EARLY ARISTOCRATIC SEALS:

AN ANGLO-NORMAN SUCCESS STORY*

Jean-François Nieus

In an influential overview of the development of sealing practices from Antiquity to the thirteenth
century, the French historian Robert-Henri Bautier stated that aristocratic seals appeared in England
with ‘considerable delay with regard to continental Europe’.1 Anglo-Norman counts and barons
would thus hardly have used seals before 1200, which, as Bautier wrote, placed them on the same
level as their counterparts from Poitou and other southern French regions, where seals were
notoriously very late to flourish. The same distinguished diplomatist and sigillographer even offered
an easy interpretation for the delayed rise of English baronial seals in the early thirteenth century,
observing that it coincided with the Magna Carta and the political victory of feudal magnates against
the authoritarian Plantagenet kings.2

Bautier was terribly wrong, as we shall see, but, in his defence, it must be said that British scholarship
was of little help in assessing the importance of aristocratic seals in twelfth-century England, since an
overview of the broadening use of seals in this country during the High Middle Ages is still lacking.
Thomas Heslop, in his survey dedicated to English seals up to 1100, essentially focused on late Anglo-
Saxon insignia, failed (quite comprehensibly) to bring to light the premises of Anglo-Norman lay
sigillography in the late eleventh century.3 Scattered pieces of evidence, such as the early seal of the
Norman knight Ilbert de Lacy (d. 1091/1100), which was briefly commented by Michael Clanchy in his
classic book From Memory to Written Record,4 or the similarly very ancient seal of Ralph de
Mortemer (fl. c. 1080–1104), mentioned by Richard Mortimer in his 2002 Battle Conference paper on
eleventh-century lay charters,5 have not really come to the attention of specialized historians (pl. 1-
2). Until now, no one has realized that these two artefacts are the oldest extant ‘baronial’ or
‘knightly’ seals, not only in England, but in the West.

* I would like to thank Richard Allen, Sir John Baker, Clément Blanc, Frédéric Boutoulle, Guilhem Dorandeu, Claude Groud-
Cordray, Samuel Mathot, Vivien Prigent, Nicholas Vincent and, especially, David Crouch for pointing me towards useful
evidence. I am also very grateful to Liesbeth Van Houts for her warm welcome at the Battle Conference.
1
Robert-Henri Bautier, ‘Le cheminement du sceau et de la bulle des origines mésopotamiennes au XIIIe siècle occidental’,
Revue française d’héraldique et de sigillographie 54–9, 1984–9, 41–84 (reprinted in idem, Chartes, sceaux et chancelleries.
Études de diplomatique et de sigillographie médiévales, Mémoires et documents de l’École des chartes 34, 1990, vol. 1,
123–66) at 65–6: ‘un retard très net par rapport au continent’.
2
Bautier, ‘Cheminement du sceau’, 69–70.
3
Thomas A. Heslop, ‘English Seals From the Mid-ninth Century to 1100’, Journal of the British Archaeological Association
133, 1980, 1–16.
4 rd
Michael T. Clanchy, From Memory to Written Record: England 1066–1307, 3 ed., Chichester 2012, 313 and pl. 1 in fine.
5
Richard Mortimer, ‘Anglo-Norman Lay Charters, 1066–c. 1100: A Diplomatic Approach’, ANS 25, 2002, 152–75 at 158–9
and 169.

1
 Plate 1: Seal of Ilbert de Lacy, used c. 1088 x 1100. Diameter: 50 mm.
(from T. Hudson Turner, ‘Original Documents’, The Archaeological Journal 4, 1847, 249-51)

Plate 2: Seal of Ralph de Mortemer, used c. 1100. Diameter: 55 mm.

Unsurprisingly, indeed, the absence of a comprehensive catalogue of early princely, comital, and
baronial seals in Latin Europe has prevented scholars from investigating the beginnings and early
spread of lay seal practice beneath the imperial and royal level on solid grounds. This is actually what
I have been working on for a while, as part of a broader research programme on the emergence of
‘signorial’ pragmatic literacy. My intention has been primarily to bring some order to the
documentary minefield of eleventh-century sealed princely charters (notoriously dotted with
forgeries!), with a particular focus on Flanders and Northern France.6 To be honest, I was not even
thinking of Normandy and England, because of Bautier obviously, and more generally, because in
scholarship, the introduction of aristocratic seals has usually been seen as a mere replication of the
rise of episcopal seals, which seem to have first prospered in Northern and Eastern France.7 The
ducal face of the famous double-sided seal of William the Conqueror may well have been duly
identified as the earliest known ‘princely’ equestrian seal, and even at times, as standing at the
beginning of the tradition;8 it nevertheless remained loosely challenged by other insignia (notably
those of the counts of Flanders and Anjou), and, in any case, the standard list of further known lay
seals up to the 1120s – a very short one indeed – did not point in the direction of the Anglo-Norman

6
By way of example, the following early princely seals enumerated by Bautier, ‘Cheminement du sceau’, 65–6, eventually
appear to be forged or, at the very least, spurious or wrongly dated: Flanders, Upper Lotharingia, Holland, Luxembourg,
Braine (sic, for Brienne?) and Clefmont.
7
Jean-Luc Chassel, ‘L’usage du sceau au XIIe siècle’, in Le XIIe siècle. Mutations et renouveau en France dans la première
moitié du XIIe siècle, ed. Françoise Gasparri, Paris 1994, 61–102 at 63–4; idem, ‘L’essor du sceau au XIe siècle’, Bibliothèque
de l’École des chartes 155, 1997, 221–34 at 222.
8
Exclusively by British scholars, notably Heslop, ‘English Seals’, 10, and Clanchy, From Memory, 313.

2
realm.9 Yet my investigations soon led to the conclusion that the key to the problem was to be
sought in this part of the medieval West.

In this paper, I intend to present briefly the most significant results of this investigation on early lay
seals and discuss their possible interpretations. The survey essentially relies upon a critical and, as far
as possible, exhaustive catalogue of all Western ‘first’ aristocratic seals up to the 1120s; here, ‘first
seal’ denotes the earliest known seal for each family or fief. Around eighty seals have been recorded
to date.10 I have been searching for not only extant wax impressions, but also evidence relating to
now lost seals: that is, drawings made by antiquaries, which are, of course, quite numerous;
descriptions of seals; sealing clauses in the text of charters; and also material traces of sealing on
original charters. Needless to say, I have encountered a whole range of critical and methodological
problems, which cannot be fully discussed here.11 I will, however, briefly point out two key concerns.
Firstly, the representativeness of my survey in terms of geographical coverage may be confidently
assured. While I have not seen all charters from all places, it was nevertheless possible to crosscheck
several transversal sets of documentation: notably, the large French and Belgian seal casts
collections,12 the recently launched database of French original charters before 1121,13 and the large
charter collections of some important monasteries whose estates or dependencies were spread
across very large areas (Savigny, Marmoutier, Cluny, Molesme, Fontevraud, Saint-Martin-des-
Champs, etc.).14 All of these wide-ranging seal and charter collections have confirmed a clustering of
evidence in and around Normandy. Secondly, and less comforting, there is the problem of the
authenticity of early sealed lay charters. Not only are a number of them forgeries, but a huge number
of late eleventh- and early twelfth-century unsealed documents – especially English documents –
were later reworked, precisely in order to provide them with the seal that they did not have. This is a
well-known situation,15 but I am afraid that the reality is even worse than commonly assumed. As it

9
See the list of twenty-five seals compiled by Pierre Bony, ‘L’image du pouvoir seigneurial dans les sceaux: codification des
signes de la puissance de la fin du XIe siècle au début du XIIIe siècle dans les pays d’oïl’, in Seigneurs et seigneuries au
Moyen Âge. Actes du 117e congrès national des sociétés savantes, Clermont-Ferrand, 1992. Section d’histoire médiévale et
nd
de philologie, 2 ed., Paris 1995, 367–401 at 387; idem, Un siècle de sceaux figurés (1135–1235). Le sceau image de la
personne en France d’Oïl, Angleterre, Écosse et pays de Lorraine, Paris 2002, 117. Nevertheless, Bony’s pioneering work on
the typology of seals is a most valuable reference.
10
I hope to be able to publish this catalogue in the not too remote future, with the afferent illustrations.
11
Methodological aspects are central in J.-F. Nieus, ‘Cataloguer les plus anciens sceaux aristocratiques (1066–1125): un défi
méthodologique’, in Le sceau dans les Pays-Bas méridionaux, Xe-XVIe siècles. Entre contrainte sociale et affirmation de soi,
ed. Marc Libert and J.-F. Nieus (forthcoming).
12
Paris, Archives nationales, Section des sceaux; Brussels, Archives générales du Royaume, Collection des moulages de
sceaux (now online: http://search.arch.be/fr/tips/99-zegelafgietsels, accessed 16/09/2015).
13
Chartes originales antérieures à 1121 conservées en France, ed. Cédric Giraud, Jean-Baptiste Renault and Benoît-Michel
Tock, online: http://www.cn-telma.fr/originaux (accessed 16/9/2015).
14
When evoking Marmoutier and Fontevraud, special reference must be made to the extraordinary collection of drawings
executed for Roger de Gaignières around 1600; for details, see Anne Ritz-Guilbert, ‘Les sceaux médiévaux au XVIIe siècle:
les dessins de sceaux dans la collection Gaignières (1642–1715)’, in Pourquoi les sceaux? La sigillographie, nouvel enjeu de
l’histoire de l’art. Actes du colloque organisé à Lille, Palais des Beaux-Arts, les 23–25 octobre 2008, ed. Jean-Luc Chassel and
Marc Gil, Villeneuve d’Ascq 2011, 45–60.
15
See, for example, Mortimer, ‘Anglo-Norman Lay Charters’, 152–75.

3
so happens, I have dismissed many more sealed charters than I included in my catalogue. It is very
likely that some reworked charters resisted my critical assessment, but I have worked with the belief
that the margin of error does not deeply affect the general picture…

My argument will run as follows. It will begin with a sketchy outline of the dissemination of
aristocratic seals, both in time and space, from their origins to the first quarter of the twelfth
century. In doing so, I will also try to call attention to the Anglo-Norman origins of aristocratic seals.
In the second part, I will tentatively address the question as to why seals flourished among the
Norman elite in the late eleventh century. And finally, the third part will be dedicated to some
aspects in the development of sealing practice beyond the Anglo-Norman realm, especially in
northern France, a region that also played an active role in the promotion of aristocratic seal culture.

