EARLY NORMANDY
Lesley Abrams
The year 2011 witnessed two 1,100th-anniversary celebrations relevant to the activities of Scandinavians overseas. On 2 September 911 a treaty designed to facilitate trade and promote good relations between the Byzantines and the Kievan Rus
was agreed in Constantinople.1 The treaty between the Frankish king Charles the
Simple and the Viking army led by Rollo that marked the beginning of Normandy
is almost virtual in comparison, in the absence of a text or a contemporary record;
but, according to Dudo of Saint-Quentin, it too was granted in 911, at Saint-Clairsur-Epte.2 The exact nature of the power of the Vikings on the Seine at that moment
is obscure: their takeover coincided with a hiatus in Frankish annal-writing, and no
contemporary evidence survives from church or court in Rouen. The earliest witness
is an original charter in King Charles’s name from March 918 which granted the
abbey of Croix-Saint-Ouen and its dependencies to the monastery of Saint-Germaindes-Prés, praeter partem ipsius abbatiae quam annuimus Normannis Sequanensibus, uidelicet Rolloni suisque comitibus, pro tutela regni.3 The charter refers to
the portio Nortmannorum but does not spell out its boundaries. From the 920s to
the 990s Reims’s historians Flodoard and Richer occasionally reported on Norman
affairs,4 but after what appears to have been almost a century without writing in
Normandy itself, there was a surge in the reign of Richard II (996–1026) – charters, hagiography, and, of course, Dudo’s history, much of it closely associated with
the dukes. These circumstances make early Normandy very dificult to reconstruct
and assess. It is not a neglected subject, however: David Bates’s lucid analysis in
1
For a translation of this text, see The Russian Primary Chronicle. Laurentian Text, ed. and trans.
S. H. Cross and O. P. Sherbowitz-Wetzor, Cambridge MA 1953, 65–8; it is discussed in S. Franklin and
J. Shepard, The Emergence of Rus 750–1200, London 1996, 103–7.
2
De moribus et actis primorum Normanniae ducum, ed. J. Lair (Mémoires de la Société des Antiquaires
de Normandie 23), 1865, 168; Dudo of St Quentin. History of the Normans, trans. E. Christiansen,
Woodbridge 1998, 48. D. C. Douglas, ‘Rollo of Normandy’, EHR 57, 1942, 417–36 at 428–9, supported
a date during the autumn of 911, but proposed that Dudo might have invented the location; Christiansen
(195, n. 201) suggested that other meetings at Saint-Clair-sur-Epte, including one in 945 between Richard
I and King Louis, could have inluenced Dudo’s choice. On this and other Viking treaties in Francia, see
P. Bauduin, Le monde franc et les vikings VIIIe–Xe siècle, Paris 2009, and his ‘Chefs normands et élites
franques, in IXe–début Xe siècle’, in Les fondations scandinaves en Occident et les débuts du duché de
Normandie, ed. P. Bauduin, Caen 2005, 181–94, esp. 189–91.
3
‘Except for the part of the abbey’s lands which we have granted to the Northmen on the River Seine,
that is, to Rollo and his companions, for the protection of the kingdom’; Recueil des actes de Charles
III, ed. P. Lauer, Paris 1940, 209–12 (no. 92).
4
Flodoard’s annal for 923 conirms that, ‘shortly before’, the Normanni had agreed an arrangement
with Charles and been given land: Les annales de Flodoard, ed. P. Lauer, Paris 1905, 16; The Annals
of Flodoard of Reims, ed. and trans. S. Fanning and B. S. Bachrach, Peterborough (Ontario) 2004, 9.
Flodoard’s Historia Remensis ecclesiae (Flodoard von Reims. Die Geschichte der Reimser Kirche, ed. M.
Stratmann, MGH Scriptores 36, Hannover 1998) was written in the 950s, Richer’s Histories in the 990s
(Richer von Saint-Remi. Historiae, ed. H. Hoffmann, MGH Scriptores 38, Hanover 2000, and Richer of
Saint-Rémi. Histories, ed. and trans. J. Lake, 2 vols, Cambridge MA 2011).
Anglo-Norman Studies 35.indb 45
24/04/2013 12:50:51
46
Lesley Abrams
his Normandy before 1066 remains the benchmark for subsequent discussion,5 and
Lucien Musset’s enduring scholarship has been extended in recent years by Francophone and Anglophone historians with intimate knowledge of both local and wider
Frankish contexts.6 Despite all this attention, there is no consensus about how things
stood in the tenth century, before the change of gear that followed the accession of
Richard II.
Although Rouen and Kiev offer many contrasts, it is a measure of Scandinavian
vitality overseas that the same year saw two such successes. The degree to which
the people involved belonged to the same world continues to be a matter for debate.
Recently I have argued for the utility of the concept of a Scandinavian diaspora in
the Viking age: deinitions of diaspora vary, but all rely on the essential element
of a shared but lexible identity, preserved over distance and shaped in dynamic
ways.7 In the early Middle Ages Scandinavians left their homelands and travelled
and settled as far aield as Greenland and the Upper Volga. As now, there were
many ways for diaspora connections to remain active, among them trade, migration,
diplomacy, marriage, war, and religious mission. The concept of diaspora allows us
to envisage a network of far-lung elite centres sharing a culture based on language,
origins, and ongoing contacts. Connections between these groups occurred across
large distances, but at the same time new regional identities were formed by the
interaction between incomer Scandinavians and their immediate localities.
This model gives some theoretical substance to the spread of a conspicuously
pan-Scandinavian culture across the North Atlantic, Britain and Ireland, northern
Europe, and along the Russian rivers. Its reach is very observable, articulated as
it was through dress accessories and art styles, a shared language and oral culture,
shared religious and mythological traditions, and a common, non-Latin, alphabet.
While Scandinavians tapped this diasporic identity and deployed their ‘Scandinavian-ness’ abroad, they also absorbed local resources and developed new, hybridized
societies driven by more proximate inluences. The model conceives of Scandinavian identity as a kind of artefact: every community would have had decisions to
make about what role it would play in their new environment. Judging from later
survivals, the Northern Isles of Scotland were at one end of the cultural spectrum
– thoroughly Scandinavianized, with Norse spoken until the modern period – and
Normandy at the other. Normandy, in fact, seems the odd one out. The tally of
material culture is extremely low: only a handful of metal objects, such as the two
oval brooches from Pîtres, no runic inscriptions, no sculpture or coins that namecheck Thor or Oðin, no preserved burials of traditional Scandinavian type.8 Nor is
5
D. Bates, Normandy before 1066, London and New York 1982. My debt to this book, especially
chapter 1, will be clear throughout this article, but references will, regrettably, be kept to a minimum for
reasons of space. Bates’s challenge that ‘a simple statement that the Scandinavian impact was small just
will not do’ (p.16) has inspired me to confront this problem.
6
Musset’s output was extraordinary. Nordica et Normannica: recueil d’études sur la Scandinavie
ancienne et médiévale, les expéditions des vikings et la fondation de la Normandie, Paris 1997, offers
a very useful but necessarily limited selection and a bibliography. See also Postérité de Lucien Musset,
ed. V. Gazeau and F. Neveux, Caen 2009, esp. P. Bauduin, ‘Lucien Musset et les débuts de la principauté
normande’, 27–34. E. Searle, Predatory Kinship and the Creation of Norman Power, 840–1066, Berkeley
CA 1988, offered a contrasting interpretation. The subject of early Normandy is vast and spans many areas
of expertise; unfortunately, only a fraction of the relevant scholarship can be cited in this short article,
which cannot do justice to the detail of the many pertinent debates. The CRAHAM website contains a
wealth of material, including up-to-date bibliographies (http://www.unicaen.fr/crahm/?lang=fr).
7
L. Abrams, ‘Diaspora and Identity in the Viking Age’, EME 20, 2012, 17–38.
8
P. Périn, ‘Les objets vikings du musée des antiquités de la Seine-Maritime à Rouen’, Recueil d’études
en hommage à Lucien Musset, Caen 1990, 161–88; E. Roesdahl, ‘What May We Expect? On the Problem
Anglo-Norman Studies 35.indb 46
24/04/2013 12:50:51
Early Normandy
47
Rouen’s ‘Viking’ archaeology a match for York’s or Dublin’s: a new street plan was
laid out, but without conirmed dates the authority behind this reconstruction of the
town’s core is uncertain.9 Furthermore, sites in Rouen more irmly dated to the tenth
century have not been accompanied, as I understand it, by any diagnostically ‘Scandinavian’ inds; but only a small proportion of the town has been excavated.10 While
Northumbria’s new elite made a point of displaying its Scandinavian connections on
metal, stone, and wood, there was apparently no such display in Normandy – and yet
its place-names match northern England’s in density and ‘Norseness’, and Scandinavian personal names were in some cases retained for generations. This apparently
contradictory relection of the Scandinavian impact provides the starting-point for
my investigation.
There is no question that early Normandy continued to have Scandinavian
connections. Richard I (942–96) and Richard II both had Scandinavian military
allies, for example.11 In terms of trade, there is little evidence of the so-called
‘Viking economy’ of mixed bullion and coin, but minting was tightly controlled
in the Carolingian period, and the Rouen regime, once established, soon issued its
own coinage.12 Bullion and foreign coins were likely to have been routinely melted
down and recoined without delay, leaving no evidence of their origin. In 2007,
however, a mixed hoard buried c. 895 was discovered at Saint-Pierre-des-Fleurs,
south of Elbeuf; its contents suggest that it was assembled in England and travelled
to Normandy in Scandinavian hands.13 Jacques le Maho has argued that Rollo and
his Vikings had already established a string of trading-places along the Seine before
911.14 While Rouen traded upstream as well, inds of Norman coins of the tenth and
of Vikings and Archaeology in Normandy’, in La progression des vikings, des raids à la colonisation, ed.
