Kinship and identity in eleventh-century Normandy:

the case of Hugh de Grandmesnil,

c. 1040e1098

Mark Hagger

Brasenose College, Oxford, OX1 4AJ, UK

Abstract

Hugh de Grandmesnil was one of the co-founders of the Norman monastery of Saint-Evroult. It was

no doubt his part in this foundation that led Orderic Vitalis, a monk of that house, to provide an account
of Hugh’s career in his

Historia ecclesiastica. The information found there provides an almost unique

opportunity to observe an individual of the eleventh century in the context of nearly all of his family
connections. This article uses that evidence first to examine Hugh’s relationships with his kinsmen and
to ask whether they acted together so as to form, in Sir James Holt’s words, a ‘mutual benefit society’,
and secondly to consider the extent to which Hugh’s identity was defined by his relations with his kins-
men. The findings of this inquiry reveal, amongst other things, that the importance of Hugh’s relation-
ships with his kinsmen varied over the course of Hugh’s career, and that the pool of kinsmen, friends,
and allies to whom Hugh could turn in time of need was equally fluid. Hugh’s career therefore stands as
a corrective to frequently held assumptions that the relationships forged by kinship and marriage
between members of the secular elite of eleventh-century Normandy remained stable throughout an
individual’s life.
Ó 2006 Published by Elsevier Ltd.

Keywords: Kinship; Family; Aristocracy; Marriage; Identity; Normandy; Orderic Vitalis

Grandmesnil is one of the more familiar Norman family names of the eleventh century.

It is a name that is known, above all, for its part in the foundation of a monastery in the
Pays d’Ouche, in association with William fitz Giroie, and other members of his family,

E-mail address:

mark.hagger@history.ox.ac.uk

0304-4181/$ - see front matter

Ó 2006 Published by Elsevier Ltd.

doi:10.1016/j.jmedhist.2006.07.005

Journal of Medieval History 32 (2006) 212e230

www.elsevier.com/locate/jmedhist

at a place now known as Saint-Evroult-Notre-Dame-du-Bois (Orne). This monastery became
the home of perhaps the greatest of the twelfth-century Anglo-Norman chroniclers, Orderic
Vitalis, whose

Historia ecclesiastica naturally recorded the deeds of the founders of his

house.

Orderic provides an almost unique opportunity to observe an individual of the eleventh

century in the context of nearly all of his family connections. The chronicler provides
a full list of the marriages made by the members of the Grandmesnil family from before
c. 1040 until 1136, allowing for a detailed study of the kinship network which surrounded
Hugh, and within which he operated, and an examination of the extent to which kinship
could affect a man’s career in the mid- to late-eleventh century. Questions concerning the
strength and uses of kinship bonds lead to related questions about how such bonds affected
Hugh’s identity. Throughout his career Hugh de Grandmesnil seems to have seen himself as
a member of the lineage of Grandmesnildan agnatic family line of the type that historians
once believed only developed in the eleventh century, but which had probably always co-
existed with the broader kin-grouping.

Further layers in Hugh’s own identity, however,

were grounded in his active association with the kin-group of the Giroie, and also in his
relationships with his wife’s family and friends in the Beauvaisis and with his kinsmen by
marriage in the Pays d’Auge. In contrast to his apparently constant sense of being the
head of the house of Grandmesnil, the importance of these relationships seems to have varied
over the course of Hugh’s career, and the pool of kinsmen, friends, and allies to whom Hugh
could turn in time of need was equally fluid. This stands in contrast to the assumption that
such relationships were secure and stable often found in the vast literature on marriages and
kinship ties in the middle ages.

Modern historians have tended to study the Giroie family rather than that of Grandmesnil: see J.-M. Maillefer,

‘Une famille aristocratique aux confins de la Normandie: les Ge´re´ au XI

sie`cle’,

Autour du pouvoir ducal Normand

XII

Sie`cles, Cahier des Annales de Normandie, 17 (Caen, 1985) (henceforth Les Ge´re´), 175e206; P. Bauduin,

‘Une famille chaˆtelaine sur les confins normanno-manceaux: les Ge´re´ (X

XIII

s.)’,

Arche´ologie Me´die´vale, 22

(1992), 309e56.

It is not possible here properly to survey the vast amount that has been written on family models. Among

the most pertinent works (in English) are M. Bloch,

Feudal society: I. the growth of ties of dependence, trans.

L.A. Manyon (London, 1961); G. Duby,

The chivalrous society, trans. C. Postan (London, 1977); K. Schmid, ‘The

structure of the nobility in the earlier middle ages’ in:

The medieval nobility, ed. and trans. T. Reuter

(Amsterdam, 1978), 37e59; S.D. White,

Custom, kinship and gifts to saints: The laudatio parentum in western

France, 1050e1150 (Chapel Hill, 1988); C.B. Bouchard, ‘Family structure and family consciousness among the ar-
istocracy in the ninth to eleventh centuries’,

Francia, 14 (1986), 639e58; C.B. Bouchard, Those of my blood: con-

structing noble families in medieval Francia (Philadelphia, 2001); D. Crouch, The birth of nobility (Harlow, 2005),
chs. 4 and 5.

On the making of marriages and relationships between kinsmen see, for example, G. Duby,

Medieval marriage:

two models from twelfth-century France, trans. E. Forster (London, 1978); J.C. Holt, ‘Feudal society and the family in
early medieval England: I. The revolution of 1066’,

Transactions of the Royal Historical Society, fifth series, 32

(1982), 193e212; J.C. Holt, ‘Feudal society and the family in early medieval England: III. Patronage and politics’
Transactions of the Royal Historical Society, fifth series, 34 (1984), 1e25; E. Searle, Predatory kinship and the
formation of Norman power (Berkeley, 1988); G. Duby, Love and marriage in the middle ages, trans. J. Dunnett
(Cambridge, 1994); J. Green,

The aristocracy of Norman England (Cambridge, 1997), chs. 10 and 11; G. Althoff,

Family, friends and followers: political and social bonds in early medieval Europe, trans. C. Carroll (Cambridge,
2004).