Brief outline, 1066–1125

To start from the beginning, which is, actually, the oldest princely seal in the West? The answer may
be twofold. If we take the German Empire into account, the title holder is the Bavarian duke Henry
VII (1042–1047), who made use of a pedestrian seal inspired by the Ottonian royal model.16 But
Henry’s seal is very isolated; we only know of one similar insignia (namely, that of a count of
Ballenstedt in 1073, if genuine) and two now lost mid-eleventh-century ducal seals (namely, those of
the dukes of Tuscany and Lower Lotharingia).17 We therefore cannot speak of any consistent German
tradition. Later aristocratic seals in the Empire will be of the equestrian style introduced by
Lotharingian princes in the late eleventh century.18 Indeed, the proper question is, which is the oldest
extant equestrian seal? Since the purported mid-eleventh-century seals of the counts of Anjou and
Flanders have been rightly dismissed in recent scholarship,19 the very first equestrian seal on record
is that of William the Conqueror (pl. 3).20 The double-sided seal die adopted by the Conqueror after
Hastings is a celebrated, though not fully elucidated, artefact. Its ‘majesty’ side, displaying the new
king of the English seated on his throne, is just a modified version of Edward the Confessor’s seal.
The great novelty is the equestrian obverse, which depicts William as Duke of Normandy, or, to
quote the intriguing hexameter legend, as patronus Normannorum. We can see a man wearing a
16
Erich Kittel, Siegel, Bibliothek fur Kunst- und Antiquitätenfreunde 11, 1970, 121 (ill. 76) and 246.
17
Kittel, Siegel, 246 and 248 (ill. 153); Die Urkunden und Briefe der Markgräfin Mathilde von Tuszien, ed. Elke and Werner
Goez, MGH. Laienfürsten- und Dynastenurkunden der Kaiserzeit 2, 1998, 14–5; Georges Despy, ‘Les actes de ducs de Basse-
Lotharingie du XIe siècle’, in La Maison d’Ardenne, Xe-XIe siècles. Actes des Journées lotharingiennes, 24–26 octobre 1980,
Publications de la section historique de l’Institut grand-ducal de Luxembourg 95, 1981, 65–132 at 130.
18
See below.
19
Olivier Guillot, Le comte d’Anjou et son entourage au XIe siècle, vol. 2, Paris 1972, 11 n. 31; Georges Declercq, ‘Van
privaatoorkonde tot vorstelijke oorkonde. De oorkonden van de eerste graven van Vlaanderen, inzonderheid voor de Sint-
Pietersabdij te Gent (10de-11de eeuw)’, in Chancelleries princières et scriptoria dans les anciens Pays-Bas, Xe-XVe siècles,
ed. Thérèse de Hemptinne and Jean-Marie Duvosquel, Brussels 2010 (= Bulletin de la Commission royale d’histoire 176/2),
41–77 at 64–71; J.-F. Nieus, ‘Cum signo auctoritatis et excellentie mee sigillo. Sceaux et identité symbolique des comtes de
Flandre à la fin du XIe siècle’, Cahiers de civilisation médiévale 58, 2015, 43–64 at 44–5.
20
On this issue, see ultimately Regesta: William I, 102–5.

4
helmet or crown, mounted on a galloping horse, and holding a shield and banner with three
streamers. This image may seem banal, but it is not, for it was completely new in the 1060s. Indeed, I
do think that William was the inventor of the equestrian princely seal. He was rather the first
territorial prince (outside of the Empire) to break the royal monopoly on sealing, but he did so
precisely when he himself became a king.

Plate 3: Seal of William the Conqueror (equestrian obverse), used 1069. Diameter: 85 mm.
(Paris, Archives nationales, Moulages des sceaux des Archives nationales, no. 9998 ; photo J.-F. Nieus)

Plate 4: Seal of Hervey I de Stretton (reused by Hervey II), early 12th century. Diameter: 50 mm.
(Kew, National Archives, E 329/381; drawing J.-F. Nieus)

William’s extraordinary seal might of course be discussed at length, notably as regards the origins
and signification of its equestrian figure, but there is only place here for a couple of remarks. On the
dating, first: William’s seal is first met on a charter in 1067 or 1069,21 but it was probably made in
1066. One has speculated on the possible existence of the equestrian ducal obverse before the
Conquest,22 but, in my opinion, it is far more plausible that both sides were created simultaneously
when William became king, for the very reason that has just been given. Slight differences in style
between the two sides of the matrix could be interpreted as a division of work between two artists,
possibly aimed at shortening the delivery time. My next remark is more of a suggestion: might a
combined reading of the iconography and text of the ducal face lead to a renewed interpretation?
The easiest way to interpret the image is to see it as a representation of William as a duke, as a
military commander holding his most important sign of recognition (i.e., the banner).23 Yet the rider
apparently does not wear a full military outfit. He does not necessarily, or not only, refer to William

21
Regesta: William I, nos. 180 and 254.
22
Regesta: William I, 103.
23
See David Crouch, The Image of Aristocracy in Britain (1000-1300), London and New York 1992, 180–90, and more
recently, Robert W. Jones, Bloodied Banners: Martial Display on the Medieval Battlefield, Woodbridge 2010, esp. 33–56.

5
as the victorious leader on the battlefield of Hastings. The very unusual reference to the patronus
Normannorum, which is echoed in a couple of ducal charters for the abbey of Fécamp,24 as well as
parallels in twelfth-century iconography might well suggest that William’s insignia is also an allusion
to a saint, a triumphant military saint, like Saint George, whose cult and imagery were already
flourishing in Western France before 1100.25

The supposition that William paved the way for seal ownership among lay lords and created the
equestrian model is not only backed by the chronology of the earliest princely seals, but also, and
most importantly, by the subsequent dissemination pattern of aristocratic seals. Considering the
situation before 1090, as reflected on Map 1, things apparently started slowly. It appears, however,
that some of the most prominent neighbours and competitors of the Conqueror on the continent
were prompt to imitate him. The quickest move came from the count of Flanders: we know that in
1070, Arnulf III (1070–1071) owned a seal, which is now lost, and that his successor Robert the
Frisian (1071–1093) consistently used an equestrian seal (pl. 7), together with a secret seal modelled
on Carolingian royal seals.26 I have argued elsewhere that Robert so to say invented the equestrian
type ‘with a sword’ in order to challenge the type ‘with a spear’ promoted by his rival.27 Early seals
are also found among Robert’s neighbours and parents, namely his nephew Count Baldwin II of
Hainault (1071–1098) and his son-in-law Count Henry III of Louvain (1079–1095).28 Southwards,
Count Fulk IV ‘le Réchin’ of Anjou (1068–1109) also had an equestrian seal made for him before
1085.29 His ally William VIII, duke of Aquitaine (1058–1086), seems to have imitated him, although
the related charter evidence is thin.30 And lastly, within the Anglo-Norman realm, the Conqueror’s

24
Regesta: William I, 87.
25
I must refer here to the ongoing research of Samuel Mathot on the subject. I am grateful to him for discussing his ideas
with me.
26
Declercq, ‘Privaatoorkonde’, 64–71; Nieus, ‘Sceaux et identité’, 44–59.
27
Nieus, ‘Sceaux et identité’, 58–9. This point is also discussed below.
28
René Laurent, Les sceaux des princes territoriaux belges du Xe siècle à 1482, Brussels 1993, vol. 1/2, 356, no. 1, and vol. 2,
pl. 164 (Baldwin II); vol. 1/1, 256–7, no. 2, and vol. 2, pl. 87, no. 2 (Henry III). The authenticity of Henry’s seal has been
convincingly reaffirmed by Alain Dierkens and David Guilardian, ‘Actes princiers et naissance des principautés territoriales:
du duché de Basse-Lotharingie au duché de Brabant (XIe-XIIIe siècles)’, in Chancelleries princières et scriptoria dans les
anciens Pays-Bas, Xe-XVe siècles, ed. T. de Hemptinne and J.-M. Duvosquel, Brussels 2010 (= Bulletin de la Commission
royale d’histoire 176/2), 243–58 at 250 and 253–4.
29
Guillot, Comte d’Anjou, vol. 2, 11–2 and 216, no. C 347 b.
30
It only consists of one original charter (c. 1079) with traces of an applied seal and a sealing clause (‘scriptum […]
confirmatum autem sigillo nostro consignavimus’): Poitiers, AD Vienne, G 490, C 4, n° 69; printed in Louis Rédet,
‘Documents pour l’histoire de l’église de Saint-Hilaire de Poitiers’, Mémoires de la Société des antiquaires de l’Ouest 14,
1847, 1–362 at 97–9, no. 91 (see Chartes originales, no. 1251, with a photograph). This early occurrence of a ducal seal is
strangely isolated, since the next mentions date from 1096 and 1107: Cartulaire de l’abbaye Sainte-Croix de Bordeaux, ed.
Ariste Ducaunnès-Duval, Archives historiques de la Gironde 27, 1892, 4, no. 3 (sealing clause; I am grateful to Frédéric
Boutoulle for this reference); François Eygun, Sigillographie du Poitou jusqu'en 1515. Étude d’histoire provinciale sur les
institutions, les arts et la civilisation d’après les sceaux, Poitiers 1938, pl. I, no. 1 (seal impression).

6
own brother, Bishop Odo, created Earl of Kent (1067), used a double-sided seal evidently inspired by
William’s matrix, known to us by a drawing in Hatton’s Book of Seals.31

Map 1: The dissemination of aristocratic seals up to 1125.

A decade later, in 1100, both counts of the Blois-Champagne conglomerate, Hugh of Troyes (1093–
1125) and Stephen-Henry of Blois (1077–1102), are the only newcomers of ‘princely’ rank in the club
of seal owners,32 but by now, several other counts and lords have joined: there is evidence for the

31
Sir Christopher Hatton’s Book of Seals, ed. Lewis C. Stenton and Doris M. Stenton, Oxford 1950, 301, no. 431, and pl. 8
between pp. 304 and 305. Further reproductions: English Romanesque Art, 1066-1200: Hayward Gallery, London 5 April-8
July 1984, London 1984, 79; Paul D.A. Harvey and Andrew McGuinness, A Guide to British Medieval Seals, Toronto 1996, 23.
32
Arnaud Baudin, Emblématique et pouvoir en Champagne. Les sceaux des comtes de Champagne et de leur entourage (fin
XIe-début XIVe siècle), Langres 2012, 60–6.

7
counts of Boulogne, Ponthieu, Dammartin, Maine, and, more eastwards, Auxerre-Nevers and
Tonnerre.33 The Viscount of Cotentin Nigel III (d. 1092) is apparently the earliest noble of lesser status
known to have owned a seal, if we can trust the reference to his lost sigillum in a pancarte dated c.
1090.34 Further barons are Ilbert de Lacy, Gilbert Tison (d. c. 1124), Ralph de Mortemer,35 and,
possibly, the Picard lords Anselm II of Ribemont (1071–1099) and Odo of Péronne (fl. c. 1090–
1095).36 With the exception of the counts of the Auxerre-Nevers-Tonnerre region, who probably
imitated the counts of Champagne, all of these people were somehow involved with the Anglo-
Norman world. Men like Hugh of Dammartin (c. 1067–c. 1095) and Anselm of Ribemont might seem
more peripheral in the picture, but this is absolutely not the case, for Hugh married Rohese de Clare-
Tonbridge,37 while Anselm married his daughters to Walter II Giffard and one of his tenants.38 It is
eventually worth noting that the 1090s also witnessed the adoption of an equestrian seal by a king
from the immediate vicinity of the realm, i.e., Duncan II, King of Scotland (1094), who simply copied
the seal of William the Conqueror or William Rufus.39

The further dissemination of lay seals during the first quarter of the twelfth century simply confirms
the trends observed before 1100. A new set of princes, located at the periphery of the previous
dispersion area, adopted equestrian seals in the early years of the twelfth century (namely, the rulers
of Brittany, Aquitaine, Burgundy, and Upper Lotharingia, not to mention Louis, the future king of
France).40 Yet, for the rest, there is no significant geographical spread beyond the regions where lay