A.-M. Flambard Héricher, Rouen 2003, 207–14; A. Nissen Jaubert, ‘Implantations scandinaves et traces
matérielles en Normandie. Que pouvons-nous attendre?’, in Les fondations, ed. Bauduin, 209–23. For
two recent inds of Thor’s hammers, see J.-C. Moesgaard, ‘Deux bijoux vikings découverts en HauteNormandie’, Annales de Normandie 59 (2), 2009, 133–40, and J. Callais, ‘A Thor’s Hammer Found in
Normandy’, in Viking Trade and Settlement in Continental Western Europe, ed. I. S. Klæsø, Copenhagen
2010, 145–7; as increasing numbers of fakes are appearing on the market, it would be reassuring if all
new inds could be examined together with authentic objects from other regions.
9
J. le Maho, ‘Rouen à l’époque des incursions vikings (841–911)’, Bulletin de la Commission
Départemental des Antiquités de la Seine-Maritime 42, 1994, 143–202; for a recent critique, see
J. Nelson, ‘Normandy’s Early History since Normandy before 1066’, in Normandy and its Neighbours,
900–1250. Essays for David Bates, ed. D. Crouch and K. Thompson, Turnhout 2011, 3–15 at 13–14.
10 Most recently, see J. le Maho, ‘Avant et après les Normands: les lieux d’échanges dans l’espace
luvio-maritime normand au haut Moyen Âge (VIIe–Xe s.)’, forthcoming. I am extremely grateful to
Jacques le Maho for his help with my questions. Compare Aspects of Anglo-Scandinavian York, ed. R. A.
Hall et al., York 2004; it is noteworthy that the earliest known written Old Norse (henceforth ON) in
Latin script occurs on a coin of the 940s from York (cununc, ON konungr, ‘king’; 336).
11 Richard I turned to the rex paganus Sihtric in 943 (Les annales, ed. Lauer, 88; The Annals, trans.
Fanning and Bachrach, 38), and Richard II hosted Svein Forkbeard, king of Denmark, and Olaf
Haraldsson, later Norway’s king and patron saint: GND, II, 16–17, 24–7.
12 Royal control may have faltered by the late ninth century, when magnates and ecclesiastical houses
also issued coins, but there is no evidence of a shift to uncoined metal; see J. Lafaurie, ‘Numismatique:
des Carolingiens aux Capétiens’, Cahiers de civilisation médiévale 13, 1970, 117–37, esp. 130–5; for
Norman coins, beginning with those of William Longsword (c. 927–42), see J. C. Moesgaard, ‘A Survey
of Coin Production and Currency in Normandy, 864–945’, in Silver Economy in the Viking Age, ed.
J. Graham-Campbell and G. Williams, Walnut Creek CA 2007, 99–121.
13 T. Cardon, J. C. Moesgaard, R. Prot, and P. Schiesser, ‘Le premier trésor monétaire de type viking
en France: denier inédit d’Eudes pour Beauvais’, Revue numismatique, 6th ser., 164, 2008, 21–40.
14 J. le Maho, ‘Les Normands de la Seine à la in du IXe siècle’, in Les fondations, ed. Bauduin, 161–79;
see also L. Musset, ‘La Seine normande et le commerce maritime du IIIe au XIe siècle’, Nordica et
Normannica, 337–49, esp. 343–49.
Anglo-Norman Studies 35.indb 47
24/04/2013 12:50:51
48
Lesley Abrams
early eleventh centuries in northern England, the Hebrides, Denmark, and Russia,
although few,15 represent the continuing activity of trading networks connecting
Scandinavian centres.
Where settlement occurred in already occupied lands, historians have used Scandinavian cultural features to judge the volume of migration, based on the thinking
that a heavy cultural imprint indicates signiicant numbers. The absence of classic
indications of Scandinavian-ness has therefore fostered doubts about the Viking
settlers of Normandy. Vincent Carpentier, for example, has recently argued that
there was no signiicant immigration: in his view Scandinavian colonization is a
misconception, and attributing a Scandinavian identity to the Normans perpetuates ‘a romantic and often racist conception of regional history’, a myth driven by
linguists ‘infatuated with normannisme’.16 ‘Normanism’ and its élucubrations xénophobes et raciales are indeed undesirable.17 Also unhelpful, however (and potentially misleading), is the equation of numbers with cultural identity. Pierre Bauduin
has taken a different approach, contending that the apparent absence of Scandinavian-ness was a result of rapid assimilation: in his view, Scandinavian identity had
no particular currency in the tenth century, politically or culturally, and was quickly
rejected. Focussing attention on the Franks and their relations with the new lords
of Rouen, Bauduin has argued that an intense process of accommodation through
the usual means – military alliances, marriages, relationships of amicitia, and spiritual kinship – integrated the leaders of la première Normandie swiftly into the
Frankish elite.18 Both of these views play down the Scandinavian contribution, but
it remains the case that Normandy’s histories – and its name – continued to advertise
its northern origins,19 and the impact of Norse speech was widely imprinted on the
landscape.
Before attempting to explain this situation, we should irst question whether
Normandy’s absence of evidence is ‘real’. It is worth remembering that at least
some of the material that now supports the presence of substantial numbers of Scan-
15
J. C. Moesgaard, ‘Renovatio Monetae et la chronologie des monnaies de Richard Ier, duc de Normandie
942/945–996’, Bulletin de la Société Française de Numismatique 66, 2011, 125–33, and ‘Les échanges
entre la Normandie et la Baltique aux Xe–XIIe siècles’, forthcoming. I am indebted to the author for
all his help with the numismatic material. For other connections, see L. Abrams, ‘England, Normandy,
and Scandinavia’, in A Companion to the Anglo-Norman World, ed. C. Harper-Bill and E. van Houts,
Woodbridge 2003, 43–62, and E. M. C. van Houts, ‘Scandinavian Inluence in Norman Literature of the
Eleventh Century’, ANS 6, 1983, 108–21, esp. 120–21.
16 V. Carpentier, ‘Les vikings en Normandie: archéologie d’un paradoxe identitaire’, Les dossiers
d’archéologie 344, March/April 2011, 72–7, esp. 76–7. For a comparison with neighbouring Brittany (and
Bretonnitude), see W. Davies, ‘Franks and Bretons: The Impact of Political Climate and Historiographical
Tradition on Writing their Ninth-Century History’, in Frankland. The Franks and the World of the Early
Middle Ages: Essays in Honour of Dame Jinty Nelson, ed. P. Fouracre and D. Ganz, Manchester 2008,
304–21.
17 G. Nondier, ‘Le mythe viking en Normandie et ses paradoxes’, in L’héritage maritime des vikings en
Europe de l‘Ouest, ed. E. Ridel, Caen 2002, 503–12 and 532–3 at 512, splendidly translated as ‘jingoistic
and racial twaddle’ (p. 532).
18 Bauduin, Le monde franc, esp. 349–51; La première Normandie (Xe–XIe siècles), 2nd edn, Caen
2006.
19 Richard I could still be referred to in Richer’s obituary notice as pyratarum dux, ‘leader of the
pirates’, for example: ed. H. Hoffmann, 307; ed. and trans. Lake, II, 436–7. There are many studies
of this theme, from Musset’s ‘L’image de la Scandinavie dans les oeuvres normandes de la période
ducale (911–1204)’, Nordica et Normannica, 213–31, R. H. C. Davis’s The Normans and their Myth,
London 1976, and Bates’s comments in Normandy, esp. 56–7, to, more recently, E. Albu, The Normans
in their Histories: Propaganda, Myth and Subversion, Woodbridge 2001, and N. Webber, The Evolution
of Norman Identity 911–1154, Woodbridge 2005.
Anglo-Norman Studies 35.indb 48
24/04/2013 12:50:51
Early Normandy
49
dinavian immigrants in northern and eastern England was unknown until relatively
recently: hundreds of coins, ingots, brooches, and other metal objects have been
discovered by metal-detectorists since the activity caught the public imagination
and a revolution in academic attitudes took place in the 1990s.20 Jane Kershaw’s
recent study of brooch-inds, for example, concluded that metalwork continued to be
imported from Scandinavia to the English Danelaw throughout the tenth century and
that imitations with a very Scandinavian look were also produced locally.21 Of the
550 brooches in her corpus, only about twenty were known before the recent surge
in metal-detecting, and new inds continue to be made. The brooches in question are
mainly small, everyday objects, not the large oval brooches that would have been
harder to lose. The recent availability of this kind of material is helping to reshape
understanding of the Danelaw’s contacts with the homelands and focus attention
on the potential role of an ongoing Scandinavian identity. Admittedly, the context
was different in England. The irst half of the tenth century saw the West Saxons
extend their hegemony into areas which had been independent Scandinavian polities for ifty to eighty years: an opposing regional identity, articulated through dress,
could have served a purpose for which there was no need across the Channel. As
metal-detecting is illegal in France, however, it cannot be entirely safe to say that
similar evidence does not exist in Normandy; twenty or thirty years ago we did not
know that it existed in England. It is impossible to say whether metal-detecting in
Normandy on the English scale would produce material with a similarly transformative potential. A modest stray-ind from the Orne, a pierced Christiana religio coin
of Louis the Pious, could be exceptional – or the tip of an iceberg. Simon Coupland
has pointed out that since such coins, worn as ornaments, were common in graves
in Scandinavia in the irst four decades of the tenth century, this ind could have
belonged to an early settler.22
There are other factors to consider: the Church would have inhibited converts
from indulging in the kind of traditional funerary rites that would be archaeologically visible today, and such customs, which went out of fashion in Denmark in the
tenth century, may not have been popular with any remaining pagans with Danish
connections. Other types of evidence may have gone missing – sculpture, for
example, which acted as a signiicant manifestation of Anglo-Scandinavian identity.