213

M. Hagger / Journal of Medieval History 32 (2006) 212e230

Background, career, and connections before c. 1063

The origins of the Grandmesnil family

are lost. Hugh’s father, Robert I de Grandmesnil,

simply emerges in the mid-eleventh century in the midst of a war against Roger de Beau-
mont and, apparently, with his toponymic. As this appears only in the pages of Orderic’s
Historia, however, and as the monk had mid-twelfth century ideas about the use of family
names, it should perhaps be noted that Hugh de Grandmesnil was the first man to appear
with the toponymic independently of Orderic’s works, appearing in the second of the two
versions of the foundation charter for Saint-Evroult as

Hugo de Grentemaisnil. This act is

dated to September 1050.

He is given the toponymic again in the witness list of an act of

1055 for Marmoutier.

Land and lordship over men provided the foundation for the power and wealth of the

secular e´lite, and was also often linked with a family’s identity. This is, of course, often
revealed in the use of a toponymic. While it has been shown that toponymics could simply
reflect a man’s residence at a particular place, and might change if and when he ceased to
live there,

this was not the case here. Hugh de Grandmesnil was lord of Grandmesnil, and

his name did not change at any point during his lifetime. The epitaph written for his tomb
by Orderic makes this connection: ‘Grandmesnil was the name his castle bore, / From
which his name earned honour with all men’.

Whether the name was at first intended as a hereditary

cognomen is open to doubt, but no

matter how it was first used, it is clear that the toponymic had been taken up by family members
to the exclusion of any other name some years before the end of the eleventh century. Thus,
Hugh de Grandmesnil’s son, William, and his progeny retained the Grandmesnil name despite
becoming lords of a huge fief in Calabria when William married Mabel, the daughter of Robert
Guiscard, duke of Apulia, Calabria, and Sicily in

c. 1089.

William himself was not unusual in

In this article, ‘the Grandmesnil family’ is intended as a convenient synonym for the group of kinsmen related by

blood in the male line to Robert I de Grandmesnil. Medieval writers generally used the terms

stirps or gens to refer to

such groups (C.B. Bouchard,

Strong of body, brave and noble: chivalry and society in medieval France (London, 1998),

67e8).

Paris, Bibliothe`que Nationale de France (henceforth BnF), MS lat. 11055 (s. xiii, Cartulary of Saint-Evroult), f.

16v, no. 15;

Orderici Vitalis ecclesiasticae historiae libri tredecim, ed. A. le Pre´vost (Socie´te´ de l’Histoire de

France, vol. 5, Paris, 1855) (henceforth, le Pre´vost), 78e80, no. iii.

Recueil des actes des ducs de Normandie de 911 a` 1066, ed. M. Fauroux (Me´moires de la Socie´te´ des Antiquaires de

Normandie, 36, Caen, 1961) (henceforth,

RADN ), no. 137. The name was generally written ‘Grentemaisnillio’ or

‘

Grentemaisnil’ by contemporaries, although Orderic once renders it ‘Mansio Grentonis’ and an act of 1066 has ‘Grento

Mansionili’, which more clearly reveals that it had a personal-name element (Orderic Vitalis, Historia ecclesiastica, ed.
and trans. M. Chibnall, 6 vols (Oxford, 1968e81) (henceforth Orderic), vol. 2, 16, 32; le Pre´vost, 189, no. xxvii;

RADN,

no. 228). On the basis of a mistranslation of the epitaph of Hugh de Grandmesnil copied into Orderic’s

Historia, Barbara

Walker has supposed that Grento was the founder of the Grandmesnil family (B. Walker,

The Grandmesnils: a study in

Norman baronial enterprise (unpublished Ph.D. thesis, University of California at Santa Barbara, 1968), 10e11). There
is no evidence to support this theory.

As was the case, for example, with Salo de Seignelay, who changed his name to Salo de Bouilly as soon as he in-

herited Bouilly from his father and stopped residing at Seignelay (C.B. Bouchard, ‘The structure of a twelfth-century
French family: the lords of Seignelay’,

Viator, 10 (1979), 46).

Evidenced by Geoffrey di Malaterra,

De rebus gestis Rogerii Calabriae et Siciliae comitis et Roberti Guiscardi ducis

fratris eius, bk 4, ch. 22, ed. E. Pontieri (Rerum Italicarum scriptores, 5/1, Bologna, 2

edn, 1927e8), 100;

The Alexiad

of Anna Comnena, trans. E.R.A. Sewter (Harmondsworth, 1969), 348; and Carte latine di abbazie Calabresi, ed. A.
Pratesi (Studi e Testii, 197, 1958), no. 34.

214

M. Hagger / Journal of Medieval History 32 (2006) 212e230

this respect. In late-eleventh-century Italy, first-generation colonists such as William de Grand-
mesnil emphasised their Norman origins. By the early years of the twelfth century, however, the
first generation of Italo-Normans to have been born in Italy were referring to themselves as

sons

of Normans, and before long high-status Italo-Normans such as Roger di San Severino and
Robert d’Eboli dropped their Norman toponymics, if they ever had them, in favour of Italian
ones.

That William’s son, Robert, retained his Norman toponymic up to the time that he

was exiled from the kingdom of Sicily in 1130 is, therefore, notable.

Closer to home, Ivo

de Grandmesnil, Hugh’s fourth son and his successor in England, also retained his family top-
onymic, despite no longer residing at, or having an interest in, the family estates in Normandy
in general, or Grandmesnil in particular.

It seems clear, then, that by the time Hugh de Grand-

mesnil died in 1098dand probably for some years previouslydthe toponymic that linked
members of the family to their seat at Grandmesnil had become both hereditary and indepen-
dent of any actual attachment to the place. In short, it had become one of the keystones of
a sense of family identity.