33
Boulogne: Mina Martens, ‘Une reproduction manuscrite inédite du sceau de Godefroid de Bouillon’, Annales de la Société
archéologique de Bruxelles 46, 1942–3, 7–26, pl. 1–3 between pp. 16 and 17 (not identified by Martens, but matches with a
damaged impression in Canterbury Cathedral Archive, DCc/ChAnt/F/130). – Ponthieu: Recueil des actes des comtes de
Pontieu (1026–1279), ed. Clovis Brunel, Paris 1930, pl. 1 between pp. LXXVI and LXXVII, no. 1. – Dammartin: Bony, Sceaux
figurés, pl. 5, no. 20 (the correct date being 1082 x 95). – Maine (lost seal): Cartulaire des abbayes de Saint-Pierre de La
Couture et de Saint-Pierre de Solesmes, ed. [Benedictine monks of Solesmes], Le Mans 1881, 35–9, nos. 24–6 and 28. –
Auxerre-Nevers: Recueil des chartes de l’abbaye de Cluny, ed. Auguste Bernard and Alexandre Bruel, vol. 5, Paris 1894, 63–
4, no. 3717. – Tonnerre (lost seal): Dijon, AD Côte-d’Or, 7 H 1651; printed in Cartulaires de l’abbaye de Molesme, ancien
diocèse de Langres (946–1250). Recueil de documents sur le nord de la Bourgogne et le midi de la Champagne, ed. Jacques
Laurent, vol. 2, Paris 1911, 40–3, no. 28.
34
Léopold Delisle, Histoire du château et des sires de Saint-Sauveur-le-Vicomte, Paris and Caen 1867, Pièces justificatives,
50–5, no. 45. A seal belonging to Nigel’s brother is similarly mentioned in a document dated 1104: ibid., 55–8, no. 46.
35
See Mortimer, ‘Anglo-Norman Lay Charters’, 157–9 and 169.
36
We only have mentions of their seals in later copies: Paris, BN, Collection Moreau, vol. 37, fols 145r–146v (Anselm of
Ribemont, 1094; copy by dom Grenier, with the indication: ‘Sceau et attache perdus’); Cartulaire de l’abbaye de Saint-
Corneille, ed. Morel, vol. 1, 44–6, no. 18 (Odo of Péronne, 1091; sealing clause). Odo’s sister and heir, Adela, also owned a
seal by 1100: see below, n. 115.
37
Jean-Noël Mathieu, ‘Recherches sur les premiers comtes de Dammartin’, Mémoires de la Société de l’histoire de Paris et
Île-de-France 47, 1996, 7–59 at 48–9; Michel Bur, ‘De quelques champenois dans l’entourage français des rois d’Angleterre
aux XIe et XIIe siècles’, in Family Trees and the Roots of Politics: The Prosopography of Britain and France from the Tenth to
the Twelfth Century, ed. Katharine S.B. Keats-Rohan, Woodbridge 1997, 333–48 at 339.
38
K.S.B. Keats-Rohan, Domesday Descendants: A Prosopography of Persons Occurring in English Documents, 1066–1166. II:
Pipe Rolls to Cartae Baronum, Woodbridge 2002, 662 (with reference to Orderic, VI, 38).
39
Harvey and McGuinness, Guide to British Medieval Seals, 5–6.
40
Brittany: Actes des ducs de Bretagne (944–1148), ed. Hubert Guillotel et alii, Rennes 2014, 120–1. – Aquitaine: Eygun,
Sigillographie du Poitou, pl. I, no. 1 (but see above, n. 30). – Burgundy: Chassel, ‘Essor du sceau’, 226. – Upper Lotharingia:
Hubert Collin, Sceaux de l’histoire lorraine, Nancy, 1988 (= Lotharingia. Archives lorraines d’archéologie, d’art et d’histoire
1), 78, no. 33. – Prince Louis (VI): Recueil des actes de Louis VI, roi de France (1108–1137), ed. Jean Dufour, vol. 1, Paris
1992, pl. I, no. 1.

8
aristocrats had already begun to use seals in the late eleventh century. What we can essentially see is
an intensification of the practice within these regions. The only exception to this observation, which
is not seen on Map 1, lies in the fact that during the decades 1100 and 1110, some dukes and
margraves from the German Empire – but dukes and margraves only – also started using equestrian
seals, modelled on the seal of the dukes of Lower Lotharingia.41 At this stage in the development of
aristocratic seals, it is clear that their ‘homeland’ was the Anglo-Norman realm as well as the
adjacent regions with a very effective Norman influence: Brittany, Anjou, Chartrain, Picardy, and
Flanders. The vast majority of seal owners up to 1120 or 1130 were cross-Channel magnates and
barons, and even those who cannot be strictly defined as ‘Anglo-Norman’ or ‘Anglo-French’ still
enjoyed strong connections with Normandy and England. For instance, the three Flemish barons in
the list are William of Ypres (d. 1164), whose affinities with England certainly predated his well-
known involvement with King Stephen,42 and Wenemar I (1074/88–1118/20) and Baldwin III (1097–
1127) of Ghent, respectively castellan of Ghent and lord of Aalst, and both members of the family of
the advocates of St Peter’s abbey in Ghent from which originated the Domesday magnate Gilbert de
Gand (d. 1095).43 Another example is Aimery I (c. 1100–1151), Viscount of Châtellerault in Poitou, a
grandchild of another hero of the Conquest, Aimery de Thouars (d. c. 1094), himself somehow
related to Brittany and England.44

I have not pushed this systematic investigation further in time, but it seems clear that by the mid-
twelfth century, most Anglo-Norman aristocrats, and even the middle- and lower-rank nobles among
them – the ‘petty knights’ vilified by Richard de Lucy (d. 1179) in his famous statement on seal
ownership45 – did own a personal seal. An ample series of these more humble seals are, for instance,
described and precisely dated in Frank Stenton’s collection of charters relative to Danelaw.46 This is in

41
Kittel, Siegel, 521–2; Andrea Stieldorf, Marken und Markgrafen. Studien zur Grenzsicherun, durch die fränkisch-deutschen
Herrscher, MGH Schriften 64, 2012, 314–28. For the Lotharingian model, see below.
42
Sealing clause in 1119: Bruges, Groot Seminarie, Archief Lo, no. 34 (printed in Cartulaire de l’abbaye Saint-Pierre de Loo,
de l’ordre de Saint-Augustin, 1093–1794, ed. Léopold Van Hollebeke, Brussels 1870, 6–7, no. 4). Impressions and drawing of
a (later?) seal in use during the 1150s: Canterbury Cathedral Archive, DCc/ChAnt/C/1127 (I am grateful to Nicholas Vincent
for this reference); Bruges, Groot Seminarie, Archief Ten Duinen-Ter Doest, no. 448 (see http://search.arch.be/fr/tips/99-
zegelafgietsels, no. 21708, accessed 16/9/2015); Saint-Omer, Bibliothèque de l’Agglomération, MS 803/1, p. 288. On
William, see Eljas Oksanen, Flanders and the Anglo-Norman world, 1066–1216, Cambridge 2012, 225–31, and J.-F. Nieus,
‘The Early Career of William of Ypres in England: A New Charter of King Stephen’, in EHR 130, 2015, 527–45.
43
Both seals are known through engravings in André Duchesne, Histoire généalogique des maisons de Guînes, d’Ardres, de
Gand et de Coucy et de quelques autres familles illustres, qui y ont esté alliées, Paris 1631, Preuves, 67 (Wenemar I, from a
charter dated c. 1101) and 193 (Baldwin III, in 1125). On Gilbert de Gand and his Flemish relatives, see Oksanen, Flanders,
192–3, 198 and 204.
44
Bony, Sceaux figurés, 25 and pl. 5, no. 21. See François Chamard, ‘Chronologie historique des vicomtes de Châtellerault
avant la fin du XIIIe siècle’, Mémoires de la Société des antiquaires de l’Ouest 35, 1870–1, 79–122 at 103–8; Jacques Duguet,
th
‘Notes sur quelques vicomtes de Châtellerault’, Bulletin de la Société des Antiquaires de l’Ouest, 4 ser. 16, 1981, 261–71.
45
The Chronicle of Battle Abbey, ed. Eleanor Searle, Oxford 1980, 214–5. See, for example, Clanchy, From Memory, 53.
46
Frank M. Stenton, Documents Illustrative of the Social and Economic History of the Danelaw, from Various Collections,
London 1920, passim.

9
sharp contrast to the situation in the rest of northern Europe at the time, where most barons were
still on their way to acquiring their very first seal.47

Alongside this chrono-geographical framing of the aristocratic sealing practice in North-West Europe
during its first half-century of existence, complementary evidence supporting a ‘Norman’ origin of
the use of seals by lay people can also be found in the remote Mediterranean countries where the
Normans had extended their domination. The relevant sources from southern Italy and the Latin East
are poorly preserved and not well studied, but they still remain evocative. In Sicily and southern Italy,
the Norman settlers were massively faced with the Byzantine habit of sealing documents with leaden
bullae. As a result, from the 1070s onwards, local barons began using their own leaden bullae copied
from Byzantine models, as well as small dies designed for wax impressions, generally made from
antique or medieval intaglios.48 However, Count Robert II of Loritello (1107–1134/37), a descendant
of Tancred de Hauteville, made use of a small equestrian seal in 1113,49 as did Count Henry del Vasto
(c. 1090–c. 1137), a follower and relative of Roger I of Sicily, in 1115.50 Count Roger (1072–1101)
himself owned a seal with an intaglio, but, around 1100, he also minted remarkable bronze coins
displaying a mounted warrior with a banner, obviously derived from contemporary North-European
seals.51 Sigillographic material from the Latin Kingdom of Jerusalem, whose early elites included a fair
number of Anglo-Norman and Italo-Norman crusaders, also provides some significant clues. From the
1110s and 1120s onwards, several counts and barons made use of similar equestrian bullae that
presented a strong resemblance to some of the oldest Western seals.52 Even more revealing are two
recently published bullae, dated around 1100 and respectively attributed to one ‘Richard the
Constable’ (probably a Norman senior official in the newly created Principality of Antioch) and one
‘Count Richard’ (most probably Richard of Salerno, regent of Edessa in 1104–1108);53 their