In England our knowledge often derives from the visible reuse of tenth-century stone
fragments in later buildings, but few Norman churches display reused fragments of
any sort, let alone any decorated in Scandinavian art styles. The reversed-S serpents
at Evrecy have been judged to have parallels in Yorkshire,23 but Evrecy seems to be
quite exceptional. A comparison of the grave furniture of the early regimes at Rouen
and York would be enlightening, but Rouen Cathedral’s cemeteries have yielded
nothing to compare with the distinctive grave slabs under York Minster.24 Many
20
In 1997 the British government initiated the Portable Antiquities Scheme (PAS), which now covers
most of England and Wales; see http://inds.org.uk/.
21 J. Kershaw, ‘Culture and Gender in the Danelaw: Scandinavian and Anglo-Scandinavian Brooches’,
Viking and Medieval Scandinavia 5, 2009, 295–325.
22 S. Coupland, ‘Un témoin numismatique de la présence des vikings en Normandie?’, Bulletin de la
Société Française de Numismatique 62, 2007, 69–71.
23 L. Musset, ‘L’église d’Evrecy (Calvados) et ses sculptures préromanes’, Bulletin de la Société des
Antiquaires de Normandie 52, 1952–4, 116–68 at 147–8 (ig. 32); M. Baylé, ‘Réminiscences angloscandinaves dans la sculpture romane de Normandie’, ANS 13, 1990, 35–48 at 35–7.
24 J. Lang, York and Eastern Yorkshire. Corpus of Anglo-Saxon Stone Sculpture, III, Oxford 1991,
38–40 and 69–75. Compare J. le Maho, ‘Les fouilles de la cathédrale de Rouen de 1985 à 1993: esquisse
d’un premier bilan’, Archéologie médiévale 24, 1994, 1–49, esp. 31–41.
Anglo-Norman Studies 35.indb 49
24/04/2013 12:50:52
Anglo-Norman Studies 35.indb 50
Figure 1. An original charter of 1006, Musée de la Bénédictine of Fécamp, no. 3, formerly 1bis (RADN, no. 9). Dimensions of entire
charter: 66 x 47 cm.
24/04/2013 12:50:52
Early Normandy
51
years of destructive warfare in the ninth and tenth centuries and a radical rebuilding
programme in the twelfth combined to obliterate most early churches in Normandy,
but when rebuilding began under the dukes it showed no obvious Scandinavian
afinities. According to Maylis Baylé, these arrived much later, as part of a more
general inluence from England after the Conquest.25 Unfortunately, the possible
importance of wooden buildings and monuments, now lost, cannot be measured.26
With these reservations in mind, on present evidence the prevailing impression
is of an immigrant society which did not embrace multiculturalism. It did, however,
use Norse. The impact of the native tongue of the newcomers has often been downplayed, thanks in part to Dudo’s statement that Richard I was sent to Bayeux for
his education.27 Dudo’s explanation, put into the mouth of William Longsword (‘the
city of Rouen much prefers the use of Roman rather than Dacian eloquence, and
Bayeux uses the Dacian more often than the Roman tongue’), does not in fact say
what some have concluded, that Norse had ‘died out’ in Rouen; in any case, scholars
have found other motives for Dudo’s remarks.28 Even in the mid-eleventh century
an ecclesiastical author showed continuing familiarity with the language when he
referred to a ship ‘qui barbara lingua isnecchia dicitur’.29 Although the Rouen regime
may well have preferred the language of the Franks, it seems unlikely that Richard
I and Richard II conducted relations with their Norse-speaking allies in anything
but Norse.30 Rural language use is likely to have been subject to different forces,
but it is more observable, being relected, in different densities, across the region: in
vernacular vocabulary such as étac (‘rocky pile’) and ho (‘small bay’), for example,
derived from ON stakkr and hóp, and isigardus, ‘ishery’, from ON iskigarðr.31
Norse topographical words were frequently combined together in meaningful ways
to produce place-names, such as Caldebec (Caudebec), from ON kaldr, ‘cold’, and
bekkr, ‘stream’,32 and Haroloz (Harleur), ON hár, ‘high’, and (perhaps) ON lóð,
lood, tide (Figure 1).33 These, like the many hundreds of other major names with
ON elements – such as toft, þveit, dalr, lundr, bekkr, haugr, and hólmr (> tot, tuit,
dal, lon, bec, houge, houmet) – appear to have been formed in a Norse-speaking
25
Baylé, ‘Réminiscences’, 45–8.
M. Baylé, Les origines et les premiers développements de la sculpture romane en Normandie, Art
de Basse-Normandie 100 (n.d.), 24; M. Baylé, ‘Remarques sur les ateliers de sculpture dans le Cotentin
(1100–1150)’, Romanesque and Gothic: Essays for George Zarnecki, [no ed.], 2 vols, Woodbridge 1987,
I, 7–13, esp. 12–13. A igure clutching his forked beard with both hands in the nave wall of Bayeux
Cathedral, probably of early twelfth-century date, has striking Scandinavian parallels; see, for example,
J. Graham-Campbell, Viking Artefacts: A Select Catalogue, 1980, 24–25 and 214.
27 Dudo, ed. Lair, 221–2; trans. Christiansen, 97.
28 For example, F. Amory, ‘The Dönsk tunga in Early Medieval Normandy: A Note’, in American
Indian and Indoeuropean Studies. Papers in Honor of Madison S. Beeler, ed. K. Klar, M. Langdon, and
S. Silver, The Hague 1980, 279–89.
29 Inventio et miracula Sancti Wulfranni, ed. J. Laporte, Rouen and Paris 1938, 79.
30 E. Ridel, Les vikings et les mots: l’apport de l’ancien scandinave à la langue française, Paris 2009,
57. François de Beaurepaire argued that the language was spoken at least until the irst half of the
eleventh century: La toponymie de la Normandie: méthodes et applications, Paris 1969, 85.
31 Ridel, Les vikings, 123.
32 De Beaurepaire, commenting on the presence of two places of this name on the Seine, suggested that
Caudebec might have originally been the Norse name for the river; Les noms des communes et anciennes
paroisses de la Seine-Maritime, Paris 1979, 57.
33 Ridel, Les vikings, 209–10; A.Vincent, Toponymie de la France, Brussels 1937, 160, and E. Nègre,
Toponymie Générale de la France, 4 vols, Geneva, 1990–8, II, 1012, interpret the irst element as a
personal name, Hrolfr.
26
Anglo-Norman Studies 35.indb 51
24/04/2013 12:50:52
52
Lesley Abrams
Figure 2. An original charter of 1051x1066, Archives Départementales de la
Seine-Maritime (RADN, no. 205). Dimensions of entire charter: 29.5 x 19.5 cm.
environment.34 New Romance verbs were created from Norse nouns, and preixes
or sufixes added to create words unknown in Norse (e.g., dellage, from ON deill,
‘part, portion’).35 Borrowed words such as iskigarðr or æcer/akr (‘acre’) were in
such regular use in the eleventh century that they appear with declensional endings
in Latin charters (see Figure 1). Some place-names, such as Clarbec (clarus, ‘clear’,
and bekkr, ‘stream’), combine Romance and Norse (Figure 2). Dating these placenames is problematic in the absence of early forms, but while they do not have to
have been coined in this period, they could not have been coined before; and there
are enough eleventh-century witnesses to indicate that they were well established
by then. Once borrowed words moved into the local language, they continued to
form minor names – Le Bec, La Hougue, La Hoguette, La Londe – for many years.36
Place-names apparently combining Old English (OE) and Old Norse, on the other
hand (such as Elbeuf, interpreted as compounded from OE wella, ‘watercourse’,
+ ON búþ, ‘temporary dwelling’),37 might, if the attributions hold up to linguistic
scrutiny, offer evidence of the kind of language spoken in the Danelaw in the tenth
century. This kind of linguistic impact relects the vitality of Norse speech in combination with the local vernacular and seems therefore to testify to a signiicant Norsespeaking presence, not just in towns or at court but also in rural society.
The inluence of a language should not simply be measured by how long it
continues to be spoken, however: more analysis of the process of borrowing, both
34
‘Toft’, ‘clearing’, ‘valley’, ‘wood’, ‘stream’, ‘hill’ or ‘mound’, ‘island’; Vincent, Toponymie, 158–61,
lists twelve second elements; see also Nègre, Toponymie, II, 1011–26.
35 Ridel, Les vikings, 119, 193.
36 F. De Beaurepaire, ‘La diffusion de la toponymie scandinave dans la Normandie ducale’, Tabularia,
Études 2, 2002, 47–56 at 52–3.