Grandmesnil (Calvados) itself is today a small village, located about twelve and a half

miles east of Falaise (Calvados), and set on the summit of a gentle spur above the River
Oudon. The earthworks of the Grandmesnils’ castle, mentioned in Hugh’s epitaph, still
stand, although they are now within the borders of the neighbouring commune of
Norrey-en-Auge (Calvados).

Aside from Grandmesnil, the foundation deeds for Saint-

Evroult and the donations Hugh made to other religious houses reveal that the Grandmes-
nils’ estates were located in two main clusters. The first was around Grandmesnil itself,
roughly in the area bounded by the valleys of the Dives and Oudon. The second, smaller,
group lay further west, between the Rivers Laize and Laizon. There were in addition more
outlying possessions on the Rivers Touques and Orne, and close to Caen (Calvados).

The

location of these lands appears to have been the main cause of Robert I de Grandmesnil’s
marriage to Hawisa fitz Giroie, which probably took place in the period 1020e30. Her fam-
ily was establishing itself in Normandy at this time, and doing so by making marriages with
neighbouring families.

G. Loud,

The age of Robert Guiscard: southern Italy and the Norman conquest (Harlow, 2000), 286. On the other

hand there was a Robert de Magny from Calvados whose name was taken from the female side by a Lombard lineage
(Loud,

The age of Robert Guiscard, 287).

Alexander of Telese,

Ystoria Rogerii regis Sicilie, Calabrie atque Apulie, bk 1, chs. 17, 20: ed. L. de Nava and D.

Clementi (

Fonti per la storia d’Italia, 112, Rome, 1991), 16, 18, 103, 105.

For evidence that Ivo continued to be known as Ivo de Grandmesnil after 1098 see

Calendar of charter rolls pre-

served in the Public Record Office, vol. 4 (London, 1912), 182; Regesta regum Anglo-Normannorum, ii. Regesta Henrici
Primi, ed. C. Johnson and H. A. Cronne (Oxford, 1956), no. 1990; Orderic, vol. 6, 304.

Orderic implies the existence of the castle while writing of the events of 1091 and expressly mentions it

in the epitaph composed for Hugh of Grandmesnil in, or around, 1098 (Orderic, vol. 4, 230, 336). For a des-
cription, picture, and a plan of the surviving earthworks see J. Decae¨ns, ‘Le Patrimoine des Grentemesnil en
Normandie, en Angleterre et en Italie aux XI

et XII

sie`cles’, in:

Les Normands en Me´diterrane´e dans le sillages

des Tancre`de, ed. P. Bouet and F. Neveux (Caen, 1994), 129, 132. The foundation deed reveals that
Norrey-en-Auge was already a settlement distinct from Grandmesnil by

c. 1050 (Orderic, vol. 2, 32; RADN,

no. 122).

RADN, no. 122, Regesta regum Anglo-Normannorum, the acta of William I (1066e1087), ed. D. Bates (Oxford,

1998) (henceforth,

Regesta), nos. 212, 50, and 166. Decae¨ns, ‘Le Patrimoine des Grentemesnil’, 130e1 has maps of

the family’s estates.

Les Ge´re´, 179, 182.

215

M. Hagger / Journal of Medieval History 32 (2006) 212e230

Hugh, who was probably born in around 1023e1025,

was almost certainly the eldest of

Robert I’s sons, although he was not the heir to the whole of his father’s patrimony. In one
of his interpolations into the

Gesta Normannorum ducum of William de Jumie`ges, Orderic

notes that Robert I de Grandmesnil ‘divided his land equally between his sons, Hugh and Rob-
ert, and entrusted his youngest son, Arnold, to their care, so that when he came of age they
would treat him correctly as their brother’.

Aside from being a call for fraternal unity, this

may have meant that Arnold was to be given a share in his father’s estates when he achieved
his majority. He had certainly acquired an interest in them by

c. 1050, as the foundation charter

for Saint-Evroult makes it clear that all three brothers held a share in these lands.

So too does

Ansfrid the Dane

Fig.1 : Table showing Hugh de Grandmesnil’s
immediate family (with some omissions)

Thurstan Goz

Giroie

Robert I = 1. Hawisa 2. = William d’Evreux

William fitz Giroie

Robert = Adelaide

Arnold d’Échauffour

Judith = Great Count Roger of Sicily

Count Ivo de Beaumont

Adeliza de Beaumont =

Hugh

Robert II

Arnold

Adeliza = Humfrey du Tilleul

Robert of Rhuddlan

Arnold du Tilleul

Robert III =

William=

Hugh

Ivo =

Aubrey

1. Agnes dau. of
Ranulf de Briquessart

Mabel dau. of
Robert Guiscard

dau. of
Gilbert de Gant

2. Emma dau. of Robert

d’Étoutteville

Robert

Ivo II

3. Lucy dau. of Savaric
son of Cana

Richard de Courcy

Ralph de Montpinçon

Adelina =

Rohais =

Matilda =

Agnes =

Hawisa

Roger d’Ivry

Robert de Courcy

Hugh de Montpinçon

William de Sai

As both Hugh and his brother Robert inherited land on their father’s death and their brother Arnold did not, it may

be supposed that Hugh had reached his majoritydor was at least close to reaching itdwhen his father died. Robert I de
Grandmesnil died sometime in the early 1040s, perhaps in 1043 (see below), so that a date of 1023e1025 for Hugh’s
birth seems about right. This would make him around 73e75 years old when he died in 1098dat which point Orderic
calls him ‘worn out with old age’ and a ‘veteran’ (Orderic, vol. 4, 336).

William de Jumie`ges,

Gesta Normannorum ducum, ed. and trans. E. van Houts, 2 vols (Oxford, 1992e1995) (hence-

forth Jumie`ges), vol. 2, 96.

RADN, no. 122.