47
Chassel, ‘Usage du sceau’, 66–8; Hubert Flammarion, ‘Le sceau du silence: sigillographie et pratiques seigneuriales au XIIe
siècle entre Marne et Meuse’, in Retour aux sources. Textes, études et documents offerts à Michel Parisse, ed. Sylvain
Gougenheim, Paris 2004, 99–113; J.-F. Nieus, ‘Des seigneurs sans chancellerie? Pratiques de l’écrit documentaire chez les
comtes et les barons du nord de la France aux XIIe-XIIIe siècles’, in Chancelleries princières et scriptoria dans les anciens
Pays-Bas, Xe-XVe siècles, ed. T. de Hemptinne and J.-M. Duvosquel, Brussels 2010 (= Bulletin de la Commission royale
d’histoire 176/2), 285–311 at 289–91.
48
Arthur Engel, Recherches sur la numismatique et la sigillographie des Normands de Sicile et d’Italie, Paris 1882; Vivien
Prigent, ‘Notes sur la tradition sigillographique byzantine dans le royaume normand de Sicile’, in L’héritage byzantin en
Italie (VIIIe-XIIe siècle). II: Les cadres juridiques et sociaux et les institutions publiques, ed. Jean-Marie Martin, Annick Peters-
Custot and V. Prigent, Collection de l’École française de Rome 461, 2012, 605–41.
49
Engel, Recherches, 96, no. 44.
50
See Pietro Burgarella, Nozioni di diplomatica siciliana, Palermo 1978, 170. Slightly later equestrian seals are those of
Matthew of Craon (ibid., 135 and pl. 15, no. 3) and Robert II of Capua (hunting type, recently discovered by Guilhem
Dorandeu; I am extremely grateful to him for sharing his results with me).
51
Philip Grierson and Lucia Travaini, Medieval European Coinage, 14: Italy (III) (South Italy, Sicily, Sardinia), Cambridge 1998,
89–91 and 608–11. Roger’s coinage was imitated by William, Duke of Apulia (1111–1127), in 1114 (see ibid., 98). Grierson
and Travaini state that ‘[t]he equestrian figure has no formal model that one can identify’ and invoke unconvincing
sculptural parallels – but also, more interestingly, the design of the Bayeux Tapestry (ibid., 90).
52
See Hans E. Mayer, Das Siegelwesen in den Kreuzfahrerstaaten, Bayerische Akademie der Wissenschaften. Philosophisch-
historische Klasse. Abhandlungen N.F. 83, 1978, and, for further illustrations, François Chandon de Briailles, ‘Bulles de
l’Orient latin’, Syria 27, 1950, 284–300.
53
André Ronde, ‘Un sceau en plomb de l’Orient latin au nom du comte Richard’, Bulletin de la Société française de
numismatique 58, 2003, 203–4; John W. Nesbitt, ‘A Seal of Richard the Constable’, Thesaurismata 37, 2007, 61–4. On the

10
equestrian image strikingly echoes the seals of two major leaders of the Crusade, namely Duke
Godfrey of Bouillon (1087–1100) and his brother Count Eustace III of Boulogne (c. 1089–1125).54

Why the Normans?

Having (hopefully) made the point about the Anglo-Norman origin of aristocratic seals, the next step
is, of course, to address the question as to why seals flourished among the Norman lay elite around
1100. There is certainly no unique and ascertained answer to this complex interrogation. At this
stage of the investigation, I am inclined to formulate three hypotheses, which may be seen as
complementary.

The first has to do with cultural transfers. We have good reasons to consider that the Anglo-Saxon
aristocracy already possessed seals and utilized them on a rather large scale, at least during the
century before the Norman Conquest.55 Concrete evidence is provided by a set of Anglo-Saxon seal
dies, of which three have been on record since the nineteenth century,56 while a fourth was retrieved
by metal detectorists as recently as 2010.57 They are all dated between the late tenth and mid-
eleventh century. If one considers just how rare later equestrian seals matrices are, whether
preserved in specific collections or resulting from random finds, these four Anglo-Saxon artefacts
might well be statistically significant evidence for the widespread use of seals among the pre-
Conquest ruling class. Moreover, all four dies exhibit the same stereotyped image of a nobleman
holding his sword. They refer to a common model that is still to be seen on the famous early twelfth-
century seal of Thor Longus (fl. c. 1113–1124), an Anglo-Saxon noble settled in Scotland.58 The
prevalence of this model is in itself further strong evidence for the existence of ordinary sealing
practices in late Anglo-Saxon times, probably in relation to various types of documents of which
nothing has been preserved. I am not saying that the Normans were confronted with such a
pervasive reality that they were immediately and massively urged to imitate it (indeed, things did not
occur that way), but the use of seals in Anglo-Saxon society may have been conspicuous enough so as
to encourage the Normans to gradually incorporate seals in their own literate practices. This is the

possible identity of Richard the Constable, see Thomas S. Asbridge, The Creation of the Principality of Antioch, 1098–1130,
Woodbridge and Rochester 2000, 184.
54
For Godfrey and Eustace’s seals, see above, n. 33, and below, n. 103.
55
Recent discussions include Jane Roberts, ‘What did Anglo-Saxon Seals Seal When?’, in The Power of Words: Essays in
Lexicography, Lexicology and Semantics in Honour of Christian J. Kay, ed. Graham D. Caie, Carole Hough and Irene
Wotherspoon, Amsterdam 2006 (= Costerus N.S. 163), 131–57; Jane K. Kershaw and Rory Naismith, ‘A New Late Anglo-
Saxon Seal Matrix’, ASE 42, 2013, 291–8; Simon Keynes, ‘Seals’, in The Wiley Blackwell Encyclopedia of Anglo-Saxon
nd
England, ed. Michael Lapidge, John Blair, S. Keynes and Donald Scragg, 2 ed., Chichester 2014, 426-8.
56
Seal matrices of Aelfric, Godwin, and Wulfric, described by Heslop, ‘English Seals’, 4–7.
57
Kershaw and Naismith, ‘New Late Anglo-Saxon Seal Matrix’, 291–8.
58
Heslop, ‘English Seals’, 15–6 (ill. 2), and ‘Seals’, 317, no. 370.

11
same kind of cultural transfer as seen in Norman Italy or the Latin East, where the Byzantine use of
bullae was somehow recycled by the new Western ruling class.

Another hypothesis relates to the practical efficacy of seals as a means to close letters and validate
charters. Indeed, the many Normans enfeoffed with lands in England after 1066 were soon
confronted with the difficulties of managing estates disseminated on both sides of the Channel, and,
in the case of their new English lands, often deliberately scattered across several distant counties.
Communication consequently became a nightmare for them. Transmitting orders in a secure and
efficient way over long distances turned into a critical issue. The use of the written word was
certainly part of the solution.59 The use of seals probably was, too. The seal affixed to a written
message made its authentication and acceptation by the recipient possible; it brought solid proof
that the document was unaltered and that it did indeed originate from its issuing person or
authority, whose name and image were to be seen on the wax impression. Furthermore, if we
consider that many potential recipients of such written ‘mandates’ produced by Anglo-Norman
landlords were likely to be intendants of Anglo-Saxon origin, possibly familiar with aristocratic seals,
there might even have been an urge to use seals in order to secure the enforcement of the lord’s
command. We do not know, of course, how widespread administrative letters and written orders
were in Anglo-Norman society around 1100, but we can reasonably infer that they once existed in
large quantities, even though nearly all of them have now disappeared.60 Nor do we know if they
were necessarily sealed. The only thing we do know for sure is that royal writs, the well-known
Anglo-Saxon form of documents adopted by the Norman administration, needed to be validated with
the king’s pendant seal.61 I would suggest that early baronial writs were likewise commonly sealed, as
were those issued by Bishop Odo as the Earl of Kent (the only ones preserved from the late eleventh
century),62 as well as those produced in the course of the twelfth century. Evidently, the general
evolution of diplomatic practice, which began to demand a single deed for each conveyance or
transaction after 1100, also gave a strong impulse to the use of seals – whether aristocratic or not –
on more solemn charters. Some great monasteries, like Marmoutier and Savigny, might even have
actively prompted certain laymen in acquiring a seal.63

However, in Normandy and England, it took decades before seals really supplanted other means of
personal corroboration, such as autograph crosses, symbolic objects, or the simple touching of the

59
See most recently D. Crouch, ‘Between Three Realms: The Acts of Waleran II, Count of Meulan and Worcester’, in Record,
Administration and Aristocratic Society in the Anglo-Norman Realm, ed. Nicholas Vincent, Woodbridge 2009, 75–90.
60
In fact, as pointed out by Judith Everard, ‘Lay Charters and the Acta of Henry II’, ANS 30, 2008, 100–16 at 103–4, this
important issue has not received much scholarly attention. The most relevant discussions to date are by Mortimer, ‘Anglo-
Norman Lay Charters’, 153–75, and Crouch, ‘Between Three Realms’, 75–90, who both tend to admit that lay aristocrats
already produced substantial numbers of charters, writs, and letters by the late eleventh and early twelfth centuries. For a
different opinion, see Clanchy, From Memory, 56–8.
61
Richard Sharpe, ‘The Use of Writs in the Eleventh Century’, ASE 32, 2003, 247–92 at 249; Mark Hagger, ‘The Earliest
Norman Writs Revisited’, Historical Research 82, 2009, 181–205 at 184.
62
Mortimer, ‘Anglo-Norman Lay Charters’, 160 n. 29.
63
This can only be a guess based on the observation that their archives provide us with several early lay seals.

12
charter.64 There were certainly other reasons for the cross-Channel elite to turn to seals. Seal
matrices were not just mere validation tools. They were meaningful markers of identity as well as
powerful symbols of status and lordship.65 This statement applies to all medieval seals, but I do think
that it is especially true in relation to the early aristocratic seals, which must have enjoyed a strong
sense of novelty and exclusiveness among their Norman and other ‘French’ owners, and probably
still somehow referred to the royal origins of the sealing practice. Moreover, the materiality of seal
dies and their visual content certainly had a particular impact on the illiterate or semi-literate nobles
for whom they were designed: beyond their basic function in authenticating charters and letters,
they could easily be regarded as an incarnation of their owner’s ability to exert power – and even, to
a certain extent, his personal virtues. At the very least, there are strong indications that the
possessors of the early seals whom we are talking about cared very much about them. Charter texts,
when announcing their seals, vibrantly proclaim that these are the mark of their virtus, potestas,
excellentia, or even majestas.66 After the sigillants’ death, their matrices were reverently kept in
family treasures and frequently reused (or accurately replicated) by the one or several of their
successors, in a manner so as to express the continuity of lordship or a special link to the deceased
forebear.67 The strong ‘symbolic’ dimension of lay sigillography therefore invites us to consider not
only the pragmatic reasons for the booming success of seals in the Anglo-Norman world and its
sphere of influence, but also the ‘ideological’ motives that might have resulted from an intricate
array of political, social, and familial factors.

We need to focus again on the dissemination process of lay seals up to the 1120s, but this time
through the lens of social networks and iconographical cross-references. The dissemination process
as we have seen was indeed fuelled by imitation, whether at a horizontal level – that is, between
competing kings, magnates, and lords (or, conversely, allied princes and lords) – or at a vertical level
– that is, between vassal and lord, as an expression of allegiance or re-joining a political community.
At the same time as the use of seals gradually spread throughout the Anglo-Norman realm and its
margins, it concomitantly went down the social ladder in Anglo-Norman society, from the king-duke
to lesser knights. This top-down mimetic process is a well-known feature of socio-political history,
but it requires careful assessment in order to avoid misinterpretation.

The Conqueror’s double-sided seal was tremendously influential. Both of its sides were reproduced
by William Rufus and Henry I when they in turn became king of England, even though they were not

64
See, for example, Clanchy, From Memory, 38–41, 256–62, and 309–18.
65
Brigitte M. Bedos-Rezak, When Ego Was Imago: Signs of Identity in the Middle Ages, Leiden and Boston 2011.
66
Crouch, Image of Aristocracy, 243; Nieus, ‘Sceaux et identité’, 64; Cartulaire de l’abbaye de Saint-Corneille de Compiègne,
ed. Émile-Épiphanius Morel, vol. 1, Montdidier 1904, 75–6, no. 3 (charter of Rainald II of Clermont, count of Vermandois,
1115: ‘fecit litteris annotari et sue majestatis insigniri sigillo’).
67
J.-F. Nieus, ‘L’hérédité des matrices de sceaux princiers au XIIe siècle, entre conscience lignagère et discours politique’, in
Pourquoi les sceaux? La sigillographie, nouvel enjeu de l’histoire de l’art. Actes du colloque organisé à Lille, Palais des Beaux-
Arts, les 23–25 octobre 2008, ed. Jean-Luc Chassel and Marc Gil, Villeneuve d’Ascq 2011, 217–39. Since the publication of
this article, I have been able to identify many more reused matrices, especially within the Anglo-Norman realm.