37 De Beaurepaire, Les noms ... de la Seine-Maritime, 70 (and 75, Étalondes).
Anglo-Norman Studies 35.indb 52
24/04/2013 12:50:52
Early Normandy
53
from the linguistic and from the social perspective, would be useful. According to
Elisabeth Ridel, the language of the earlier Germanic invaders of Gaul added perhaps
400 to 500 words to the Romance vernacular and affected its grammar. In contrast,
only 145 or so Norse borrowings have been identiied in Normandy.38 As they relate
primarily to the maritime and commercial spheres, these borrowings have been
judged to relect only a minor, ‘niche’, contribution by the newcomers. This may be
too dismissive. Put simply, borrowing seems to occur because enough speakers of a
second language use it as an alternative to a native word, or because something new
needs a name. For example, new vocabulary for ships’ rigging is understood to have
been introduced into Normandy because the majority of users were Norse-speakers,
and/or because Scandinavian immigrants brought new technology with them. On
this principle, the place-name Le Tingland in the northern Cotentin, if it derives
from ON thing, ‘assembly’,39 could indicate that there was a signiicant new population there, or that a new form of government was introduced (or both). On the other
hand, there is no surviving Norse contribution to military vocabulary in Normandy,40
although it can hardly be denied that the Vikings constituted a signiicant military
presence. It might follow that if a Scandinavian community took up farming on the
local model, as is suggested by the continuity of estate structures, there may have
been little impact on language. And so it seems: only a small number of examples of
imported ‘rural vocabulary’ have been identiied – forlenc (OE furlang, ‘furlong’),
wendinc (OE wending, ‘turning’), hovelland (OE heafodland, ‘headland’), croute
(‘croft’), gare (OE gara, ‘gore’), and estrac (OE straca, ‘strip’)41 – and mansloth,
from ON mannshlútr, a common word for a holding in the Danelaw, occurs in a
single charter.42 It would be worth exploring if some of the Old English terms could
instead be of Norse origin; in the absence of contemporary Scandinavian evidence
they have usually been understood as straightforward borrowings from England.
Without tenth-century charters, these terms necessarily appear in much later documents, although OE æcer (or ON akr), which occurs across the province, is attested
as early as 1006 (hacreis, see Figure 1). Musset argued that all the terms were likely
to be contemporary, and he associated them with early tenth-century immigrants
from the Danelaw, thanks to a reference in the Anglo-Saxon Chronicle to a jarl
Thorketill leaving England with his army to go ofer sæ on Froncland.43
More signiicant, perhaps, are the borrowed terms relating to landholding that are
embedded in place-names. Although most major towns kept their names, a radical
38
Ridel, Les vikings, 108–14; E. Ridel, ‘The Linguistic Heritage of the Scandinavians in Normandy’, in
Scandinavians and Europe, 800–1350: Contact, Conlict and Co-Existence, ed. J. Adams and K. Holman,
Turnhout 2004, 149–59.
39 J. Renaud and E. Ridel, ‘Le tingland: l’emplacement d’un þing en Normandie?’, Nouvelle revue
d’onomastique 35–6, 2000, 303–6.
40 Note, however, two thirteenth-century examples of ON húskarl, ‘retainer’: Rogerius Huscaille and
vallis Huscalli (Ridel, Les vikings, 104).
41 L. Musset, ‘Pour l’étude des relations entre les colonies scandinaves d’Angleterre et de Normandie’,
and ‘Les apports anglais en Normandie de Rollon à Guillaume le Conquérant (911–1066)’, Nordica
et Normannnica, 145–56 and 447–66 (at 454–60); M. Arnoux and C. Maneuvrier, ‘Le pays normand:
paysages et peuplement (IXe–XIIIe siècles)’, Tabularia, Études 3, 2003, 1–27, esp. 14–17; see also
G. Chartier, ‘De quelques toponymes normands’, Nouvelle revue d’onomastique 35–6, 2000, 265–301.
42 D. C. Douglas, ‘The Rise of Normandy’, Pro. Brit. Acad. 33, 1947, 101–30 at 103–4; L. Musset, ‘Un
type de tenure d’origine scandinave en Normandie: le mansloth’, in Nordica et Normannica, 441–6.
43 ‘Les apports’, 449; ASC 916A (there is some confusion over the date, which has been altered in the
manuscript; The Anglo-Saxon Chronicle: A Collaborative Edition, MS A, ed. J. Bately, Cambridge 1986,
66). That this refers to Normandy is an assumption.
Anglo-Norman Studies 35.indb 53
24/04/2013 12:50:52
54
Lesley Abrams
renaming occurred in the countryside. Around 1055 the abbey of Saint-Florent, near
Saumur, asked Duchess Matilda for help in its effort to reclaim ancient possessions
in the Cotentin which it had been granted some time before by one of the dukes: a
charter asserts that ‘if by chance any of these [places] should be unknown to you
or yours, you should not suppose that they have ceased to exist, but rather that
they have lost their names’. Flottemanville – ‘villa of Flotman’, or ‘villa of the
seamen’ – seems to have been one of the estates involved, exemplifying the problem.44 New names include a small number which compound the imported element
thorp with Scandinavian personal names (e.g., Torgistorp); eleven simplex forms
with a deinite article are also recorded (Le Torp, Le Tourps). This suggests that
the term was adopted into the local dialect, probably with the same meaning as
its English counterpart, ‘secondary settlement’.45 Thorp presumably entered the
lexicon in Normandy because it was useful. It is not clear how many rural sites
which local inhabitants would have described as secondary settlements in their own
language existed before the Viking takeover; it does seem, however, that a word was
borrowed for that purpose when at least some people were speaking Norse, from
which we might conclude that changes in rural organization were taking place at the
time. Much more common than thorp is delle, ‘portion of workable land’ (ON deill).
It has been suggested that a delle could have been the unit of distribution in the tenth
century, or perhaps the component of a new type of tenure; it is very well attested
in the minor names of the Danelaw.46 Place-names in -tuit or -thuit, from ON þveit,
‘clearing’, testify to land clearance, probably associated with an intensiication of
farming in previously wooded land.47 The many names in -tot (OE ‘toft’) plus a
personal name, both Norse (Hatteintot) and Frankish (Robertot), could represent a
general reorganization of landholding, especially as in some parts of Normandy the
older -acum names often seem to have been replaced by names in -tot.48
New place-names were frequently formed with Scandinavian personal names.
Historians recognize that many factors determine naming strategies, but in the
context of early Normandy it seems reasonable (in principle, if not in every case)
to accept the likes of names such as Hásteinn and Thorgisl as plausible ethnic identiiers. Although Rollo’s family abandoned Scandinavian names, their retention by
others seems signiicant in terms of cultural politics and family identity. Jean Adigard
des Gautries suggested, for example, that three women’s names which are unknown
in Scandinavia – Anschella, Osmunda, and Torfreda – might have been created in
Normandy as calques on male names.49 Some names suggest origins elsewhere,
44
Quorum si forte aliqua tibi tuisque fuerint incognita, non tamen credas hec omnino non esse, sed
potius nomina perdidisse; P. Marchegay, ‘Chartes normandes de l’abbaye de Saint-Florent’, Mémoires
de la Société des Antiquaires de Normandie 30, 1880, 666–8, no. 2. I am grateful to Richard Sharpe for
comments on this passage. On the identiication, see Musset, ‘Essai sur le peuplement’, 396.
45 J. Adigard des Gautries, ‘Études de toponymie Normannique I: les noms en -torp’, Études germaniques
6, 1951, 3–10; although few in number, they are found across Normandy (with a cluster in the Cotentin);
for England see P. Cullen, R. Jones, and D. N. Parsons, Thorps in a Changing Landscape, Hatield 2011.
46 Arnoux and Maneuvrier, ‘Le pays normand’, 15–16; like Ridel (Les vikings, 95), they have associated
its diffusion with the spread of open-ield farming, although the date for this remains uncertain.
47 L. Musset, ‘Essai sur le peuplement de la Normandie (VIe–XIIe siècle)’, Nordica et Normannica,
389–402 at 398; Å. K. H. Wagner, ‘Les noms de lieux issus de l’implantation scandinave en Normandie:
le cas des noms en -tuit’, in Les fondations, ed. Bauduin, 241–52.
48 De Beaurepaire, ‘La diffusion’, 53.
49 J. Adigard des Gautries, ‘Les noms de personnes d’origine scandinave dans les obituaires de
Jumièges’, in Jumièges: congrès scientiique du XIIIe centenaire, 2 vols, Rouen 1955, I, 57–67 at 67.
Anglo-Norman Studies 35.indb 54
24/04/2013 12:50:52
Early Normandy
55
such as OE Dunstan and Beorhstan, represented in the place-names Dénestanville
and Brestanville.50 No such individuals appear in written sources, which could place
the names in the tenth century, before charters came back into use.51 Danestanvilla is
attested in 1051,52 and at least one of several Englesquevilles (anglisca uilla) occurs
in a charter of 1014.53 The English æthelings, exiled in Normandy from 1016 for over
twenty years,54 might conceivably provide a later context for some of these placenames, as would 1066. Musset made the point, however, that, after the Conquest,
English names had little prestige and were unlikely to have been imported then.55
Gaelic personal names, found only in western Normandy, are similarly and almost
inevitably unattested until they appear in charters: Niellus in 1013x1020, for example.56 Contacts between western Normandy and the Celtic parts of Britain had a long
history, however, and some may precede (or indeed post-date) the Viking period.57
Dicuil, Duncan, and Niall are found in the place-names Digulleville, Doncanville,
and Néhou,58 and Néel, Beccan, Murdac, and Patric became family names. Why
Gaelic personal names were retained as family names and English ones were not is
unclear: perhaps differences of status and lordship were involved.59 The irony is that
while English and Gaelic personal names may potentially be associated with Scandinavians who came to Normandy via one of the overseas settlements, they draw
attention away from the much more numerous ‘bog-standard’ Norse names, whose
bearers could have come from anywhere in the Scandinavian diaspora, including
directly from the homelands.