216

M. Hagger / Journal of Medieval History 32 (2006) 212e230

an act for the ducal abbey of Montivilliers in which Hugh, Arnold, and Robert II all consented
to the gift by a certain Hawisadundoubtedly their motherdof half of the manor of O (Orne)
with the church and tithe.

Like many of their contemporaries in Normandy, therefore, the sons

of Robert I de Grandmesnil divided their patrimony between themselves, rather than acknowl-
edging the right of the eldest son alone to inherit.

Hawisa was the sister of William fitz Giroiedanother man to loom large in the pages of

Orderic’s

Historia. William was a tenant of the Belleˆme family for the lands he held at Hau-

terive and, possibly, Saint-Ce´neri-le-Ge´rei (both Orne). His tenurial position was, perhaps
unexpectedly, to affect Hugh’s own early career. Hugh’s first known action was to assist Ivo
de Belleˆme, bishop of Se´es (Orne) and lord of Belleˆme, to drive members of the Surdon family
from the cathedral church at Se´es, which they had taken over and were using as a fortress, in
around 1049.

Hugh de Grandmesnil himself had no known independent connection with

Bishop Ivo, Se´es, or the Belleˆmois that would have caused him to come to the bishop’s aid.
His presence at Se´es was, therefore, almost certainly the result of his kinship with William
fitz Giroie, who would have wished to support Bishop Ivo in his struggle, not only because Wil-
liam then held Hauterive from him, but also because of the two men’s shared sympathies for
Count Geoffrey Martel of Anjou which made them natural political allies.

This provides one example of how Hugh de Grandmesnil’s early career was dominated by

his relationship with his Giroie kinsmen. Undoubtedly the most significant effect of the close
relationship between Hugh, his brother Robert, and their uncle, however, was the outcome of
their joint decision to found the monastery of Saint-Evroult in around 1050.

This decision

meant that Hugh and Robert II had to give up their plans to establish an abbey at Norrey-
en-Auge, where a body of monks under Abbot Gilbert of Conches had already commenced
monastic life.

Williamdif the speech put into his mouth was not simply the creation of

Orderic’s imaginationdprovided some justification for this change of plans by pointing out
that the necessary resources were lacking at Norrey, although this had not, so far as we
know, bothered the monks already living there, and by remarking on the sanctity of Saint
Evroult (d. 706)

who had established a monastery on the same spot in the Pays d’Ouche.

Although William apparently did not mention it in his speech, both parties would have been

Regesta, no. 212.

For a discussion of the inheritance practices found in Normandy at this time and later see D. Bates,

Normandy before

1066 (London, 1982), 118e20 and E.Z. Tabuteau, ‘The role of law in the succession to Normandy and England, 1087’,
Haskins Society Journal, 3 (1991), 141e69. There is no evidence that the division of the inheritance was made with any
regard to a distinction between lands that Robert I had inherited and those he had acquired.

Jumie`ges, vol. 2, 114. This disturbance may have had its originated in the hanging of Walter de Surdon by Robert de

Belleˆme in the late 1030s or early 1040s (Jumie`ges, vol. 2, 56). The incident may be dated to shortly before the Council
of Reims in 1049 at which Bishop Ivo was censured for his actions (Jumie`ges, vol. 2, 116e8).

For the Giroie’s support of Count Geoffrey see Orderic, vol. 2, 46e8, 66e8; Bates,

Normandy before 1066, 80; Les

Ge´re, 183. Importantly, Bishop Ivo had ‘made peace with the Giroie and all his other neighbours’ following the muti-
lation of William fitz Giroie at the hands of William Talvas, Bishop Ivo’s brother, in

c. 1046 (Jumie`ges, vol. 2, 112). It is

likely that William fitz Giroie held Saint-Ce´neri-le-Ge´rei from Geoffrey de Mayenne, a vassal of Count Geoffrey Mar-
tel, by this date.

Jumie`ges, vol. 2, 136; Orderic, vol. 2, 14e6.

Orderic, vol. 2, 20. Jumie`ges, vol. 2, 136, in contrast, gives the impression that monastic life had not yet started at

Norrey.

St Evroult had been born into a wealthy family of Bayeux in 626. He later took up an eremitical life, gathering

a group of followers around him, and eventually establishing fifteen monasteries in and around Saint-Evroult (see
Orderic, vol. 3, xvexx, 264e302).

217

M. Hagger / Journal of Medieval History 32 (2006) 212e230

aware that this move would also transfer the monastery from an area already securely held by
the Grandmesnil into one where William was looking to increase his own power and influence,
but where the Grandmesnil family had very little property.

Orderic’s record of these discus-

sions in both his

Historia ecclesiastica and in his interpolations in the Gesta Normannorum

ducum make it clear that it was William fitz Giroie who had decided where the monastery
should be located, and that the Grandmesnil brothers agreed to change their own plans to
accord with his.

Robert de Grandmesnil subsequently decided to become a monk at Saint-Evroult, no doubt

in the knowledge that his position as a co-founder of the monastery would ensure that he would
achieve high office in his new vocation. Indeed, he became prior almost immediately after
taking his vows in 1050, and then in June 1059 was made abbot.

As abbot, Robert II de

Grandmesnil was not afraid to use his Giroie relatives to bolster his authority. During his short
rule (June 1059eJanuary 1061) Robert II turned to his cousin, Arnold d’E

´ chauffour, for help

against the abbey’s men, Baudri and Vigor de Boquence´, who were refusing to obey him
and his monks and ‘inflicting all kinds of annoyances on their men’.

Abbot Robert gave

Arnold the lordship of the Boquence´ brothers for his lifetime. After suffering under Arnold’s
burdensome rule for a short period, Baudri and Vigor begged to be restored to the abbey
and promised good behaviour from then on.

Although we know relatively little of Hugh de Grandmesnil’s career during the 1050s, it is

clear that it cannot have been unaffected by the rebellions of his Giroie kinsmen in around
1059.