13
(yet) duke of Normandy, which obviously helped to prolong the influence of the original equestrian
prototype until the early twelfth century.68 Its impact can be measured both within and outside the
Norman Empire. Beyond Norman frontiers, it was strikingly taken over without any modification by
Duncan II during his brief reign as king of Scotland in 1094,69 while on the Continent, although princes
often opted for a somehow modified representation, several of them – such as Duke Odo I of
Burgundy (1079–1103) and Count Ralph I of Vermandois (c. 1115–1152)70 – just abided by the image
of the king-duke with horse and banner. However, such literal imitations mostly took place within the
Anglo-Norman political realm. Several counts indeed appropriated King William’s archetype at the
turn of the eleventh and twelfth centuries: we have solid evidence for such borrowings by the counts
of Surrey, Huntingdon, Aumale, and, on the Norman borders, Ponthieu and Perche;71 the well-
preserved seal impressions of Gui of Ponthieu (thanks to the reuse of the latter’s matrix by his
successors)72 and David of Huntingdon both show a very careful copy of their model. One generation
later, in the 1110s and 1120s, we can even see lesser barons adopting an equestrian seal matrix
charged with exactly the same depiction, including the banner. The only well-dated contemporary
example is the seal of Peter fitz William of Studley, attached to a charter dated c. 1125,73 but some
later documented seals were most probably cut during the same period, such as those of Hervey I of
Stretton (fl. 1086) (pl. 4), Hugh IV of Gournay (c. 1116–1178/9), and the Picard Lord Gerard I of
Picquigny (1122/6–1178).74 Besides, slightly modified designs also appear on a range of other comital
and baronial seals – reverted images, for instance, with the horse riding to the left, as can be seen on
the seals of Eustace III of Boulogne, Robert de Beaumont, count of Meulan and earl of Leicester
(1081–1118), and Rotrou II, count of Perche (1099–1144), or barons such as Eustace de Breteuil

68
Frank Barlow, William Rufus, Berkeley and Los Angeles 1983, ill. 1b between pp. 74 and 75; Facsimiles of English Royal
Writs to A.D. 1100 Presented to Vivian Hunter Galbraith, ed. Terence A. M. Bishop and Pierre Chaplais, Oxford 1957, XIX.
There is unfortunately no decipherable impression left of Duke Robert Curthose’s seal: see Charles H. Haskins, Norman
Institutions, New York, 1918, 72–3. It is worth noting that in the 1140s, Geoffrey Plantagenet, invested with Normandy, still
reproduced Henry I’s equestrian types on his new double-sided seal: Bony, Sceaux figurés, 27 and pl. X-XI, nos. 53 and 58.
69
Harvey and McGuinness, Guide to British Medieval Seals, 5–6. On the subsequent development of seal usage among the
Scottish nobility, see Cynthia J. Neville, Land, Law and People in Medieval Scotland, Edinburgh 2010, 81–99.
70
Burgundy: Chassel, ‘Essor du sceau’, 226; Bony, ‘Image du pouvoir’, 395, pl. I, no. 1. – Vermandois: see below, n. 126.
71
Surrey: Gustave Demay, Inventaire des sceaux de la Normandie, Paris 1881, 7, no. 40; Early Yorkshire Charters, vol. 8, ed.
William Farrer and Charles T. Clay, Leeds 1949, 66–73, nos. 7–15 (formerly sealed charters of William de Warenne, 1088 x
1118). – Huntingdon: Harvey and McGuinness, Guide of British Medieval Seals, 43 (and 44, ill. 37). – Aumale: Bony, Sceaux
figurés, pl. IX, no. 48. – Ponthieu: see above, n. 33. – Perche: Paris, BN, lat. 5441/2, 300.
72
See Nieus, ‘Hérédité des matrices’, 233–4 and 238 (ill. 5).
73
Sir Christopher Hatton’s Book of Seals, ed. Stenton and Stenton, 366–7, no. 528, and pl. IX between pp. 360 and 361.
74
Stretton: there is no known representation of Hervey I’s insignia, but we can safely infer that the extremely archaic seal
of Hervey II (c. 1157–1184/93), documented in 1161 x 5, is in fact his grandfather’s seal: Roger H. Ellis, Catalogue of Seals in
the Public Record Office: Personal Seals, vol. 1, London 1979, 63, no. P 763, and pl. 21; for the date, see ‘The Staffordshire
st
Chartulary, Series I-II’, ed. George Wrottesley, in Collections for a History of Staffordshire, 1 series, vol. 2, Lichfield 1881,
178–276 at 252–4. – Gournay: Gustave Demay, Inventaire des sceaux de la Flandre, vol. 1, Paris 1873, 122, no. 969 (see also
a drawing of the same seal, reused by Hugh V, in N.-R.-P. Potin de La Mairie, Recherches historiques sur la ville de Gournay,
Gournay 1844, between pp. 106 and 107). – Picquigny: Nieus, ‘Hérédité des matrices’, 238, ill. 7.

14
(1103–1136), Philip de Briouze (c. 1095–c. 1135), William de Saint-Clair (fl. 1119–1161), and William
d’Espinay (c. 1093–c. 1140).75

The extent of royal influence on Anglo-Norman (and ‘Anglo-French’) aristocratic seals can also be
verified otherwise. Firstly, the concept of the double-sided seal, aimed at illustrating the fact that its
owner enjoyed two holdings with distinct titles, was soon embraced by some great magnates: Bishop
Odo came first, followed by Robert de Beaumont and Rotrou II (count of Perche and, after 1113, lord
of Bellême), and later, in the 1120s and 1130s, by Stephen of Brittany (1093–1135/6; count of
Penthièvre and lord of Richmond), Waleran of Meulan (1118–1166; count of Meulan and, in 1138,
earl of Worcester), and Gilbert of Clare (d. 1148, created earl of Pembroke in 1138).76 Even the king
of France was to take up the idea when he would become duke of Aquitaine in 1137.77 Secondly, it
must be noted that when King Henry I changed his equestrian seal in 1110 or soon after – I follow
here Judith Green – and then replaced the ducal banner with a sword (pl. 5),78 this brand-new type
‘with a sword’, which had never before been used in the Anglo-Norman realm, almost immediately
began to adorn aristocratic insignia: Nigel d’Aubigny had a sword on his seal before 1114 (pl. 6),79 as
did the counts of Ponthieu, Warwick, and probably Chester in the following years,80 and many more

75
Boulogne: see above, n. 33. – Meulan-Leicester: D. Crouch, The English Aristocracy, 1070–1272. A Social Transformation,
New Haven and London 2011, ill. 7 between pp. 174 and 175. – Perche: see above, n. 60. – Breteuil: Bony, Sceaux figurés,
23 and pl. IV, no. 18, erroneously attributed to Eustace III of Boulogne; Bony’s source is a drawing made for Gaignières,
taken from a charter of Fontevraud that can be dated 1119 x 23 (Paris, BN, lat. 5480/1, 278–9, confirmed by Eustace in
1123: Grand cartulaire de Fontevraud, ed. Jean-Marc Bienvenu, Robert Favreau and Georges Pon, vol. 2, Archives
historiques du Poitou 63, 2005, 618–9, no. 653). – Briouze: Facsimiles of Early Charters in Oxford Muniment Rooms, ed.
Herbert E. Salter, Oxford 1929, nos. 2 and 3. – Saint-Clair: Demay, Inventaire des sceaux de la Normandie, 56, no. 510;
Louis Douët d’Arcq, Collection de sceaux, vol. 2, Paris 1867, 65, no. 3513, from a deed of Savigny issued between 1127 and
1130 (my thanks to Richard Allen for this dating). – Espinay: Weston S. Walford and Albert Way, ‘Examples of Mediaeval
Seals’, The Archaeological Journal 13, 1856, 62–76 at 62–5; Walter de Gray Birch, Catalogue of Seals in the Department of
Manuscripts in the British Museum, vol. 2, London 1892, 305, no. 6108.
76
Odo: see above, n. 31. – Robert: see above, n. 75. – Rotrou: see above, n. 71. – Stephen: Early Yorkshire Charters, vol. 4,
ed. William Farrer and Charles T. Clay, Wakefield 1935, 9–12, no. 9 (with ill.). – Waleran: Edmund King, ‘Waleran, Count of
Meulan, Earl of Worcester (1104–1166)’, in Tradition and Change: Essays in Honour of M. Chibnall, ed. Diana Greenway,
Christopher Holdsworth and Jane Syers, Cambridge 1985, 165–81 at 167–8 (and ill. between pp. 176 and 177); Bony, Sceaux
figurés, 26 and pl. VIII, nos. 40–1. – Gilbert: John H. Round, ‘The Introduction of Armorial Bearings into England’,
Archaeological Journal 51, 1894, 43–8, between pp. 46 and 47 (incorrect drawing); Anthony R. Wagner, ‘A Seal of [Richard]
Strongbow in the Huntington Library’, The Antiquaries Journal 21, 1941, 128–32 (photograph of Richard’s seal, which is
actually Gilbert’s, reused by his son).
77
Martine Dalas, Corpus des sceaux français du Moyen Âge. 2: Les sceaux des rois et de régence, Paris 1991, 146–7; Bony,
Sceaux figurés, pl. II and VII, nos. 7 and 37.
78
Judith A. Green, ‘Le gouvernement d’Henri Ier Beauclerc en Normandie’, in La Normandie et l’Angleterre au Moyen Âge.
Actes du Colloque de Cerisy-la-Salle, 4–7 octobre 2001, ed. Pierre Bouet and Véronique Gazeau, Caen 2003, 61–73 at 65
(contra P. Chaplais, ‘Seals and Original Charters of Henry I’, EHR 75, 1960, 260–75 at 263–5).
79
Charters of the Honour of Mowbray, 1107–1191, ed. Diana E. Greenway, London 1972, LXXXII and pl. in initio.
80
Ponthieu: Recueil des actes, ed. Brunel, LXXVI and pl. 1, no. 2. – Warwick: The Newburgh Earldom of Warwick and its
Charters, 1088–1253, ed. D. Crouch, Publications of the Dugdale Society 48, 2015, 46 and 80–2, no. 38. – Chester: T.A.
Heslop, ‘The Seals of the Twelfth-Century Earls of Chester’, in The Earldom of Chester and its Charters: A Tribute to Geoffrey
Barraclough, ed. Alan T. Thacker, Chester 1991, 181 and pl. 2.

15
seal owners in the course of the 1120s.81 Mimesis, especially royal mimesis, was definitely a crucial
process in early Anglo-Norman lay sigillography.

Plate 5: Third seal of King Henry I, engraved c. 1110. Diameter: 82 mm.

Plate 6: Seal of Nigel d’Aubigny, used 1109 x 1114. Diameter: 60 mm.