50 F. de Beaurepaire, ‘Les noms d’anglo-saxons contenus dans la toponymie normande’, Annales de
Normandie 10, 1960, 307–16 (with map); G. Fellows-Jensen, ‘Les noms de lieu d’origine scandinave
et la colonisation viking en Normandie’, Proxima Thulé 1, 1994, 63–103 (99 for a map, also in Nordica
et Normannica, 440). De Beaurepaire has since identiied more English names, such as Cynewald (in
Canouville) and Edwin (Yainville): Les noms… de la Seine-Maritime, 56 and 166.
51 Fellows-Jensen, ‘Les noms’, 96.
52 De Beaurepaire, Les noms… de la Seine-Maritime, 66 (without clarifying whether it occurs in an
original document or a later copy).
53 RADN, no. 15, an apparent original or contemporary copy; de Beaurepaire, ‘Les noms d’anglosaxons’, 311–12; F. de Beaurepaire, ‘Les noms de lieux du Calvados attestés entre 911 et 1066’, Annales
de Normandie 2, 1952, 209–28, and 3, 1953, 22–36 and 135–48 at 22.
54 S. Keynes, ‘The Æthelings in Normandy’, ANS 13, 1990, 173–205.
55 Musset, ‘Les apports’, 463.
56 RADN, no. 23; L. Musset, ‘Participation de vikings venus des pays celtes à la colonisation scandinave
de la Normandie’, Nordica et Normannica, 279–96 (map at 440); E. Ridel, ‘From Scotland to Normandy:
The Celtic Sea Route of the Vikings’, in West over Sea: Studies in Scandinavian Sea-Borne Expansion
and Settlement before 1300, ed. B. Ballin Smith, S. Taylor, and G. Williams, Leiden 2007, 81–94 (map at
85). Dudo was keen to associate the dukes with Ireland: ed. Lair, 127, 192, 265, 282; trans. Christiansen,
14, 69, 140, 156.
57 J.-M. Picard, ‘L’Irlande et la Normandie avant les Normands (VIIe–IXe siècles)’, Annales de
Normandie 47, 1997, 3–24.
58 Musset, ‘Participation’, 288–91; Ridel, ‘From Scotland to Normandy’, 81–3; F. de Beaurepaire, Les
noms des communes et anciennes paroisses de la Manche, Paris 1986, 157, for La Meurdraquière and
the family name Murdach.
59 Fellows-Jensen, ‘Les noms’, 100. De Beaurepaire suggested that a pagus-name, *Helgenes, and a
place-name, Helgevilla (Helleville), potentially identiied a local Scandinavian lord, Helgi (Les noms
… de la Manche, 136–7); there are complications with the spellings, however. For Breton names in
Normandy, compare É. von Torhoudt, ‘Les Bretons dans les diocèses d’Avranches et de Coutances (950–
1200 environ): une approche onomastique de la question de l’identité’, in Bretons et normands au moyen
âge: rivalités, malentendus, convergences, ed. P. Bouet, B. Merdrignac, F. Neveux, and J. Quaghebeur,
Rennes 2008, 113–44.
Anglo-Norman Studies 35.indb 55
24/04/2013 12:50:52
56
Lesley Abrams
Adigard des Gautries identiied eighty-two certain (and dozens more potential)
Scandinavian personal names in Normandy,60 many of them – from the familiar
Hásteinn and Thorgisl to others, such as Óbeini or Ámundr, unattested in Scandinavia – compounded with the Frankish place-name element -ville. Names such as
Hastingivilla and Amundivilla (now Hattenville and Emondeville) appear in the
earliest surviving records, and it has sometimes been assumed that they were coined
by the Scandinavians themselves. More investigation is required, however, to look
for signs that these names were formed in Norse, as shown, for example, by the
genitive in Helperby, Yorkshire (Hjalpar by, ‘settlement of Hjalp’).61 Genitives of
names in -ville are, if anything, in Latin, and (without the beneit of detailed study)
they have the look of names coined by Romance-speaking locals or Latinate recordkeepers to refer to the new cohort of Scandinavian lords.62 It seems too that only
Norse personal names, and not common nouns or adjectives, were compounded
with -ville, although the identiication of almost all doubtful or ambiguous irst
elements as personal names (as in Flottemanville, above) may falsely reinforce this
contrast. If real, it would distinguish the names in -ville from England’s ‘Grimstonhybrids’, whose irst elements include a greater range of Norse word types.63 The
earliest surviving original charter, dated 1006, includes the grant of the church of
Scrotiuilla (now Ecretteville) (see Figure 1).64 It is unclear whether Skrauti of Scrotiuilla was the man who held the land in 1006 or the original Scandinavian recipient;
if the latter, the name would represent the moment of takeover and distribution of
lands to new holders, potentially as early as 911. It is perhaps more likely, however,
that this kind of name changed as landholders succeeded one another. A connection between names and lordship was certainly understood in the twelfth century:
Robert of Torigni observed that one of the sisters of Richard II’s wife, Gunnor,
married Turulfus of Pont-Audemer, the son of a certain Torf, ‘after whom several
towns are called Torfuille to this day’.65 I suspect that names were ixed only when
documents became common. Names in -ville were also compounded with Frankish
names or titles (Quiberville from Guibertus, Contevilla from conte), and although
the assumption is often made that they pre-date 911, the lack of early records leaves
the context of their formation, and their relation to Scandinavian names in -ville,
open to question.
Many years ago Musset pointed out that a comparison of polytiques of the ninth
century with documents of the twelfth would give the impression that nothing much
had happened on Normandy’s large estates in the intervening centuries, whereas the
names and place-names in surviving charters would suggest that a complete bouleversement had swept away all existing settlements.66 The challenge is to explain
60 J. Adigard des Gautries, Les noms de personnes scandinaves en Normandie de 911 à 1066, Lund
1954, 71–243.
61 For a summary of the English evidence, see G. Fellows-Jensen, ‘Danish Place-Names and Personal
Names in England: The Inluence of Cnut?’, in The Reign of Cnut: King of England, Denmark and
Norway, ed. A. R. Rumble, London 1994, 125–40 at 134.
62 The medial vowel of Scrotiuilla, for example, could be a Latin genitive; the ON genitive would be
Skrauta.
63 For this category of name, see G. Fellows-Jensen, Scandinavian Settlement Names in Yorkshire,
Copenhagen 1972, 112–21.
64 RADN, no. 9; illustrated in C. H. Haskins, Norman Institutions, Cambridge MA 1918, plate 1.
65 GND, II, 268–9. De Beaurepaire gave c. 1034 as the earliest attestation of Tourville-sur-PontAudemer, without identifying the source: Les noms des communes et anciennes paroisses de l’Eure,
Paris 1981, 200.
66 Musset, ‘Les apports’, 251.
Anglo-Norman Studies 35.indb 56
24/04/2013 12:50:52
Early Normandy
57
this puzzle. One possible solution is to envisage a continuity of local administration, coupled with signiicant displacement of landholding personnel and some reorganization of holdings, some of which occurred in a Norse-speaking environment.
Further, more detailed study would reine or replace this general model with a better
one. One problem to be faced is that, although most scholars have taken the modern
map of identiiably Scandinavian place-names as a representation of the distribution of Scandinavian settlement, it does not necessarily represent the situation in
the tenth century. Some topographical names, as we have seen, were a product
of language borrowing, and could date from any time after the words from which
they are formed entered the lexicon. Some early place-names, on the other hand,
will have disappeared. Ridel has warned that subsequent urbanization and industrialization could have been responsible for blank areas on the distribution map.67
Other, more contemporary factors may also have been in play, judging from the
English evidence. David Parsons identiied twenty-ive names in Domesday Book
compounded from OE tun and a Scandinavian personal name in Suffolk, but none
in neighbouring Essex; yet the ratio of Norse to English personal names among
landholders in both counties was identical.68 He pointed out that Domesday Book
revealed a marked distinction in social organization in Suffolk and Essex, the former
having numerous free peasants, the latter none. Leaving aside the argument about
the origins of free peasants, this example suggests that variations in social structure
and distinctions of tenure could lie behind different patterns of name-coining.
Although this exercise cannot be repeated in Normandy, its place-names, carefully handled, are nevertheless a major source of information for the historian. Their
analysis is not trouble-free, however, as Frank Stenton pointed out as long ago as
1945.69 Early spellings are essential to put the study on a secure footing, yet few
are available in the absence of tenth-century charters or an early province-wide
survey. Documents, when they do appear, tend to be preserved in large ecclesiastical archives, and areas without major religious houses are consequently poorly
represented. Names in charters have all been Latinized and have therefore been
transformed twice: by the transition to writing and by the inluence of the local
vernacular.70 Lost names have not been systematically collected, and names for
smaller places – hamlets, ields, landscape features – have often been neglected.
