed in Auguste Eckel, Charles le Simple: Annales de l’histoire de France à l’époque carolingienne, Bibliothèque de l’École des hautes études 124 (Paris: Émile Bouillon, 1899), pages 60-90; Prentout, Étude critique, pages 196-250; Steenstrup, Normandiets Historie, pages 50-55; David C. Douglas, “Rollo of Normandy,” English Historical Review 57 (1942): pages 426-30; David Bates, Normandy before 1066 (London: Longman, 1982), pages 8-9; Eleanor Searle, Predatory Kinship and the Creation of Norman Power, 840-1066 (Berkeley: University of California Press, 1988), pages 40-47; Neveux, La Normandie, pages 29-31. Cf. Dudo 2.24-29. The only full-scale study of Charles the Simple remains Eckel, Charles le Simple; although dated and marred by excessive reliance on Dudo, it remains useful as a general narrative of events in the Frankish world during Charles’ reign. For a brief but current overview of the 10th century Frankish world, see Jean Dunbabin, “West Francia: The Kingdom,” in The New Cambridge Medieval History, Volume 3, ed. Timothy Reuter (Cambridge: Cambridge University Press, 1999), pages 372-97. For the interesting argument that Rollo was first officially settled in Neustria by Robert the Strong c. 889, and that the agreement of 911 was simply a confirmation of the earlier settlement, see Felice Lifshitz, “La Normandie carolingienne: Essai sur la continuité, avec utilisation de sources négligées,” Annales de Normandie 48 (1998): pages 505-24; in this case, Dudo’s King Charles is Charles the Fat.
 Flodoard of Rheims, Historia Remensis ecclesiae, edited by Martina Stratmann, MGH, Scriptores 36 (Hannover: Impensis Bibliopolii Hahniani, 1998), 4.14. On Flodoard, see the exhaustive study of Michel Sot, Un historien et son Église au Xe siècle: Flodoard de Rheims (Paris: Fayard, 1993). Douglas, “Rollo of Normandy”, interprets this as meaning Rollo was at Chartres; cf. Dudo 2.22-23. Richer has a garbled account of Rollo’s dealings with the Franks, in which the adventures of Robert of Neustria on the Loire are conflated with Rollo; Richer of Rheims, Historiae, edited by Hartmut Hoffmann, MGH, Scriptores 38 (Hannover: Impensis Bibliopolii Hahniani, 2000). For a thorough discussion of Richer, his work, and its context, see Jason Glenn, “Political History: The Work of Richer of Saint-Remigius” (Ph. D. diss, University of California, Berkeley, 1997). Of the four chapters of this dissertation, 1 was published as Jason Glenn, “The Composition of Richer’s Autograph Manuscript,” Revue d’histoire des textes 27 (1997): pages 151-89, 2 as Jason Glenn, “The Lost Works of Richer: The Gesta Adalberonis and Vita Gerberti,” Filologia Mediolatina 4 (1997): pages 153-90, and chapter 4 is forthcoming as Jason Glenn, “Two Views of a Frankish Civil War,” Journal of Medieval History. A revised version of the dissertation is promised as Political History: The Work and World of Richer of Reims. On the Battle of Chartres, see Prentout, Étude critique, pages 191-6, who surprisingly does not overtly question Rollo’s participation at Chartres, although he gives no reason to accept it; Steenstrup, Normandiets Historie, pages 40-45; Searle, Predatory Kinship, pages 42-43, who accepts the involvement at Chartres of the Northmen of Rouen and, seemingly, Rollo.
 Bibliothèque nationale, ms. lat. 4280, fol. 102r (letter of Herveus of Rheims) and fol. 106v (letter of Pope John X); the conversion of Rollo and the Northmen of Rouen is discussed in Olivier Guillot, “La conversion des Normands peu après 911: Des reflets contemporains à l’historiographie ultérieure (Xe-XIe siècles),” Cahiers de Civilisation Médiévale 24 (1981): pages 101-16, 181-219, with relevant excerpts printed at page 102, note 8; Prentout, Étude critique, pages 250-60; Steenstrup, Normandiets Historie, pages 79-82; Douglas, “Rollo of Normandy,” pages 432-4; Bates, Normandy before 1066, pages 11-12; cf. Dudo 2.30-31. It should be noted that Dudo records absolutely no historical information about Rollo’s reign; after the “treaty” and his “conversion,” there remains in Dudo only a fable about Rollo’s justice, and his death.
 Planctus for William Longsword, Verse 2. For bibliographical references and a discussion of the Planctus as a source for Rollo, see Appendix 2 below.