This visual propinquity to the king makes the most striking feature of the social dissemination of seals
in the Anglo-Norman aristocracy even more remarkable: more than half of the ‘first’ seals catalogued
up to 1125 originate not from first-rank magnates, but from untitled lords, some of whom were even
apparently of quite modest standing. Norman counts and post-Conquest English earls were
presumably prompt to adopt seals, but this is not properly reflected in surviving sigillographic
evidence.82 In this respect, excepting Bishop Odo, they are preceded by the ‘peripheral’ counts of

81
Henry de Port (1096–1136/53): Facsimiles of Early Charters in Oxford, ed. Salter, no. 14. – Ralph III de Tosny (c. 1102–
c. 1126): see de Gray Birch, Catalogue of Seals, vol. 2, 358, no. 6469, for Ralph IV’s seal, which, due to its design, must have
been his grandfather’s matrix (to be compared with Roger de Tosny’s very fashionable seal in Bony, Sceaux figurés, pl. X,
no. 51). – Peter I de Saint-Hilaire (1121/9–57/68): Demay, Inventaire des sceaux de la Normandie, 57, no. 518. – Robert IV
(fl. c. 1120) or Robert V fitz Erneis (d. by 1172): ibid., 30, no. 258. – Baderon of Monmouth (c. 1125–76): R.H. Ellis, Catalogue
of Seals in the Public Record Office: Personal Seals, vol. 2, London 1981, 73, P 1756 (and pl. 22). – Richard de Verly (fl.
1115/8–47/53): bone matrix found in 1849, now at St. Albans Museum (ref. 1990.812); see Albert Way, ‘Notice of a Seal
Formed of Bone, Discovered in the Abbey Church, St. Albans’, in Three Papers Read at the Meeting of the St. Albans
Architectural and Archaeological Society, October 23, 1850, London 1851, 11–5, and de Gray Birch, Catalogue of Seals,
vol. 2, 366, no. 6508 (speaks of ‘a doubtful seal’, probably because of the crude style of the engraving). While several
among the aforementioned seals are not documented before the mid-twelfth century, they all share common features that
strongly suggests that they were cut in the same period, i.e., Henry I’s reign.
82
The seals belonging to the earls created in England before Stephen’s reign appear in the following order (early seized
earldoms excluded): Surrey (1088 x 1118: see above, n. 71), Leicester (1107 x 18: see above, n. 75), Huntingdon (1116 x 18:
see above, n. 71), Warwick (1115 x 19: Newburgh Earldom, ed. Crouch, 56–7, no. 7), Chester (1120 x 29: see above, n. 80),
Buckingham (c. 1130?, see Facsimiles of Early Charters in Oxford, ed. Salter, no. 44, and Demay, Inventaire des sceaux de la
Normandie, 6, no. 30), and Gloucester (1122 x 47: Earldom of Gloucester Charters. The Charters and Scribes of the Earls and
Countesses of Gloucester to A.D. 1217, ed. Robert B. Patterson, Oxford and New York 1973, 24 and pl. XXXI, ill. A;
Nicholas Vincent, ‘The Seals of King Henry II and his Court’, in Seals and their context in the Middle Ages, ed. Phillipp

16
Boulogne, Ponthieu, Maine, and Dammartin.83 The true Anglo-Norman specificity lies in the very
early appearance of sigillants emanating from the middle ranks of the nobility: for the period up to
1100, I have already named Nigel viscount of Cotentin, Ilbert de Lacy, Gilbert Tison, and Ralph de
Mortemer. These men were, of course, no humble knights; the former belonged to an old and
prestigious lineage, and the latter were considerable – though newly enriched – Domesday tenants-
in-chief. Yet they did not pertain to the greater Anglo-Norman ruling elite. This reality seems to be
somehow acknowledged in Ilbert and Ralph’s seals (those of Nigel and Gilbert are lost): Ilbert’s shows
an unarmed knight (looking more like a squire transporting his master’s shield!) and Ralph’s is not of
the equestrian type, but instead adorned with a lion (pl. 1-2). However, it would not be long before
barons assumed a fully equestrian seal with sword or spear: the first precisely dated ones – but there
may be many others – are those of Nigel d’Aubigny, lord of the great honour of Mowbray (between
1109 and 1114), and Eustace, lord of Breteuil and Pacy (before 1119). Eustace’s seal is a copy of the
matrix made for his powerful neighbour Robert of Beaumont, count of Meulan and Leicester: what
we see here is thus a baron imitating a magnate.84 But Nigel’s seal, for its part, is obviously a direct
imitation of the newly cut matrix of King Henry I (pl. 6).85 It is probably worth remembering here that
Nigel, a ‘new man’ raised to a prominent position in order to serve the king’s purposes, exerted
authority as local justiciar in Yorkshire and Northumberland, and probably had custody of York castle
and its administrative nexus from c. 1107 onwards.86 For him, taking over the royal image was by no
means a usurpation or pretentious expectation in sharing the king’s power, but more likely a
symbolic reminder that Nigel acted as Henry’s trusted representative in local affairs. The group of
early sigillants actually includes several men who were royal servants at some point in their career:
William I de Tancarville (d. 1129), royal chamberlain, apparently sealed a charter (if genuine) for
Saint-Georges de Boscherville c. 1114;87 Walter (d. c. 1129), sheriff of Gloucestershire and constable
of England, surely validated several documents around 1120;88 likewise Henry de Port (d. 1136/53),
sheriff of Hampshire and justiciar in Kent,89 or William II de Courcy (d. c. 1140), seneschal of King
Henry I, in the 1120s.90 Henry de Port’s insignia – the only surviving one from the aforementioned list

R. Schofield, Oxford 2015, 7–33 at 19). – Those owned by the Norman counts can be traced as follows: Mortain (1100 x 04:
sealing clause in MS. Paris, BN, lat. 12878, fol. 280-r-v; for the date, see J.H. Round, Calendar of Documents Preserved in
France, Illustrative of the History of Great Britain and Ireland , vol. 1, London 1899, 436, no. 1208), Eu (1109: Achille Peigné-
Delacourt, ‘Charte de Henri, comte d’Eu. Donation et confirmation de dons à l’abbaye Saint-Lucien de Beauvais, an 1109’,
Mémoires de la Société académique de l’Oise 3, 1856, 550–7), and Aumale (1106 x 16: see above, n. 71).
83
See above, n. 33.
84
See above, n. 75, for both seals.
85
See above, nn. 78–1.
86
Charters of the Honour of Mowbray, ed. Greenway, XXIV–XXV.
87
Rouen, AD Seine-Maritime, 13 H 15 (tag for seal); printed by Albert Besnard, Monographie de l’église et de l’abbaye Saint-
Georges de Boscherville (Seine-Inférieure), Paris 1899, LXXI, from a modern copy.
88
Kew, National Archives, Duchy of Lancaster, Ancient Deeds, DL 25, no. 3 (tag for seal; printed in Ancient Charters, Royal
and Private, Prior to A.D. 1200, ed. H.J. Round, vol. 1, London 1888, 19, no. 11); Historia et Cartularium Monasterii Sancti
Petri Gloucestriae, ed. William H. Hart, vol. 1, London 1863, 188 and 246, nos. 75 and 167 (sealing clauses); The Cartulary of
Shrewsbury Abbey, ed. Una Rees, vol. 1, Aberyswyth 1975, 1–3, no. 1 (idem).
89
Facsimiles of Early Charters in Oxford, ed. Salter, no. 14.
90
Alençon, AD Orne, H 2008 (slit for tag); see J.H. Round, Calendar of Documents Preserved in France, vol. 1, 432, no. 1196.

17
– is, just like Nigel d’Aubigny’s seal, a (clumsy) copy of the royal horseman with a sword. It is
therefore tempting to suggest that the various administrative duties assumed by members of the
Anglo-Norman ruling class operated as a strong incentive for acquiring seals. In fact, a compelling
illustration of the relation between lay seals and royal service is provided by a well-known drawing of
the nearly contemporary seal of Richard Basset (d. c. 1144), the multi-sheriff and royal justice of
Henry I.91 Affixed to a charter dated 1127 x 34,92 Richard’s seal displayed a vivid allegory of the
combat between Justice and Evil, with a knight on foot fighting a demoniac creature.93 This last
example also illustrates the intense creativity of Anglo-Norman lay sigillography in the twelfth
century.

Beyond Normandy

Let us now cross the borders of ducal Normandy in order to scrutinize the development of seal usage
beyond the Anglo-Norman realm. By 1125, as we have seen, all surrounding regions, from Flanders
to Brittany, were somehow acquainted with the new aristocratic device. The most faraway groups of
sigillants were located in Anjou and Poitou (the lords of Montreuil-Bellay and Châtellerault),94 in
eastern and southern Champagne (counts of Brienne, Nevers-Auxerre, and Tonnerre)95 and, with
much greater density, in a large area including Flanders, Artois, and Picardy, as well as some
Lotharingian counties across the frontier of the German Empire. The bulk of this important group was
already observable before 1100. This is essentially where Anglo-Norman innovation and imperial
experiments in ducal sealing practice came together.

Two key developments, each associated with a major seal, need to be highlighted here. A first
momentum was the adoption of an equestrian seal by the count of Flanders in the early 1070s.
Count Arnulf III, killed in battle in 1071, apparently used a seal, but it has not been preserved.96 His
uncle and successor, Robert the Frisian, chose an equestrian seal whose mounted warrior is seen
holding a sword (pl. 7). This seal, long thought to be a forgery but definitely authentic, can be
interpreted as a quick response to the invention of the equestrian type by William the Conqueror, an
enemy of Robert since he had opposed his accession to the county of Flanders in 1071. It recalls that

91
On him, see J.A. Green, ‘Basset, Richard (d. in or before 1144)’, in ODNB (online edition:
http://www.oxforddnb.com/view/article/1647, accessed 22/9/15).
92
Basset Charters, c. 1120 to 1250, ed. William T. Reedy, Pipe Roll Society new series 50, 1995, 111–2, no. 173.
93
See Sir Christopher Hatton’s Book of Seals, ed. Stenton and Stenton, 276, no. 407, and pl. III between pp. 80 and 81, as
well as the subsequent observations drawn from the (now badly damaged) original impression by Heslop, ‘Seals’, 318, no.
372.
94
Montreuil-Bellay: see Guillot, Comte d’Anjou, vol. 2, 12 n. 32 (regrettably without reference). – Châtellerault: Bony,
Sceaux figurés, 25 and pl. V, no. 21.
95
Brienne: Cartulaires de l’abbaye de Molesme, ed. Laurent, vol. 2, no. 498 (sealing clause). – Nevers-Auxerre: Auguste
Coulon, Inventaire des sceaux de la Bourgogne, Paris 1912, no. 110. – Tonnerre: Dijon, AD Côte-d’Or, 7 H 1651 (slits for tag);
printed in Cartulaires de l’abbaye de Molesme, ed. Laurent, vol. 2, 40–3, no. 28A.
96
See Declercq, ‘Privaatoorkonde’, 64.