When François de Beaurepaire compiled his gazetteer of commune-names of the
Manche in 1969, he was obliged to exclude the names of another 18,708 recorded
minor places.71 More recent work has aimed to absorb minor names; however,
because they are most often collected from the modern map, their analysis is even
more complicated by issues of transmission and date than that of major names.72
67
Les vikings, 123–5; See also G. Gaudefroy, ‘Survivances de l’empreinte norroise dans la partie
orientale de la Seine-Maritime (ancien pays du Talou): l’apport de la microhydronymie et de la
toponymie’, Nouvelle revue d’onomastique 35–6, 2000, 209–63.
68 L. Abrams and D. N. Parsons, ‘Place-Names and the History of Scandinavian Settlement in England’,
in Land, Sea and Home: Proceedings of a Conference on Viking-Period Settlement, ed. J. Hines, A. Lane,
and M. Redknap, Leeds 2004, 379–431 at 423.
69 F. M. Stenton, ‘The Scandinavian Colonies in England and Normandy’, repr. in Preparatory to
Anglo-Saxon England, ed. D. M. Stenton, Oxford 1970, 335–45, esp. 337. For a thorough, more recent,
review of the material, see Fellows-Jensen, ‘Les noms’ (with a shorter version in English, ‘Scandinavian
Place-Names and Viking Settlement in Normandy: A Review’, in Namn och Bygd 76, 1988, 113–37).
70 Adigard des Gautries, Les noms, 244–9.
71 De Beaurepaire, Les noms… de la Manche, 17.
72 Arnoux and Maneuvrier, ‘Le pays normand’, and Deux abbayes de Basse-Normandie. Notre-Damedu-Val et le Val-Richer, Flers 2000 (Le Pays Bas-Normand 237–8). See also L. Musset, ‘Les domaines
Anglo-Norman Studies 35.indb 57
24/04/2013 12:50:52
58
Lesley Abrams
Perhaps because of obstacles such as these, place-names remain a regrettably
underdeveloped resource for historians of early Normandy.73 Like all sources, they
need care, but there is much to be learned from the study of names in charters, for
example, or the mapping of types of place-name in relation to geology and land
use; and while further analysis of place-names across the region may increasingly
enlighten us about the process of settlement, their interpretation in closely local
contexts may prove to be even more revealing.
Words and names are not being enlisted here to suggest that the Scandinavians
who came to Normandy overwhelmed the locals, nor that they brought a packaged,
ethnically deined way of life with them which they imposed when they settled.
Trying to reconstruct the nature of early Norman society by hunting ethnically Scandinavian elements, preserved in fragmentary form in later texts, has yielded only
a small harvest, and it would be misleading to aggregate the evidence and apply
it across the territory that became Normandy. Independent enterprise and freedom
from Frankish constraint may account for the hints in later periods of the exercise
of different social norms – of outlawry, for example, or the death penalty for theft,
or maritime law, such as rights to shipwreck; but such elements would have been
only a part of the new way of doing things made possible by settlement. The English
Danelaw, for example, was different from the rest of England in part because it was
‘Danish’, but also because it was conquered. If Charles’s grant gave Rollo access
to existing institutions, it may also have offered an opportunity to push for more,
to demand beneicial tenures in exchange for political support, along the lines of
the privileges requested from the French king by Alan Barbetorte in 939, for example.74 To what extent the regime that took over on the Seine in 911 expected the
authority that it had been granted to be long-term can never be known: there were
precedents, some quite short-lived. Stenton, among others, deemed the arrangement
of 911 merely to have provided the Vikings with a base for future operations, but
Musset conceived of an organized, managed, transfer of authority.75 It may be optimistic to imagine a consistent programme: the grants of 924 and 933, conceded by
a rival (and ultimately unviable) king, might have produced quite different results
from that of 911. We do not really know how or when land changed hands, especially in those regions far removed from Rouen. Royal land could conceivably have
been appropriated wholesale; it was in ducal hands in the eleventh century.76 Written
evidence suggests little delay in the distribution of ecclesiastical property: when
Gerard of Brogne wanted to restore Saint-Wandrille c. 944, even the support of
Richard I failed to convince the men holding the lands to hand them over, because,
they said, their possessions had been hard-won.77
In the last third of the ninth century Scandinavians took over the territories of
two long-established English kingdoms and substantial parts of another. York’s new
de l’époque franque et les destinées du régime domanial du IXe au XIe siècle’, Bulletin de la Société
des Antiquaires de Normandie 49, 1942–5, 7–97 at 55–8, Gaudefroy, ‘Survivances’, and Chartier, ‘De
quelques toponymes’, and ‘La microtoponymie anglo-scandinave du département de l’Orne’, Annales de
Normandie 55, 2005, 125–34.
73 My knowledge of this ield is far from comprehensive, however, and important work may be in
progress.
74 At least according to the Chronicle of Nantes: F. Neveux, ‘L’héritage des vikings dans la Normandie
ducale’, in L’héritage maritime, ed. Ridel, 101–17 at 113.
75 Compare Stenton, ‘The Scandinavian Colonies’, 344–5, and Musset, ‘Les domaines’, 48–9.
76 Musset, ‘Les domaines’, 49–51, and ‘Origines et nature du pouvoir ducal en Normandie jusqu’au
milieu du XIe siècle’, Nordica et Normannica, 263–77 at 266.
77 Inventio et miracula, ed. Laporte, 23–4.
Anglo-Norman Studies 35.indb 58
24/04/2013 12:50:52
Early Normandy
59
rulers may have achieved their position with the assistance of the Church,78 but
the only surviving evidence of a formal agreement is the treaty relating to East
Anglia between the Viking king Guthrum and Alfred, king of the neighbouring West
Saxons, made some time between 880 and 890 and preserved in a late eleventh- or
early twelfth-century collection.79 This frið (‘peace’) treated the two rulers as equals,
and its text deined the border, established procedures for dispute resolution, and
regulated trade. It also gave the settlers a legal identity. As the surviving treaties
between Kiev and Byzantium deal with Scandinavians off their patch, the treaty of
Alfred and Guthrum is arguably more relevant to understanding the arrangement
between Charles and Rollo. Yet Guthrum’s acquisition of East Anglia may have
been less oficial and more opportunistic. The role of the treaty is unclear – whether
it merely recognized Guthrum’s takeover of some (putative) pre-existing royal infrastructure, for example, or was actually instrumental in establishing his position as
the ruler of a deinite territory. After a complete dislocation, minting revived in East
Anglia in the 880s with imitations of West Saxon coins, followed c. 895 by the
Saint Edmund pennies, which made explicit reference to the last independent East
Anglian king.80 Although the concentration of major Scandinavian place-names is
less dense in East Anglia than in other parts of England, there appears to have been
a greater impact on hundred-names (at least ten of Norse origin), and thus perhaps
on administration.81
Consensus on the issue of continuity in Normandy remains elusive.82 Considering the volatility of the irst half of the tenth century, in particular, there was
probably signiicant disturbance, yet there is no evidence for a revolution in rural
administrative infrastructure: only four pagi acquired new Scandinavian names.83
Jean Yver argued that enough Carolingian infrastructure survived to allow Rollo and
his successors to use the Carolingian legacy to ‘make’ Normandy a proper state;84
but could this have been true everywhere in equal measure? And was the entire
territory under royal authority to the same degree before 911? A licensed takeover,
as opposed to informal or unauthorized possession, would presumably have encouraged the continuity of whatever public power had previously been exercised in the
interest of a centralized authority. On the other hand, the notion of ‘public power’
78
Historia de Sancto Cuthberto. A History of Saint Cuthbert and a Record of his Patrimony, trans. T. J.
South, Woodbridge 2002, 53 and 87–8.
79 The Laws of the Earliest English Kings, ed. and trans. F. L. Attenborough, New York 1963, 98–101;
S. Keynes and M. Lapidge, Alfred the Great. Asser’s ‘Life of King Alfred’ and Other Contemporary Sources,
Harmondsworth 1983, 171–2; P. Kershaw, ‘The Alfred-Guthrum Treaty: Scripting Accommodation and
Interaction in Viking-Age England’, in Cultures in Contact: Scandinavian Settlement in England in the
Ninth and Tenth Centuries, ed. D. M. Hadley and J. D. Richards, Turnhout 2000, 43–64.
80 M. Blackburn, ‘Currency under the Vikings. Part 1: Guthrum and the Earliest Danelaw Coinages’,
and ‘Part 2: The Two Scandinavian Kingdoms of the Danelaw, c. 895–954’, British Numismatic Journal
75, 2005, 18–43, and 76, 2006, 204–26.
81 K. I. Sandred, ‘Viking Administration in the Danelaw: A Look at Scandinavian and English HundredNames in Norfolk’, in Developments around the Baltic and the North Sea in the Viking Age, ed.
B. Ambrosiani and H. Clarke, Stockholm 1994, 269–76; Abrams and Parsons, ‘Place-Names’, 415–16.
82 Bauduin, La première Normandie, 25–34, Bates, Normandy, 95. Compare, for example, F. Lifshitz,
‘La normandie carolingienne: essai sur la continuité, avec utilisation de sources négligées’, Annales de
Normandie 48, 1998, 503–24, L. Musset, ‘Gouvernés et gouvernants dans le monde scandinave et dans
le monde normand (XIe–XIIe siècle)’, Nordica et Normannica, 415–37, and Searle, Predatory Kinship,
3–5.