 Several late chronicles recount the siege of Chartres; none of them associate it with Rollo or the “foundation of Normandy.” Annales Sanctae Columbae Senonensis, edited by Georg Heinrich Pertz, MGH Scriptores 1 (Hannover: Impensis Bibliopolii Hahniani, 1826), s.a. 911; John of Bèze, “Chronicle of Saint-Pierre de Bèze,” in Chronique de l’Abbaye de Saint-Bénigne de Dijon, suivie de la Chronique de Saint-Pierre de Bèze, ed. E. Bougaud and Joseph Garnier, Analecta Divionensia 9 (Dijon: Darantière, 1875), 280; “Chronicle of Saint-Bénigne de Dijon,” ibidem, 115.
 For good reviews of Norman toponymic studies, see Gillian Fellows-Jensen, “Scandinavian Place-Names and Viking Settlement in Normandy: A Review,” Namn och Bygd 76 (1988): pages 113-37, updated and translated into French as Gillian Fellows-Jensen, “Les noms de lieux d’origine scandinave et la colonisation viking en Normandie: Examen critique de la question,” Proxima Thulé 1 (1994): pages 63-103; Jean Renaud, Les Vikings et la Normandie (Rennes: Ouest-France, 1989), pages 153-98. The definitive studies remain the numerous articles of Adigard des Gauries, mainly in the Annales de Normandie 1951-1959, collected in Jean Adigard des Gautries, Onomastica minora Normanniae: Recueil d’études sur les noms de lieux et les noms de personnes d’origine scandinave en Normandie, Studia nordica (Paris: Société des études nordiques, 2002).
 The point is made succintly by Searle, Predatory Kinship page 44: “[Frankish rulers dealing with Northmen] meant as little as possible, we may plausibly guess.” Bates, Normandy before 1066, page 8, agrees: “The grant of lands to Rollo and his followers must therefore be seen as a typical response of the harassed western European ruling classes to the Viking menace. The ‘Treaty of St-Clair-sur-Epte’ was made between a Frankish king whose successors might easily seek to overthrow it and a Northman chief who could not guarantee to control the new settlers.”
 A dozen years before Raoul’s “grant” to Guillaume, Robert of Neustria conceded “Brittany with the pagus of Nantes” to the Northmen of the Loire, on condition that they accept Christianity. Flodoard of Rheims, Les Annales de Flodoard, edited by Philippe Lauer, Collection des textes pour servir à l’étude et à l’enseignement de l’histoire 39 (Paris: Picard, 1905), s.a. 921: “Rotbertus . . . Britanniam ipsis [i.e, Normannis qui Ligerim fluvium occupaverunt] quam vastaverant cum Namnetico pago concessit quique fidem isti coeperunt suscipere.” But by this time, the practice of conceding territory to the Northmen, often on condition of conversion, was a century-old tradition in the other major theatre of Northman operations, the Low Countries. Louis the Pious gave Rüstringen, at the mouth of the Wissen, to Haraldr, recently-deposed king of the Danes, after the conversion of Haraldr and his family before Louis. Annales regni Francorum inde ab a. 741 usque ad a. 829, qui dicuntur Annales Laurissenses maiores et Einhardi, edited by Friedrich Kurze, MGH Scriptores in usum scholarumcis 6 (Hannover: Impensis Bibliopolii Hahniani, 1895), s.a. 826: “Herioldus cum uxore et magna Danorum multitudine veniens Mogontiaci apud sanctum Albanum cum his quos secum adduxit baptizatus est; multisque muneribus ab imperatore donatus per Frisiam qua venerat via revesus est. In qua provincia unus comitatus qui Hriustri vocatur eidem datus est ut in eum se cum rebus suis si necessitas exigeret recipere potuisset.” Lothair I conceded the island of Walcheren, off the mouth of the Meuse, to Haraldr. Annales de Saint-Bertin, edited by Félix Grat, Jeanne Vielliard and Suzanne Clémencet, Société de l’histoire de France. Publications, série anterieure à 1789 470 (Paris: Klincksieck, 1964), s.a. 841: “Herioldo…Gualacras aliaque uicina loca huius meriti gratia in benificium contulit [Hlotharius].” At about the same time, Ragnarr received from Charles the Bald Turnhout, near the mouth of the Scheldt, from which Charles subsequently expelled him. Rimbert, Vita Anskarii auctore Rimberto accedit vita Rimberti, edited by Georg Waitz, MGH Scriptores in usum scholarumcis 55 (Hannover: Impensis Bibliopolii Hahniani, 1884), c. 21: “Nam cum cella supradicta Turholt in partem cessisset venerandi regis Karoli ipse eam a servitio quod pater suus disposuerat amovit et vobis bene cognito dedit Raginario.” Lothair gave the area around Dorestadt, on the Rhine near its mouth, to Haraldr’s brother Hrórekr, and in a story with suspicious parallels to the “Treaty of Saint-Clair-sur-Epte,” Charles the Fat later granted the same