18
the Flemish count, as margrave of Flanders, was a peer to the duke of Normandy, and therefore
shared his newly self-awarded right to use a ‘princely’ seal.97 The keen interest that Robert accorded
to seal usage is clearly shown by the fact that the vast majority of his charters were sealed, as well as
by his sporadic use of a ‘secret’ or ‘signet’ seal (sometimes used as a counterseal, which is a rarity for
the period) and even by spectacular graphical sceneries around his applied wax impressions in some
monastic charters.98 In this context, the promotion of the equestrian type ‘with a sword’ appears as a
thoughtful response to the creation of the equestrian type ‘with a banner’. The sword was an equally
significant symbol of authority (which William also exhibited on the royal side of his seal), but it was
also graphically dissimilar enough so as to provide the counts of Flanders with their own visual
identity – an identity that Robert the Frisian’s successors would devotedly preserve by copying his
seal die for generations.99 This alternative type ‘with a sword’ remained rather marginal up to the
1120s, in sharp contrast to its increasing success from the second quarter of the twelfth century
onwards. Indeed, before 1100, as far as we know, it was only adopted by the rival family branch of
Hainault, by the Flemish castellans of Ghent, and, curiously, by the count of Dammartin (if
Gaignières’ drawing can be trusted).100 Some years later, however, the dukes of Aquitaine and
Brittany also adopted the equestrian type ‘with a sword’;101 I am inclined to interpret their choice as
a result of their joint participation in the First Crusade with Robert II of Flanders. These
considerations bring us back to the adoption of the same style by King Henry I around 1110.102 I
would suggest that the king-duke switched from banner to sword for his new ducal seal because, in
the then state of lay sigillography, the former remained much more associated with princes than the
latter, which had begun to become widespread among Anglo-Norman aristocracy and, consequently,
had already lost much of its strength as a symbol of princely status. Ironically, this royal move
ensured the dazzling success of the type ‘with a sword’, which had completely overshadowed the
banner type by the mid-twelfth century.

97
This argument is developed in Nieus, ‘Sceaux et identité’, 56–9.
98
Nieus, ‘Sceaux et identité’, 46–57.
99
Nieus, ‘Sceaux et identité’, 58–9.
100
See above, nn. 28 (Hainault), 33 (Dammartin), and 43 (Ghent). A similar seal, allegedly owned by Count Conrad I of
Luxembourg as early as 1083, is most probably the result of retrospective sealing by Conrad II in the 1130s: Nieus, ‘Cum
signo auctoritatis’, 59, n. 103.
101
Aquitaine: Eygun, Sigillographie du Poitou, pl. I, no. 1. – Brittany: see above, n. 40.
102
See above, n. 78.

19
 Plate 7: Seal of Robert the Frisian, Count of Flanders, engraved 1071. Diameter: c. 60 mm.
(Saint-Omer, Bibliothèque de l’Agglomération, MS. 803/1, 128 ; reproduced by permission)

Plate 8: Seal of Godfrey of Bouillon, Duke of Lower Lotharingia, used 1096. Diameter unknown.
(17th-cent. drawing, printed by P.F.X. de Ram, ‘Notice sur un sceau inédit de Godefroid de Bouillon’,
Bull. de l’Acad. royale des Sciences, des Lettres et des Beaux-Arts de Belgique 13, 1846, 355–60; drawing J.-F. Nieus)

A second sigillographic monument with a far-reaching influence is the seal of Godfrey of Bouillon,
duke of Lower Lotharingia from 1087 until his death in Jerusalem in 1100. It has generally gone
unnoticed in scholarship, probably because it is only known through antiquarian drawings, all taken
from one lost wax impression from the abbey of Afflighem in Brabant.103 Godfrey’s image is a slight
variation of the Conqueror’s equestrian die, with the spear being tilted forward instead of resting on
the shoulder (pl. 8). It can best be compared to the very similar representation of Manasses II, count
of Guînes (c. 1090–1136/8);104 this is not surprising given that Godfrey belonged to the comital family
of Boulogne. There is every reason to think that Godfrey was the very first duke in the Empire to
have an equestrian seal. He had only been preceded by the Lotharingian counts of Hainault and
Louvain, both neighbours and kinsmen of the count of Flanders. From then on, many German dukes
and margraves began to commission seals of the same sort. The matrix of the duke of Carinthia,
documented in 1103,105 is nothing other than a poor copy of Godfrey’s insignia, while various other
ducal seals from the 1110s and 1120s derive from an intermediary model, i.e., the seal of Godfrey’s
successor in Lower Lotharingia, another Godfrey, count of Louvain (1095–1139, duke 1106–29) and
ancestor of the future dynasty of the dukes of Brabant. Godfrey of Louvain’s seal is in turn a variation
of Godfrey of Bouillon’s die, with a horse not galloping but at walk.106 It was, for instance, copied by
the dukes of Austria and Bavaria and by the margrave of Meissen.107 Later on, during the twelfth and

103
Pierre F.X. de Ram, ‘Notice sur un sceau inédit de Godefroid de Bouillon’, Bulletin de l’Académie royale des Sciences, des
Lettres et des Beaux-Arts de Belgique 13, 1846, 355–60; Martens, ‘Reproduction manuscrite’, 7–26.
104
See Duchesne, Histoire généalogique, Preuves, 39, and a more reliable drawing in the eighteenth-century cartulary of
Saint-Bertin: Saint-Omer, Bibliothèque de l’Agglomération, MS. 803/1, 182 and pl. III in fine.
105
Kittel, Siegel, 251, ill. 156b.
106
Laurent, Sceaux des princes, vol. 1/1, 257, no. 3, and vol. 2, 90–1, no. 3.
107
Kittel, Siegel, 251, ill. 157 and 158 (Bavaria and Meissen); Oskar F. von Mitis and Franz Gall, Die Siegel der Babenberger,
Publikationen des Instituts für Österreichische Geschichtsforschung. Reihe 3, 3 1954, 5, nos. 2-3 (Austria).

20
thirteenth centuries, the Godfreys’ archetype would be fully integrated by the German nobility as the
ordinary depiction of princes.

To sum up, the equestrian seal was introduced in the German Empire by Lotharingian princes, who
had in turn received it from their Flemish and Anglo-Flemish relatives. Under Flemish (and then,
Anglo-Norman) influence, the sword type was to become a universally accepted standard in
‘Western Francia’ by the mid-twelfth century, while the banner type, though originally inspired by
the Conqueror’s seal, would eventually appear as characteristically ‘German’ imagery, which indeed
fitted very well with the customary and enduring importance of banners in the Empire as a symbol of
princely authority and feudal tenure.

The contribution of northern France to the development of a true aristocratic seal culture is actually
more far-reaching than just the invention of the type ‘with a sword’. This is evidently no wonder
given the acknowledged importance of this particular region in the creation of the social and cultural
identity of Anglo-French nobility during the High Middle Ages.108 In my opinion, at least two further
northern French premières deserve to be brought to light.

The first one is noblewomen’s seals. Over 145 non-royal feminine seals from the Anglo-Norman
realm were calendared by Susan Johns in 2003.109 None of them, however, dates from the early age
of aristocratic sealing practice. The oldest extant examples are those of Mathilda of Wallingford
(1122 x 47) and Alice, wife of Gilbert fitz Richard de Clare (1136 x 38);110 further evidence indicates
that feminine seals did not spread in England and Normandy before the mid-twelfth century.
However, feminine seals did exist 50 years prior: I have come across five of them, ranging from 1096
to 1120.111 They were cut for the following noblewomen, listed here in chronological order for the
extant data: Ida of Lotharingia, wife of Count Eustace II of Boulogne (1096);112 Clemencia of
Burgundy, wife of Count Robert II of Flanders (1099 x 1100);113 Adela, countess of Vermandois

108
See most recently David Bates, The Normans and Empire: The Ford Lectures Delivered in the University of Oxford during
the Hilary Term 2010, Oxford 2013, 170–1.
109
Susan M. Johns, Noblewomen, Aristocracy and Power in the Twelfth-Century Anglo-Norman Realm, Manchester and New
York 2003, Appendix 1.
110
Johns, Noblewomen, 126. A charter of Matilda Peverel dated 1119 x 29 refers to her personal seal, but it may have been
reworked: Charters of the Redvers Family and the Earldom of Devon, 1090–1217, ed. Robert Bearman, Devon and Cornwall
Record Society N.S. 37, 1994, 157, no. 3.
111
I do not include the supposed seal of Adela of Blois discussed by Kimberly A. LoPrete, Adela of Blois: Countess and Lord
(c. 1067–1137), Dublin 2007, 534–48. No certainty can be had that Adela owned a seal before she retired as a nun at
Marcigny in 1120; the form of a purportedly sealed letter from c. 1104 (ibid., 515–6, no. 121, and 540) looks spurious to me.
The earliest firm evidence for a countess of Champagne using a seal dates from 1137 (Baudin, Emblématique, 137).
112
de Ram, ‘Notice sur un sceau inédit’, 360, ill. 2; Martens, ‘Reproduction manuscrite’, pl. 1–3 between pp. 16 and 17.
113
Laurent, Sceaux des princes, vol. 1/1, 185, no. 95, and vol. 2, 73, no. 95: plaster cast of a seal once appended to a charter
issued in 1101 x 3, destroyed in 1940 (for the date, see Erik Van Mingroot, ‘Liste provisoire des actes des évêques de
Cambrai de 1031 à 1130’, in Serta devota in memoriam Guillelmi Lourdaux, ed. Werner Verbeke, Marcel Haverals, Rafaël De
Keyser and Jean Goossens, vol. 2, Louvain 1995, 13–55 at 33, no. 22). An original letter from Countess Clemencia for Cluny
in 1099 x 1100 had also been sealed: Recueil des chartes de l’abbaye de Cluny, ed. Auguste Bernard and Alexandre Bruel,
vol. 5, Paris 1894, 836–8, no. 3733bis.

21
(1103);114 Adela, lady of Péronne (1095 x 11);115 Emma of Arques, successively wife of Nigel de
Monville and Count Manasses II of Guînes (1120).116 These women did, of course, belong to the
upper nobility and exerted personal authority at some point in their adult life, either in quality of
heiress, dowager widow, or ‘regent’ in the absence of their husband. Female lordship was a relatively
unexceptional phenomenon in Artois and Picardy during the High Middle Ages,117 which might
evidently help in interpreting the early appearance of feminine seals in the region. Yet I would not go
as far as to say that this feature resulted from an established northern French particularism; the
explanation is therefore not fully convincing. The iconography of the four seals whose appearance is
documented reveals two possible sets of influences. All four already displayed the classical, pointed-
oval, pedestrian type, with the image of a standing lady holding a book and/or a lily. The standing
figure and the lily are to be seen on the seal of Queen Matilda (1100–1118), Henry’s first wife (which
may have been inspired by a lost seal of the Conqueror’s wife, Matilda of Flanders).118 The seal of
Emma, countess of Guînes, strongly recalls this English royal seal, with the orb simply replaced by a
book (pl. 10). Countess Ida’s seal, however, with only the book as attribute, might perhaps bear
testimony to her Lotharingian origins (pl. 9).119 She was the daughter of Duke Godfrey II of Lower
Lotharingia (d. 1069), and thus a stepdaughter of Beatrix of Bar (d. 1076), second wife of Godfrey and
Duchess of Tuscany, who owned a remarkable ‘majesty’ seal, adorned with a sitting woman holding a
book in front of her.120 Ida’s seal may also be compared with a couple of other very early German
female seals, which similarly emphasize the presence of the book as the main attribute.121 The
repeated appearance of this object – certainly the psalter, a symbol of piety – is all the more striking