83 Haga, Sarnes, and Helgeres (*Helgenes) in the Cotentin and Hulmus in the Hiémois; Fellows-Jensen,
‘Les noms’, 75.
84 J. Yver, ‘Les premières institutions du duché de Normandie’, Settimane di studio del centro italiano
di studi sull’alto medioevo 16, 1968, 299–366 and 589–98, esp. 365–6.
Anglo-Norman Studies 35.indb 59
24/04/2013 12:50:53
60
Lesley Abrams
may overestimate the importance of regulation from above in the ordinary matters
of everyday life. Certainly, what became Normandy had a recognizably different
identity from other regions of northern France: Musset pointed out how certain
features of social organization, such as rural mayors and the tripartite organization of society, stopped at the borders of the province, as did its distinctive rural
terminology. In the absence of evidence for the genesis of these characteristics, he
hypothesized that they were imposed by the ducal administration of estates.85 The
secret of Normandy’s success – the exceptional control exercised by its dukes – by
this reckoning could be back-projected to the kings’ transfer of rights and lands.
Rollo and his army were not the only Vikings on the scene, however. When
Thorketill left England for Francia,86 he may have been acting on intelligence:
thanks to the disturbances of the ninth century and the instability of the early tenth,
it would have been a good time for sons to pursue opportunities discovered by a
generation of raiding fathers. Resistance to Rouen’s attempt to expand into western
Normandy in 933 seems to have been Scandinavian-led, for example: Dudo identiied its leader as Riulf (ON Herjolfr?).87 Richard I’s allies of the 940s and 960s,
spectacularly invoked by Dudo as Viking mercenaries,88 may have been fresh legs
from overseas or, alternatively, men independent of Rouen and of the inluence of
its archbishop (if Dudo’s description of them as pagans is to be trusted). In 945,
after the murder of William Longsword and renewed Frankish attempts to claw back
power in Normandy, the men of Rouen appealed to a certain Harald Nordmannus,
who ‘was in command at Bayeux’.89 This evidence, admittedly spare, nevertheless
suggests that in the tenth century Scandinavian power in Normandy may have been
maintained by shifting personal relationships between men whose authority was not
always institutional or embedded in local traditions. Some may have depended on
treaties or other unattested arrangements with Bretons, Franks, or even the Vikings
of Rouen.90 Rural settlements and commercial communities based on urban markets
and river trade are almost undocumented, but, as we have seen, their Norse vocabulary is vividly relected in the local vernacular and in place-names. Arguably, Scandinavian identity would have played out differently in contexts such as these. If so,
a much more motley Scandinavian presence – of different origins and social status,
each group with its own relationship with a local Frankish matrix – could have
characterized the tenth century.
85
Musset, ‘Origines’, 271–7, ‘Les domaines’, esp. 77–8; see also Bates, Normandy, 147–88; for the
role of the aristocracy in this achievement, see L. Musset, ‘L’aristocratie normande au XIe siècle’, in La
noblesse au moyen age XIe–XVe siècles: essais à la mémoire de Robert Boutruche, ed. P. Contamine,
Paris 1976, 71–96; Bates, Normandy, 99–106, 111–21, 134–5.
86 See note 43.
87 Dudo, ed. Lair, 187–90 and 218; trans. Christiansen, 64–8, and 94; the description of Riulf as a
‘manifold blasphemer’ may hint that he was pagan, though ‘perjurer’ could simply indicate disloyalty to
Rouen; Adigard des Gautries pointed out that a Frankish name, Hairulf, is also represented in the region:
Les noms, 62–3.
88 Dudo, ed. Lair, 239–42, and 276–7 and 282; trans. Christiansen, 114–17, and 150–51 and 156–7.
89 Qui Baiocis praerat: les annales, ed. Lauer, 98, The Annals, trans. Fanning and Bachrach, 42; Dudo,
ed. Lair, 239–40, trans. Christensen, 114 and 214, n. 358; É. van Torhoudt, ‘La résistance franco-bretonne
à l’expansion normande dans le nord-ouest de la Neustrie (924–954): une marche de Normandie?’, in Le
pouvoir et la foi au moyen âge en Bretagne et dans l’Europe de l’Ouest, ed. J. Quaghebeur and S. Soleil,
Rennes 2010, 601–17 at 614, argued that Harald remained on the scene as an ally of the Franks into
the 950s; B. Hudson, Viking Pirates and Christian Princes. Dynasty, Religion, and Empire in the North
Atlantic, Oxford 2005, 65–8, identiied him as a ‘disaffected noble’ from Denmark who subsequently
founded a dynasty in the Irish Sea.
90 Van Torhoudt, ‘La résistance’, 614.
Anglo-Norman Studies 35.indb 60
24/04/2013 12:50:53
Early Normandy
61
The ‘miscellaneous character’ of Scandinavian settlement has long been acknowledged91 – in particular, the difference between upper and lower Normandy. This has
been explained in various ways: by the settlement of armies independent of Rouen,
by different degrees of pre-existing infrastructure (and different levels of damage
to it), and, for the west, by close connections with Brittany or the Irish-Sea zone.92
A disunited Church underlines this picture. Whatever the extent of the grant to
Rollo, it was far from coterminous with the ecclesiastical province, although the
arrangement seems to have included a partnership with Rouen’s archbishop. His
archdiocese was seriously compromised; for much of the tenth century only Rouen
and, perhaps, Evreux, had a continuously resident bishop.93 A few great abbeys were
re-established in the tenth century, but the full diocesan structure seems only to have
been reconstituted in the 990s.94 An episcopal vacancy at Avranches from the 860s
to the 990s suggests that Rouen either had insuficient inluence or lacked the political will to embark on restoration there. As for Coutances, according to its twelfthcentury history, Rollo arranged for its bishop to be transferred to Rouen, ‘and there
he worked as if he were in his own see’.95 This claim may be a later invention, but,
as François Neveux has pointed out, the diocese of Coutances was not re-established
in situ until 1049.96 Although Rollo and his successors did their best to appropriate
Coutances’s episcopal power, it was some time before they could place bishops
there and exploit their inluence directly. This is not to say that anywhere without
a bishop was ‘godless’; there is important evidence for rural churches.97 But it is
likely that different kinds of religious inluence and different degrees of centralized
ecclesiastical power would have coloured local identities.
If there were numerous Viking constituencies in the tenth century, it would have
been in the interests of the ducal family to erase them from memory. Dudo, of
course, presented a picture of a single (and singular) province forged by destiny:
The land of Normandy will never be subject to the protection of more than a single
lord.… For when Rollo was banished from the conines of Dacia he conquered
[Normandy] for himself in one piece, and it was not divided up by anyone thereafter.98
It has long been recognized that the reality was quite different. Although Dudo has
in many ways been restored to historical favour in recent years, no one disputes
91
Douglas, ‘The Rise’, 104.
Bates nevertheless emphasized Rouen’s connections in western Normandy in the eleventh century:
Normandy, 103. Éric van Torhoudt makes a case for (at least) ecclesiastical continuity in his doctoral
thesis, Centralité et marginalité en Neustrie et dans le duché de Normandie: maîtrise du territoire et
pouvoirs locaux dans l’Avranchin, le Bessin et le Cotentin (VIe–XIe siècles), Paris (Diderot), 2008, which
I have unfortunately not yet been able to consult; see also van Torhoudt, ‘La résistance’.
93 R. Allen, ‘The Norman Episcopate, 989–1110’, 2 vols, PhD thesis, Glasgow 2009, 6–28. I am grateful
to the author for sending me a copy of his thesis.
94 L. Musset, ‘Un millénaire oublié: la remise en place de la hiérarchie épiscopale en Normandie autour
de 990’, in Papauté, Monachisme, et Théories politiques: études d’histoire médiévale offertes à Marcel
Pacaut, ed. P. Guichard, M.-T. Lorcin, J.-M. Poisson, and M. Rubellin, 2 vols, Lyon 1994, II, 563–73.
95 Allen, ‘The Norman Episcopate’, 6; for a translation of this section of the text, see The Normans in
Europe, ed. and trans. E. van Houts, Manchester 2000, 38–40 at 39.
96 Neveux, ‘L’héritage des vikings’, 115.
97 I should like to thank David Petts for sending me a draft version of his ‘Churches and Lordship in
Western Normandy AD 800–1200’.
98 Tellus Northmanniae non nisi unius senioris unquam tuebitur advocatione, nec debet esse divisum
quod decet esse continuum. Rollo enim, Daciae inibus exterminatus, hanc sibi integre vindicavit, et non
dispertita ab ullo postea exstitit: Dudo, ed. Lair, 237 ; trans. Christiansen, 113.
92
Anglo-Norman Studies 35.indb 61
24/04/2013 12:50:53
62
Lesley Abrams
that contemporary concerns inluenced his depiction of the past.99 The land that
became Normandy was neither a political nor a monetary unit in 911,100 but the
actual accomplishment of unity is dificult to date. Historians cite different routes
to this achievement – internal colonization, the formation of a new aristocracy,
monastic reform, episcopal revival, written bureaucracy, saints’ cults – with necessarily different timelines. There is little to override Dudo’s duke-centred spin:
Flodoard’s principal focus was on matters of concern to the archdiocese of Reims.