It is not known if Hugh himself was directly involved, as Orderic does not name

him as taking part in any of the events. Whatever the precise cause, ‘the quick-tempered
duke gave full rein to his anger, and singling out the knights Ralph de Tosny, Hugh de Grand-
mesnil and Arnold d’E

´ chauffour and their men deprived them of their lands, and drove them

into long exile without any proof of guilt’.

Orderic suggests that Hugh’s exile was a conse-

quence of the unspecified ‘disturbances of the time’, while at the same time insinuating that
it was really due to Roger de Montgommery’s plotting.

If this fails to make adequately clear

the importance of the part played by kinship and family in this expression of the ducal anger, it
is at least made explicit in the case of Abbot Robert II de Grandmesnil, who was also caught up
in the aftermath of the rebellions. He was exiled from Normandy in January 1061, in part, per-
haps, because he had derided the duke, making insulting comments about him to his prior (who
showed himself to be an untrustworthy confidant),

but mainly because of his kinship with the

Giroie.

Orderic, vol. 1, vii.

Jumie`ges, vol 2. 140e2; Orderic, vol 2. 20.

Orderic, vol. 2, 58, n.4e60, 74; V. Gazeau,

Recherches sur l’histoire de la principaute´ Normande (911e1204): I. Les

abbe´s Be´ne´dictins de la principaute´ Normande (911e1204): II. Prosopographie des abbe´s Be´ne´dictins (911e1204)
(Me´moire d’habilitation a` diriger des recherches, Universite´ de Paris I-Pantheon-Sorbonne, 2002), 412 and 301.

Orderic, vol. 2, 80e2.

Orderic, vol. 2, 78e80.

Orderic, vol. 2, 90.

Orderic, vol. 2, 90. Duke William’s treatment of the burghers of Alenc¸on, and Henry I’s of Luke de la Barre (Ju-

mie`ges, vol. 2, 124; Orderic, vol. 6, 352e4), illustrate that jokes at the duke’s expense might be treated as attempts to
undermine his authority.

Orderic, vol. 2, 90; Jumie`ges, vol. 2, 154. As Robert had trained to be a knight at the duke’s court (Orderic, vol. 2,

40), it may be that his offence was considered to be all the greater.

218

M. Hagger / Journal of Medieval History 32 (2006) 212e230

Abbot Robert II subsequently made his way to Italy, where, once again, his kinship to the

Giroie family proved to be a blessing as William fitz Giroie’s son, William de Montreuil,
gave him half of the town of Aquino (Campania). Although Robert seems to have been
appointed abbot of Robert Guiscard’s foundation of Sant’Eufemia soon after his arrival in
the Norman-controlled south of Italy, the subsequent marriage of Robert II’s half-sister Judith
to Great Count Roger of Sicily in 1062 did not do his prospects any harm. Indeed, the marriage
may have led to further favour in the form of Robert II’s receipt of the ducal abbeys at Venosa
(Basilicata) and Mileto (Calabria).

In turn, both the favour in which Robert II was held by

Guiscard and the marriage between Judith and the Great Count opened the way for William
de Grandmesnil’s marriage to Matilda, Robert Guiscard’s daughter, albeit that it occurred three
years after Robert II de Grandmesnil had been poisoned at the hands of a malcontent Saracen
baker.

Meanwhile, back in Normandy, Arnold d’E

´ chauffour had burned the town of Saint-

Evroult ‘and for many hours he and his minions stormed into every corner of the monastery,
brandishing their naked swords and clamouring for Abbot Osbern’s blood’,

as a result of

the deposition of Robert II de Grandmesnil from his abbacy and his replacement by the duke’s
nominee, Osbern. What Robert II de Grandmesnil felt about this is not known, but Arnold’s
actions might suggest that Robert’s kinsmen considered it their duty to avenge the injusticesdif
such they weredhe had suffered on their behalf.

Hugh and Robert II de Grandmesnil’s kinship with the Giroie, therefore, certainly

affected their decisions, and probably had a significant impact on how they and other mem-
bers of the family saw themselves. It certainly affected how outsiders saw them. Hugh’s
brother, Robert II, was said to have ‘stemmed from the courageous family of Giroie’,

and his sister, Adeliza is described by Orderic as being ‘of the distinguished Giroie fam-
ily’.

Robert II’s exile from Normandy is put down by Orderic to the fact that the duke

was ‘raging against him and all his kindred’,

and because he ‘stemmed from the

courageous family of Giroie’.

It may also be noted that in Duke William’s charter for

Saint-Evroult of 1050, Robert fitz Giroie witnesses immediately before ‘Hugh his nephew’,
which indicates not only Robert fitz Giroie’s superior standing in comparison to that of
Hugh de Grandmesnil, but also provides additional evidencedfor once independent of
Ordericdto show that Hugh was located by his contemporaries within the Giroie family.

That William fitz Giroie was able to exert the influence he could over his young nephews
was almost certainly due to the premature death of their father; and his untimely demise is
also likely to account for Hugh and Robert’s close identification with their mother’s family
in the years before

c. 1061.

Orderic, vol. 2, 100e2. It seems clear from both Orderic’s

Historia ecclesiastica and Geoffrey di Malaterra’s De

rebus gestis Rogerii Calabriae that Robert II de Grandmesnil had already established himself in Norman Italy before
Judith arrived (Orderic, vol. 2, 100e2; Geoffrey of Malaterra,

De rebus gestis Rogerii Calabriae, bk 2, ch. 19, ed. Pon-

tieri, 35.

Orderic, vol. 4, 22.

Orderic, vol. 2, 92.

Jumie`ges, vol. 2, 154.

Orderic, vol. 4, 136.

Orderic, vol. 2, 90.

Jumie`ges, vol. 2, 154.

RADN, no. 122.