territory to Goðfreðr, “king of the Northmen,” along with Lothair’s daughter, on condition of conversion to Christianity. Annales de Saint-Bertin, s.a. 850: “[Roricum] Hlotharius cum comprimere nequiret in fidem recipit eique Dorestadum et alios comitatus largitur”; Annales Fuldenses, edited by Friedrich Kurze, MGH Scriptores in usum scholarumcis 7 (Hannover: Impensis Bibliopolii Hahniani, 1891), s.a. 882: “Imperator…Gotafridum de fonte baptismatis levavit et quem maximum inimicum et desertorem regni sui habuerat, consortem regni constituit. Nam comitatus et beneficia quae Rorich Nordmannus Francorum regibus fidelis in Kinnin tenuerat eidem hosti suisque hominibus ad inhabitandum delegavit”; Regino of Prüm, Chronicon cum continuatione Treverenski, edited by Friedrich Kurze, MGH Scriptores in usum scholarumcis 50 (Hannover: Impensis Bibliopolii Hahniani, 1890), s.a. 882: “Novissime rex Godofridus Nordmannorum ea conditione christianum se fieri pollicetur si ei munere regis Fresia provincia concederetur et Gisla filia Hlotharii in uxorem daretur. Quae ut optaverat adeptus baptizatus est et ex sacro fonte ab imperatore susceptus”; “Annales Vedastini,” in Annales Vedastini et Xantenses, ed. Bernhard von Simson, MGH Scriptores in usum scholarumcis 12 (Hannover: Impensis Bibliopolii Hahniani, 1909), s.a. 882: “Godefridus vero rex ad eum [Karolum] exiit: cui imperator regnum Frisonum quod olim Roricus Danus tenuerat dedit coniugemque ei dedit Gislam filiam Hlotharii regis Nortmannosque e suo regno abire fecit.”
 For Charles’ reign before his deposition, see Eckel, Charles le Simple, pages 60-115
 For the conflict between Charles and the Robertines, see Eckel, Charles le Simple, pages 116-35, and Philippe Lauer, Robert Ier et Raoul de Bourgogne, rois de France, 923-936: Annales de l’histoire de France à l’époque carolingienne, Bibliothèque de l’École des hautes études 188 (Paris: Émile Bouillon, 1910).
 Flodoard of Rheims, Annales, s.a. 924: “Nordmanni cum Francis pacem ineunt sacramentis per Hugonem et Heribertum comites Seulfum quoque archiepiscopum absente rege Rodulfo: ejus tamen consensu terra illis aucta Cinomannis et Baiocae pacto pacis eis concessae.” This is discussed in Douglas, “Rollo of Normandy,” pages 429-30; Neveux, La Normandie, page 31. Dudo’s suggestion that Rollo received all of Normandy (plus his choice of Flanders or Britanny) from Charles in 911 has long been replaced by what I call the three-stage model, in which Normandy, through the “grants” of 911, 924, and 933, more or less achieved the duchy’s final form by 933. The three-stage model is accepted by Prentout, Étude critique, pages 180-87; Jean-François Lemarignier, Recherches sur l’hommage en marche et les frontières féodales, Travaux et mémoires de l’Université de Lille, Nouvelle Série, Droit et Lettres 24 (Lille: Bibliothèque Universitaire, 1945), pages 9-10, (although he spends much of the book discussing give and take on the Norman frontiers, these are presented as adjustments of a stable border existing from 933: “Dès lors, la Normandie est fixée dans ses traits, peut-on dire, définitifs,” page 11); Douglas, “Rollo of Normandy,” pages 128-30; Michel de Boüard, “Le duché de Normandie,” in Histoire des institutions françaises au Moyen Age, volume 1 : Institutions seigneuriales, ed. Ferdinand Lot and Robert Fawtier, (Paris: Presses Universitaires de France, 1957), pages 2-6; Lucien Musset, “Considerations sur la genèse et la trace des frontières de la Normandie,” in Nordica et Normannica: Recueil d’études sur la Scandinavie ancienne et médiévale, les expéditions des Vikings et la fondation de la Normandie, Studia nordica 1 (Paris: Société des études nordiques, 1997), passim; Jean Yver, “Les premières institutions du duché de Normandie,” in I Normanni e la loro espansione in Europa nell’alto medioevo, Settimane di studio del Centro italiano di studi sull’alto medioevo 16 (Spoleto: Presso la Sede del Centro, 1969), page 312; Bates, Normandy before 1066, pages 8-9 and page 265, map 2. John Le Patourel, The Norman Empire (Oxford: Clarendon Press, 1976), pages 3-14, argues for a loosening of the three-stage theory, although he follows Lemarignier in seeing a relatively precocious development of a well-defined territorial principality. Searle first proposes a more radical break from the three-stage theory in Searle, “Fact and Pattern.”; Eleanor Searle, “Frankish Rivalries and Norse Warriors,” Anglo-Norman Studies 8 (1985): pages 198-213, and develops the argument further in Searle, Predatory Kinship, especially pages 40-54 and 100-05.