114
Charles du Cange, Histoire de l’état de la ville d’Amiens et de ses comtes, Amiens 1840, 272–3, n. 1 (diplomatic mention).
115
Opera Diplomatica et Historica, ed. Aubertus Miraeus and Johannes F. Foppens, vol. 4, Louvain 1748, 512, no. 11 (sealing
clause; the dating 1095 x 11 is deduced from a confirmation by the count of Flanders in 1093 x 1111: Actes des comtes de
Flandre, 1071–1128, ed. Fernand Vercauteren, Brussels 1938, 129–30, no. 51); Charters of St-Fursy of Péronne, ed. William
M. Newman and Mary A. Rouse, Cambridge (Mass.) 1977, 26–7, no. 7 (sealing clause in 1110). We have an eighteenth-
century description, by dom Grenier, of an impression appended to a charter dated 1126: Adela’s seal was ‘ovale’ and
showed ‘une dame la tête couverte d’un voile, tenant de la main droite une sorte de fleur de lys avec sa tige ; l’empreinte
est enfoncée ; il reste de l’inscription : “sigill… …elidis Peronensis”‘ (Paris, BN, Collection de Picardie, vol. 255, fol. 146r-v).
116
Duchesne, Histoire généalogique, Preuves, 39.
117
Patrick Corbet, ‘Entre Aliénor d’Aquitaine et Blanche de Castille. Les princesses au pouvoir dans la France de l’Est’, in
Mächtige Frauen? Königinnen und Fürstinnen im europaïschen Mittelalter (11.-14. Jahrhundert), ed. Claudia Zey, Vorträge
und Forschungen 81, 2015, 225–47.
118
Harvey and McGuinness, Guide to British Medieval Seals, 48, ill. 42; Heslop, ‘Seals’, 305, no. 336.
119
On her, see Nicolas Huyghebaert, ‘La mère de Godefroid de Bouillon: la comtesse Ide de Boulogne’, in La Maison
d’Ardenne, Xe-XIe siècles. Actes des Journées lotharingiennes, 24–26 octobre 1980, Publications de la section historique de
l’Institut grand-ducal de Luxembourg 95, 1981, 43–63, and Heather J. Tanner, Families, Friends, and Allies: Boulogne and
Politics in Northern France and England, c. 879–1160, Leiden and Boston, 2004, passim.
120
Urkunden und Briefe, ed. Goez and Goez, 13–14 and pl. 8 between pp. 4 and 5.
121
In particular, the seals of Clemencia, countess of Gleiberg (documented in 1141), and, if genuine, of Adelaid of
Orlamunde (as early as 1097): see Kittel, Siegel, 277–8, ill. 181 and 182.

22
since it completely disappeared from later feminine aristocratic seals to the benefit of lilies and birds
of prey.122

Plate 9: Seal of Ida of Lotharingia, wife of Count Eustace III of Boulogne, used 1096. Dimensions unknown.
th
(17 -cent. drawing, printed by P.F.X. de Ram, ‘Notice sur un sceau inédit de Godefroid de Bouillon’, Bull. de
l’Acad. royale des Sciences, des Lettres et des Beaux-Arts de Belgique 13, 1846, 355–60; drawing J.-F. Nieus)

Plate 10: Seal of Emma of Arques, wife of Count Manasses II of Guînes, used 1120. Dimensions unknown.
(from A. Duchesne, Histoire généalogique des maisons de Guînes, d’Ardres, de Gand et de Coucy, Paris 1631, Preuves, 39)

This first generation of female seals is extremely significant, as it appears to be nearly as old as the
earliest male seals and even concomitant with the first extant seals of Anglo-Norman and Capetian
queens.123 Just how widespread feminine seals were in the early twelfth century is difficult to say,
since their conservation is obviously highly defective, but they certainly remained relatively rare
objects. For instance, when, shortly after 1103, the Countess of Roucy Sybilla de Hauteville – a
Norman princess – had to validate a charter in place of her recently deceased husband Ebles II, she
simply reused his seal matrix.124 Feminine seals seemingly had not yet spread beyond Flanders and
Picardy.

A second northern French première is heraldic seals, and probably even heraldry tout court. This is
not a completely new statement, but the situation may now be formulated on much firmer ground
than before.125 Heraldic seals (that is, seals somehow adorned with heraldic devices, notably on the

122
See, for instance, Johns, Noblewomen, 128–31. The book will only survive as an attribute on ecclesiastical seals, be it
conventual or abbatial seals (representing the patron saint or the abbess). Our early feminine seals are actually very close
to their monastic counterparts, which were obviously a source of inspiration (see Heslop, ‘Seals’, 305, no. 336).
123
Namely those of Queen Matilda (1116 x 18) and Queen Bertrade de Montfort (1115).
124
Cartulaire de Saint-Nicaise de Reims (XIIIe siècle), ed. Jeannine Cossé-Durlin, Paris 1991, 213–4, no. 38.
125
J.-F. Nieus, ‘Pourquoi les armoiries? Culture chevaleresque et construction identitaire de la haute aristocratie au XIIe
siècle’ (forthcoming). For earlier scholarship on the subject, very much influenced by Michel Pastoureau since the 1970s,
nd
see M. Pastoureau, Traité d’héraldique, 2 ed., Paris 1993, 26–32; idem, ‘La naissance des armoiries’, in Le XIIe siècle.

23
shield or banner of the equestrian figure) are all the more interesting here, as they build further
bridges between the Anglo-Norman and northern French aristocratic groups.

To summarize my results very briefly, the earliest known emblems appear on a handful of seals
belonging to nobles from the Artois and Vermandois regions. The very first one dates from the
1110s. It consists of a ‘checky’ device displayed on the banner held by Count Ralph I of Vermandois
on his equestrian seal.126 Exactly the same device was engraved on the crude seal matrix of Ralph’s
stepfather, Count Rainald II of Clermont-en-Beauvais, who had at the time remarried his mother,
Adela of Vermandois.127 This is the celebrated ‘échiqueté de Vermandois’. Further emblems are
encountered in the 1120s and 1130s on the seals of the count of Saint-Pol and the lord of Coucy, and
possibly those of the count of Roucy and the lord of Guise.128 These are the men who most probably
shaped and promoted the heraldic system and certainly had the idea of inscribing heraldic devices on
seal dies. Both concepts quickly crossed the Channel under the impulse of the great Anglo-Norman
families. As early as the late 1130s, there is sigillogaphic evidence that the Beaumont twins, Waleran
of Meulan and Robert of Leicester, had adopted the checky coat of arms of the Vermandois family, to
which they were allied through their mother.129 A sequence of remarkable seals also indicates that
the Clare family was already bearing its well-known chevronny device,130 which might possibly be
interpreted – though not without hesitation – as a reminder of the Clare’s connection with the
renowned family of Roucy, which, like the Vermandois, boasted illustrious Carolingian descent.

* * *

Mutations et renouveau en France dans la première moitié du XIIe siècle, ed. Françoise Gasparri, Paris 1994, p. 103–23;
idem, L’art héraldique au Moyen Âge, Paris 2009, 19–28. A recent publication in English is Paul A. Fox, ‘Crusading Families
rd
and the Spread of Heraldry’, The Coat of Arms, 3 series 8, 2012, 59–84.
126
Bony, Sceaux figurés, 25 and pl. V, no. 24. This celebrated seal used to be dated c. 1135, but it is already mentioned in
1110, 1120 (sealing clauses), and 1126 (description by dom Grenier): Charters of St-Fursy, ed. Newman and Rouse, 26–7,
no. 7; Claude Hémeré, Augusta Viromanduorum vindicta et illustrata, Paris 1643, Preuves, 39–40; Paris, BN, Collection de
Picardie, vol. 255, fol. 146r-v (as well as Collection Moreau, vol. 52, fol. 108r).
127
Louis Douët d’Arcq, Collection de sceaux, vol. 1, Paris 1863, 432, no. 1041 (undated impression); Cartulaire de l’abbaye
de Saint-Corneille, ed. Morel, vol. 1, 75–6, no. 3 (sealing clause in 1115).
128
Saint-Pol: J.-F. Nieus, ‘L’avoine des Candavène. Retour sur l’emblème des comtes de Saint-Pol et la naissance des
armoiries’, Archiv für Diplomatik, Schriftgeschichte, Siegel- und Wappenkunde 52, 2006, 191–212, at 199–201. – Coucy:
Recueil des chartes de l’abbaye de Clairvaux au XIIe siècle, ed. Laurent Veyssière, Collection de documents inédits sur
l’histoire de France. Section d’histoire et philologie des civilisations médiévales, série in-8° 32, 2004, 48–9, no. 20, and 612,
no. 36. – Roucy: Brussels, Archives générales du Royaume, Collection de moulages de sceaux, no. 5294 (online:
http://search.arch.be/fr/tips/99-zegelafgietsels, accessed 16/9/2015). – Guise: Donald L. Galbreath and Léon Jéquier,
nd
Manuel du blason, 2 ed., Lausanne 1977, 23, ill. 8.
129
See D. Crouch, ‘The Historian, Lineage and Heraldry, 1050–1250’, in Heraldry, Pageantry and Social Display in Medieval
England, ed. Peter Coss and Maurice Keen, Woodbridge 2002, 17–37 at 29–33.
130
Crouch, ‘Historian, Lineage and Heraldry’, 32–5.

24
The study of the origins and early diffusion of aristocratic seals reveals the importance and interest of
these artefacts as a complementary source for the political, social, and cultural history of the Anglo-
Norman ‘empire’ and its periphery. For too long, scholars have approached these tiny pieces of
coloured wax as lovely, though useless, oddities of medieval diplomatics. Yet things are now slowly
changing, in Great Britain as well as in other countries, under the impulse of a new (or renewed)
enthusiasm for topics such as the role of images, the use of the written word, and the shaping of
individual and corporate identities in medieval societies.131 However, much remains to be done in
terms of the preservation and cataloguing of seals, self-evidently, as well as in terms of a true
incorporation of seal studies into research agendas dealing with the Anglo-Norman and European
aristocracies. Even within the restricted sphere of early princely and baronial seals, on which the
present study has focused, further investigations still need to be conducted in various directions. In
particular, a better understanding of the uses of these seals in diplomatic practice would be highly
desirable, as it would clarify the apparently random character of the recourse to sealing – with or
without other signs of validation – for the authentication of English and Norman lay charters until
well into the first half of the twelfth century. This kind of study, of course, requires sharp scrutiny not
only of the seals themselves, but also of the related documents in their archival as well as social
environment. A similar emphasis on the broader context will be necessary to interpret the
iconographical propinquities that we have noticed here and there between pairs or in larger sets of
seals. For example, such prominent men like Stephen-Henry of Blois, Robert de Beaumont, and
Eustace of Breteuil shared a fully identical equestrian design on their respective seal dies: these
connections are undoubtedly significant, and therefore deserve more focused attention from
political historians.

131
The papers collected in the proceedings of two recent conferences give a good illustration of this revival: Pourquoi les
sceaux? La sigillographie, nouvel enjeu de l’histoire de l’art. Actes du colloque organisé à Lille, Palais des Beaux-Arts, les 23–
25 octobre 2008, ed. Jean-Luc Chassel and Marc Gil, Villeneuve d’Ascq 2011; Seals and their Context in the Middle Ages, ed.
P.R. Schofield, Oxford 2015.

25