219

M. Hagger / Journal of Medieval History 32 (2006) 212e230

After the Giroie, c. 1063-1098

When Hugh de Grandmesnil was recalled from exile in

c. 1063, he was around 38 to 40

years old. For all his life, he had been closely associated with his Giroie kinsmen. That was
now to change. Although Hugh himself was restored to his pre-

c. 1061 position, his Giroie

relatives were not. Although Arnold d’E

´ chauffour, the chief fomenter of the rebellion, was

allowed to return to Normandy within two or three years of being exiled, he was murdered
soon afterwards as the result of a plot formulated by Mabel de Belleˆme.

The Giroie family

then remained landless and in exile until Saint-Ce´ne´ri-le-Ge´rei was restored to Robert de
Saint-Ce´ne´ri in 1088dan event that does not seem to have resurrected the close connection
between the two families.

That is not to say that Hugh was now bereft of kinsmen to turn to for support when it

was needed. Before his exile in

c. 1061, he had established himself in the Beauvaisis in the

French Vexin. This new direction in Hugh’s career is best evidenced by his marriage with
Adeliza, the daughter of Ivo, count of Beaumont-sur-Oise (Val d’Oise),

probably before

his exile in

c. 1061, and in his appointment, at an uncertain date, as custodian of part

of Neufmarche´ (Eure), which stood on the frontier between Normandy and the Vexin.

Orderic speaks of Roger de Montgommery recommending Hugh for this post because he
‘was jealous of his neighbour’s courage and anxious to devise some misfortune for
him’.

This may be a hint either that Hugh’s exile in

c. 1061 had been partly caused

by a fear that he would assist the men of the Beauvaisis in their attacks on Normandy,
or, if Hugh were given custody of the town only after

c. 1063, that Roger was hoping

that the appointment could be manipulated to re-ignite the duke’s suspicions about Hugh’s
loyalty and bring about a second, no doubt more permanent, exile. That may be why
Orderic takes such pains to celebrate Hugh’s successes, and to defend his later failure in
battle against Count Ralph of Amiens-Valois-Vexin.

It seems likely that Hugh spent at least part of his exile at Beaumont-sur-Oise with his wife

and her kinsmen. The friendships he cultivated there, either before or during his exile, were to

Orderic, vol. 2, 122e4.

Orderic, vol. 4, 154e6. The petition for a new confirmation charter for Saint-Evroult made jointly by Robert III de

Grandmesnil and Robert de Saint-Ce´ne´ri to King Henry I in 1128 (BnF, MS lat. 11055, f. 20ve21r, no. 18) is the only
evidence for any continuing association between the two families and is not enough on its own to suggest that the once
strong relationship between the two houses had been restored. The longer version of this act found in

Gallia Christiana,

11:Instr. 204e10 (apparently ‘

ex authentico’) is a forgery probably concocted from a number of later twelfth-century

acts in the abbey cartulary.

Orderic, vol. 4, 230, 338. Beaumont-sur-Oise is 30 kilometres north of Paris.

Pierre Bauduin and Daniel Power have both recently suggested that Hugh was active at Neufmarche´ in the 1050s (P.

Bauduin,

La premie`re Normandie (X

sie`cles). Sur les frontie`res de la Haute Normandie: Identite´ et construction

d’une principaute´ (Caen, 2004), 274e6; D. Power, The Norman frontier in the twelfth and early-thirteenth centuries
(Cambridge, 2004), 339.). There is nothing in Orderic’s narrative that prevents an interpretation by which Hugh was
appointed after his return from exile in

c. 1063, however. Orderic expressly states that there was an interval between

the removal of Geoffrey de Neufmarche´ from office, at unknown date, and Hugh’s appointment (Orderic, vol. 2,
130). It is also the case that the only fighting that Orderic mentions in which Hugh was involved must date to after
Hugh’s return to Normandy in

c. 1063. This dating is provided by Orderic’s story of the injury and entry into Saint-

Evroult of Richard de Heudicourt. He was received into the monastery by Abbot Osbern, who had been appointed
as abbot only after Robert II de Grandmesnil had been deposed and Hugh had been exiled (Orderic, vol. 2, 132).

Orderic, vol. 2, 130.

Orderic, vol. 2, 130e2. Orderic emphasises that Hugh had inferior numbers when confronting Count Ralph.

220

M. Hagger / Journal of Medieval History 32 (2006) 212e230

serve him well. Waleran de Breteuil in the Beauvaisis was to speak up for Hugh and the other
exiles in

c. 1063; an intervention which appears to have been instrumentaldalong with the

duke’s need for manpower in his fight against the Manceauxdin making possible Hugh’s return
to Normandy.

Later, in 1091, Waleran’s son Theobald, along with Count Matthew of Beaumont-

sur-Oise, were to come to Hugh’s assistance during his war against Robert de Belleˆme.

Their

supportive actions indicate that Hugh’s relationship with his kinsmen and friends in the
Beauvaisis had remained more dynamic in the interval than the silence of the surviving sources
might suggest.

It may be noted that Hugh was by no means the only member of the Norman secular

e´lite to make a marriage with a family from the Beauvaisis at this time. Earl Hugh of
Chester and Gilbert fitz Richard both married daughters of Count Hugh of Clermont-en-
Beauvaisis. Although these matches were probably made about three decades later than
Hugh’s,

it is likely that all were made with the intention of extending the influence of

the dukes of Normandy over neighbouring lands.

There is certainly some evidence to sug-

gest that the process was beginning to work even in William the Bastard’s day. It was, for
example, evidently considered worthwhile to have William sign a diploma of King Philip I
of France concerning the freedom of the church of Saint-Quentin of Beauvais from all out-
side jurisdiction.

Around four years after Hugh’s return to Normandy, Duke William conquered England.

Hugh was present at the battle of Hastings, where he might have had an unsettling moment
if he was that ‘vassal of Grandmesnil’ whose reins broke in the middle of the battle, carrying
him towards the English line until his horse shied away.

He was subsequently made the

first known Norman sheriff of Hampshiredsomething that has not been previously
recognised

which may reveal that the king had been satisfied with his performance at

Neufmarche´. Despite this, Hugh’s interest in England seems to have been weak. In

c. 1068

he abandoned the country, and whatever possessions he might have been given there,

when

his sexually-frustrated wife threatened to take a lover.