 Flodoard of Rheims, Annales, s.a. 925, discussed in Prentout, Étude critique, pages 275-6; Douglas, “Rollo of Normandy,” page 434.
 Flodoard of Rheims, Annales, s.a. 927: “Karolus igitur cum Heriberto colloquium petit Nordmannorum ad castellum quod Auga vocatur ibique se filius Rollonis Karolo committit et amicitiam firmat cum Heriberto.” Discussed in Douglas, “Rollo of Normandy,” page 435; Eckel, Charles le Simple, pages 131-34.
 Rollo’s death is discussed in Prentout, Étude critique, pages 272-78; cf. Richer of Rheims, Historiae, 1.50.
 Adémar is the first to mention the marriage: “Acceptamque in conjugium Adelam, filiam Rosi Rotomagensis, genuit ex ea Willelmum Caput stupe.” Adémar de Chabannes, Chronique, edited by Jules Chavanon, Collection de textes pour servir à l’étude et à l’enseignement de l’histoire 20 (Paris: Picard, 1897), pages 143-42. Dudo expands on the story. Dudo of Saint-Quentin, De moribus, 3.47. William of Jumièges’ adaptation of Dudo contains additional information, including the name Gerloc and her missionary activity. William of Jumièges, Orderic Vitalis, and Robert of Torigni, The Gesta Normannorum Ducum, two volumes, edited and translated by Elisabeth M. C. van Houts, Oxford Medieval Texts (Oxford: Clarendon Press, 1992-1995), 3.3.
 Ruinus: Adémar de Chabannes, Chronique, pages 198, 202; Roso: Adémar de Chabannes, Chronique, pages 139, 140, 144, 148. Ruinus is from an earlier draft of the Chronique and Roso from a later draft. It is probably a coincidence that Richer has a chapter heading “Rollonis pyratae interitus suorumque ruina”, Richer of Rheims, Historiae, 1.50. On Adémar’s work, see Richard Landes, Relics, Apocalypse, and the Deceits of History: Ademar of Chabannes, 989-1034, Harvard Historical Studies 117 (Cambridge, MA: Harvard University Press, 1995), especially chapters 6 and 7. Rotlo: “Chronicon de Gestis Normannorum in Francia,” in Lamberti S. Audomari Canonici Liber Floridus: Codex authographus bibliothecae universitatis Gandavensis, ed. Albert Derolez (Ghent: J. Story-Scientia, 1968), s.a. 895; although this is probably a much later source. See Appendix 1 below for a discussion of the Chronicon. Flodoard, Richer, Dudo, and the scribe of the 918 charter all call him Rollo and we can safely assume that Rollo is what he was known by in the Latin Frankish community of the tenth century.
 Is it possible that in seeking an origin for Rollo/Hrollaugr, the bards found the Hrollaugr of the Heimskringla. However, when they realised that it was historically impossible for him to be Rollo of Normandy, they invented a brother for him, based upon the Göngu-Hrólfssaga and named Hrólfr? However, it is possible that the reason that Hrollaugr’s brother Hrólfr is mentioned in a poem which, if genuine, was written by their brother Einarr. Even if the poem is not genuine, it is very likely that the appearance of Hrólfr in the family (as an indistinct entity) predates the probably false identification of this Hrólfr with Rollo of Normandy. In the part of Flodoard’s annals that mention Rollo of Normandy, another Viking named Ragenold (equivalent to Old Norse Rognvaldr) is also mentioned in nearby passages but not with any implication of a connection to Rollo. To Icelandic scholars, the names Ragenold and Rollo occuring so close together might very well have reminded them about the Rognvaldr and his son Hrollaugr in the Icelandic sources. If they noticed that it was not reasonable to identify Hrollaugr (an Icelandic settler) and Rollo, they might have taken the opportunity to identify Rollo instead with Hrollaugr’s obscure brother Hrólfr. Or, as an alternate scenario, maybe it occurred in two stages, with Rollo being identified with Hrollaugr in an unknown earlier manuscript, and the identification later being switched to Hrólfr by a later writer who noticed the problem.