In particular, evidence for the extension of ducal power goes hand in hand with the
evidence of documents, noticeably scarce until the eleventh century. Like Dudo,
who insisted that Norman society was regulated from the start by Rollo’s laws
(decreta, iura, leges, statuta), charters asserted that possession of land was guaranteed by the dukes’ written record.101
This seems, however, to obscure a substantial period of almost exclusively oral
process. As Gilduin Davy has pointed out, historians have given a possibly spurious
written status to a series of acts in the name of Rollo and William Longsword which
are referred to in later documents and often classiied as ‘lost’.102 In contrast, a
charter of Richard II for Saint-Ouen conirms grants made by Rollo, William, and
Richard I explicitly without charters (absque cartarum notamine concessit).103 This
may be more representative of how things worked in the tenth century. If the new
rulers did employ an inherited Carolingian administration to entrench their authority,
in the region dependent on Rouen or more widely across the province, they seem
to have done so without leaving written traces. Many aspects of government and
landholding continued to be oral elsewhere in Francia, of course, but the absence of
charters in tenth-century Normandy seems to mark a break with previous practice.
The disruption of monastic holdings may have temporarily removed one literate
cohort from the picture, but the surviving diocesan Church could have supplied
the dukes with scribes had they wanted them.104 Perhaps it was partly a question of
recipients: Scandinavians could have been happy to receive their new lands without
documentation, possession being validated through oral ritual, but when religious
houses entered the picture they required written charters to deine and guarantee
their rights of lordship. Alternatively, a time-lag in the written expression of regalian rights might be expected when counts, not kings, came to exercise them; in
Autun and Auxerre, however, counts issued charters modelled on royal diplomas in
the early tenth century.105 On the other hand, there is a curious comparison to hand
99
See, for example, L. Shopkow, ‘The Carolingian World of Dudo of Saint-Quentin’, JMH 15, 1989,
19–37; P. Bouet, ‘Dudon de Saint-Quentin et Virgile: L’«Enéide» au service de la cause normande’,
Recueil d’études en hommage à Lucien Musset, Cahier des annales de Normandie 23, 1990, 215–36;
F. Lifshitz, ‘Dudo’s Historical Narrative and the Norman Succession of 996’, JMH 20, 1994, 101–20;
Bauduin, La première Normandie, 61–83.
100 Bates, Normandy, xii, Bauduin, La première Normandie, 78–83; see Moesgaard, ‘A Survey’, 101,
for the Seine as a currency boundary.
101 Dudo, ed. Lair, 200, 205, and 245; trans. Christensen, 76, 81, and 119; M. Arnoux, ‘Disparition ou
conservation des sources et abandon de l’acte écrit: quelques observations sur les actes de Jumièges’,
Tabularia 1, 2001, 1–10; G. Davy, ‘Le scribe, le droit et le prince: recherche autour de l’utilisation de
l’écrit juridique par les ducs de Normandie de la in du Xe à la in du XIe siècle’, Tabularia 2, 2002, 1–18,
esp. 8–12.
102 Davy, ‘Le scribe’, 3.
103 RADN, no. 53 (c. 1025–6). Three of Richard I’s six surviving charters involve non-Norman parties.
104 Davy, ‘Le scribe’, 6, described eleventh-century secular clergy as ‘vectors of the translation of the
Carolingian cultural patrimony into Normandy’.
105 G. Koziol, ‘Political Culture’, in France in the Central Middle Ages, 900–1200, ed. M. Bull, Oxford
2002, 43–76 at 49.
Anglo-Norman Studies 35.indb 62
24/04/2013 12:50:53
Early Normandy
63
across the Channel: no charters survive in England from 909 to 924, when King
Edward the Elder was moving northwards and eastwards into the Danelaw and, we
might assume, rewarding his followers with land taken from Scandinavian holders
as well as conirming the possession of those who weathered the political change.106
Questions about Dudo’s authority extend to the issue of Normandy’s origins. We
may wonder whether his emphasis on the Danish connection, for example, derived
from Richard II’s friendship with Svein Forkbeard, king of Denmark c. 987–1014
(more bankable at the time of writing, perhaps, than an association with Norway’s
Olaf Haraldsson (king c. 1015–28, 1029–30)). Dudo’s work was known in Norway
at least by the later twelfth century, through William of Jumièges’s Gesta,107 but
west Scandinavian tradition differed from Dudo in giving Normandy’s founder a
Norwegian ancestry.108 This too may have derived from contemporary concerns – the
desire to link Orkney with the kings of England after the Conquest, for example;109
if it did preserve earlier traditions, they would have to have survived for a very
long time. The almost total absence of memory in Normandy of Olaf Haraldsson,
baptized in Rouen and widely culted in Scandinavian circles overseas,110 is striking.
Instead, Norman hagiography of the eleventh century concentrated on saints of the
distant past and, like Dudo, projected an uncompromising, ducally led, regional
unity back to the beginning.111
The ways in which the Norman Church repressed any impetus to express a Scandinavian cultural identity would repay further investigation. If Northumbrian sculpture of the tenth century portraying Thor and Sigurd had the blessing of the Church,
its apparent absence from Normandy would suggest that the latter’s religious establishment was less culturally accommodating. Obstacles to the continued expression
of Scandinavian culture may have come more from the Frankish aristocracy, on
the other hand. Without more evidence, it is dificult to judge whether indigenous
political structures, ecclesiastical culture, and social customs were just so strong
that they drove out most imported alternatives in the early stages of settlement or
whether this end-result can be explicitly attributed to the subsequent hegemony of
Rouen.
The state of the evidence in early Normandy has often been seen as a problem:
given the absence of other elements of Scandinavian culture, the evidence of
language and place-names has seemed to some to be anomalous or in conlict with
the argument that the Vikings of Rouen swiftly embraced Frankish politics and
identity. In my view, Normandy’s personal names and place-names are not problematic: they testify to the presence of Norse-speaking immigrants. What remains to be
106
D. N. Dumville, Wessex and England from Alfred to Edgar, Woodbridge 1992, 151–3.
Theodoric referred to it in his Historia de antiquitate regum Norwagiensium (c. 1177x1187): Theodoricus Monachus: An Account of the Ancient History of the Norwegian Kings, trans. D. and I. McDougall, London, 1998, 17.
108 See Historia Norwegie, ed. I. Ekrem and L. B. Mortensen, trans. P. Fisher, Copenhagen 2003, 66–9
(written 1160x1175) and Snorri Sturluson’s genealogy of the Norman dukes in his Heimskringla, written
in the 1220s or 1230s: Heimskringla. History of the Kings of Norway, trans. L. M. Hollander, Austin TX
1991, 79 and 259.
109 I owe this suggestion to Alex Woolf.
110 GND, II, 26–7; H. Antonsson, ‘The Cult of St. Ólafr in the Eleventh Century and Kievan Rus’,
Middelalderforum 1–2, 2003, 143–60; the only Norman evidence that I know of is a mention of Olaf’s
baptismal cloak in a fourteenth-century inventory: J. Dubois, ‘Le trésor des reliques du Mont-SaintMichel’, in Millénaire monastique du Mont Saint-Michel, 5 vols, Paris 1967–93, I, 501–93 at 531–2.
111 F. Lifshitz, The Norman Conquest of Pious Neustria: Historiographic Discourse and Saintly Relics,
684–1090, Toronto 1995, 137–219; S. K. Herrick, Imagining the Sacred Past: Hagiography and Power
in Early Normandy, Cambridge MA 2007.
107
Anglo-Norman Studies 35.indb 63
24/04/2013 12:50:53
64
Lesley Abrams
understood is why those who spoke Norse did not act more like their contemporaries
in the settlements of England, Scotland, Ireland, and Rus. One traditional solution
– the idea that they did not come directly from Scandinavia and were therefore less
committed to the culture of the homelands – can be rejected, given that Scandinavians in other overseas settlements developed and maintained strongly diasporic
identities. Other tentative ways of explaining why Normandy looks so different
from its Scandinavian cousins have been offered here. One – that the evidence of
material culture is lost or remains undetected – must be left open. Another – that the
Scandinavian experience in Normandy was not monolithic, and that we should avoid
transposing the evidence from Rouen across the province – builds on the observations of others. It may be helpful for future study to disaggregate Normandy and to
conceive of more regional variation in the early stages of Scandinavian settlement,
thanks to the legacy of pre-Viking differences, the uneven survival of infrastructure,
different degrees of independence from Rouen, and the variables of immigration –
settlers of different occupations and status arriving at different times, from different
starting-points. If the grant of 911, dependent on conversion and compliance, made
new lords into successors, backed by the Church, this was not necessarily how
everyone irst gained land or social, economic, or political power. If such heterogeneity did characterize the irst stages of settlement, its disappearance from memory
can be accounted for by the subsequent success of the ducal regime. Its monopoly of
the record suggests that evidence from the eleventh century is unlikely to be much
of an indicator for the tenth. Normandy retained its connections with the Scandinavian diaspora, and its Scandinavian roots were not forgotten; but cultural memory
was overlaid by a new myth, the gens Normannorum, and Scandinavian identity was
reformulated and repackaged in the service of the dukes.112
112
Some of the research for this paper was done during a period of leave generously funded by the
Leverhulme Trust. I would like to thank Tom Lambert, George Molyneaux, and David Parsons for
reading a draft and Pierre Bauduin for many instances of assistance.
Anglo-Norman Studies 35.indb 64
24/04/2013 12:50:53