It seems from this that his prestigious

marriage and links to the Vexin were more important to Hugh, in

c. 1068, than his position in

Englanddboth in terms of possessions and also his role in the king’s service.

Orderic, vol. 2, 106.

This is dealt with later in this article.

Green,

The aristocracy of Norman England, 353.

Power,

The Norman frontier, 232e3.

Regesta, no. 28.

Orderic, vol. 2, 174; William de Poitiers,

Gesta Guillelmi, ed. and trans. R.H.C. Davis and M. Chibnall (Oxford,

1998), 134; Wace,

Roman de Rou, l. 8437, trans. G.S. Burgess (Woodbridge, 2004), 186.

Orderic, vol. 2, 220. Orderic uses the word ‘

præsidatus’ to describe Hugh’s position over the Gewissae (Hampshire),

which should probably be translated as ‘sheriff’ rather than ‘governor’ (see Orderic, vol. 1, 348). Hugh is omitted from
the list of sheriffs of Hampshire found in J. Green,

English sheriffs to 1154 (Public Record Office Handbooks, no. 24,

London, 1990), 44. It may be supposed that, as sheriff, Hugh worked closely with William fitz Osbern, who had been
given custody of the castle at Winchester by the king so that he could govern the country to the north as regent during
the king’s absences in Normandy (Poitiers,

Gesta Guillelmi, 164 and notes 1e3)

It may be that the houses that Hugh held in Southampton in 1086 had been given to him while he was sheriff

(

Domesday Book: text and translation, ed. J. Morris, 38 vols (Chichester, 1975-86), henceforth GDB if Great Domesday

Book or

LDB if Little Domesday Book with foliation of Great or Little Domesday as appropriate and Morris’s numer-

ation given,

GDB, f. 52a; Hants

x S3).

Orderic, vol. 2, 220.

221

M. Hagger / Journal of Medieval History 32 (2006) 212e230

Hugh was not the only one to abandon England at this time. Although Orderic states that

‘many others departed the country heavy at heart’, the only other man he actually names as
returning home at the same time and for the same reason is Hugh’s brother-in-law, Humfrey
du Tilleul. This is, in fact, the only time that Orderic provides evidence of Hugh de Grand-
mesnil and Humfrey du Tilleul working together, despite the fact that Humfrey had married
Hugh’s sister, Adeliza de Grandmesnil, by about 1040.

The marriage may have been due

to territorial considerations, as le Tilleul (Calvados), now part of the commune of Saint-
Georges-en-Auge, is located about five miles north of Grandmesnil.

Alternatively or

additionally, the marriage may have been brought about as a result of Humfrey’s kinship
to Thurstan Goz, who occupied the office of

vicomte of the Hie´mois during the early

1040s. This relationship was as follows. Humfrey was the son of a certain Ansfrid ‘of Dan-
ish stock’. Thurstan Goz too ‘was the son of the Dane, Ansfrid’. Hugh d’Avranches is
described as the cousin (

consobrinus) of Robert of Rhuddlan, Humfrey’s eldest son. It

may therefore be deduced with some conviction that Humfrey du Tilleul was the brother
of Thurstan Goz, and the uncle of Richard Goz,

vicomte of Avranches, whose son was

Earl Hugh of Chester.

This close kinship with Thurstan, who was a political rival of

the Grandmesnils’ own

beˆte noir, the Montgommery family, could also have made Humfrey

a man worth courting.

c. 1043, Thurstan Goz ‘became inspired by treacherous zeal’ and rebelled against

Duke William. He ‘hired royal soldiers as mercenaries whom he welcomed as his accom-
plices for the defence of the fortress of Falaise, because he did not wish to serve the duke’,
but was defeated by forces under Rodulf de Gace´ and went into exile.

The rebellion

provides a likely

terminus ante quem for the marriage of Adeliza to Humfrey, but there

is a tantalising possibility that Thurstan’s revolt and the marriage were actually linked.
In book VIII of the

Historia ecclesiastica, Orderic has Roger de Beaumont declare to Rob-

ert Curthose that ‘I fought against the rebels in your father’s minority, when Roger of Spain

The date of the marriage is based on what is known of Robert of Rhuddlan’s and Thurstan Goz’s careers. Or-

deric tells us that Robert was knighted in England by King Edward the Confessor. This event must have occurred
by January 1066, and probably after Normans began to attend his court in any number from

c. 1052 (F. Barlow,

Edward the Confessor (London, 1970), 190e2). A squire was generally knighted when he came of age, probably in
his mid to late teens. Raoul de Cambrai, for example, was 15 when the emperor knighted him, and Fulk le Re´chin
was 17 when knighted by Geoffrey Martel, although William Marshal, who was of course a younger son, was
around 20 when he was knighted in 1167 (Bouchard,

Strong of body, 80; J. Bradbury, ‘Fulk le Re´chin and the

origin of the Plantagenets’, in:

Studies in medieval history presented to R. Allen Brown, ed. C. Harper-Bill, C.J.

Holdsworth, J.L. Nelson (Woodbridge, 1989), 29; D. Crouch,

William Marshal: court, career and chivalry in the

Angevin Empire (Harlow, 1990), 16, 30). So, working backwards, Robert was probably born no later than c.
1048. As Thurstan Goz rebelled in

c. 1043 and was driven into exile (Jumie`ges, vol. 2, 102), it seems unlikely

that the marriage with Humfrey took place after his disgrace. Indeed, there is further reason for believing the mar-
riage was made before Thurstan’s exile, as discussed below.

L.C. Loyd,

The origins of some Anglo-Norman families (Harleian Society Publications, 103, 1951), 85. There is no

source outside Orderic to provide corroboration that Humfrey really used this toponymic.

Orderic, vol. 4, 136; Jumie`ges, vol. 2, 100. For the Goz family see L. Musset, ‘Les origines et le patrimoine de

l’abbaye de Saint-Sever’,

La Normandie Be´ne´dictine au temps de Guillaume le Conque´rant (X

sie`cle) (Lille, 1967),

358e60 and 365e5; L. Musset, ‘Une famille vicomtale: les Goz’,

Documents de l’histoire de Normandie (Toulouse,

1972), 94e8.

Jumie`ges, vol. 2, 102.

222

M. Hagger / Journal of Medieval History 32 (2006) 212e230

(de Tosny)

and his sons Elbert and Elinand and many others perished’.

Orderic does not

mention Robert I de Grandmesnil by name here, but he does say that he died in this same
battledalbeit without any mention of a rebelliondin book III of the

Historia.

As this

engagement has been dated to around 1040, it may be that it and Thurstan Goz’s rebellion
were part of the same insurrection, and that the marriage between Humfrey and Adeliza had
been intended to bind the conspirators more closely together.

If the rebellions in the 1040s were the

raison d’eˆtre of the marriage, then the death of Robert

I de Grandmesnil and exile of Thurstan Goz would be sufficient to explain the apparent lack of
any relationship between Hugh and Humfrey in the years following. Although Humfrey may
have celebrated his links with the Grandmesnil family in the names chosen for his two eldest
sons, Robert and Arnold, both of which were names associated with the Grandmesnil and
Giroie families,

Humfrey’s eldest son, Robert of Rhuddlan, apparently did not gain any ben-

efit from his kinship to Hugh de Grandmesnil. The pattern of his patronage to Saint-Evroult,
where he gave the abbey possessions that he himself held from Earl Hugh of Chester, indicates
that he held no land of his uncle, but had been forced to rely instead on the patronage of his
father’s family to make his way in England and Wales. Although there are some coincidences
in the men’s careers that might be interpreted as hinting at a continuing relationshipdthey both
held manors in Byfield (Northamptonshire),

witnessed King William I’s act for Saint-Evroult

together,

and both rebelled in favour of Robert Curthose in 1088

such a conclusion would

probably be misplaced. Robert of Rhuddlan’s activity in the rebellion of 1088, for example,
seems to have been limited to his being besieged at Rochester (Kent), while Hugh de Grand-
mesnil was active in Leicestershire, and Robert’s patronage and ‘love’ of Saint-Evroult is
attributed by Orderic to the fact that it was the place ‘where his brothers Arnold and Roger
were monks, and his father and mother and kinsfolk lay buried’, rather than to any connection
to the Grandmesnil family.

Even if the Tilleul family were not close to the Grandmesnil, they and others were very

much aware of the kinship between them. Orderic Vitalis tellingly locates Robert of Rhuddlan

Roger de Tosny was called ‘of Spain’ because of his part in the reconquista of Spain from the Moors (L. Musset,

‘Aux origines d’une classe dirigeante: Les Tosny, grands barons Normands du X

au XIII

sie`cle’,

Francia, 5 (1977),

52e3). He should not be confused with members of the Epaignes family, whose toponymic was frequently rendered ‘

Hispania’ in contemporary sources (Loyd, Anglo-Norman families, 51e2).

Orderic, vol. 4, 206.

Orderic, vol. 2, 40. Robert I de Grandmesnil’s death in this battle is also mentioned in one of Orderic’s interpolations

into the

Gesta Normannorum ducum of William de Jumie`ges. Orderic does not make it clear on whose side Robert I

fought (Jumie`ges, vol. 2, 96).

This bears comparison with the events of 1075, when the rebellion against King William I of that year was hatched

at the marriage of Roger de Breteuil’s daughter to Ralph de Gael, both of them among the rebels (D.C. Douglas,

William

the Conqueror (London, 1964), 231e2). Count Waleran of Meulan may also have married three of his sisters to ‘leading
castellans’ to cement more firmly his alliances with men who were to rebel with him against King Henry I in 1123e4
(Orderic, vol. 6, 332).

For what Christian names can tell us about family relationships see, for example, Green,

The aristocracy of Norman

England, 345e7; Bouchard, Strong of body, 73.

GDB, f. 224ced (Northants

xx 22. 1, 23. 12).

Regesta, no. 255.

Orderic, vol. 4, 124; William of Malmesbury,

Gesta regum Anglorum, ed. and trans. R.A.B. Mynors, R.M. Thomson

and M. Winterbotton (Oxford, 1998), vol. 1, 546;

The Anglo-Saxon chronicle, trans. D. Whitelock (London, 1961), 167

(‘E’ Chronicle, s.a. 1088).

Orderic, vol. 4, 136.

223

M. Hagger / Journal of Medieval History 32 (2006) 212e230

and his monkish brother, Arnold, in Norman society by referring to each of them as ‘the
nephew of Hugh de Grandmesnil’.

Arnold himself seems to have considered himself a part

of the Grandmesnil family, and must also have thought that his kinsmen would have a reciprocal
view of the relationship. Such is clear from his exploitation of that kinship, in the name of
Saint-Evroult, when ‘he travelled as far as Apulia and Calabria and Sicily to ask for support
for his church from the loot acquired by his kinsmen. He visited his brother, William, abbot
of Sant’Eufemia, and his cousin William de Grandmesnil and other wealthy kinsfolk in Italy,
and by gentle coercion secured from them all that he could in order to bestow it on his
monastery’.

In this way at least, the Tilleul family’s maternal kin admitted their relationship.

Although Humfrey du Tilleul never recovered his position in England, fortune smiled once

more on Hugh de Grandmesnil in the years following his precipitate return to Normandy. Prob-
ably by 1081 and certainly by 1086, Hugh held over a hundred houses in Leicester, 74 manors
in Leicestershire, 17 manors in Warwickshire, and 20 manors in Northamptonshire in chief,
another five manors in Leicestershire from Countess Judith, and a number of more dispersed
estates elsewhere. These possessions were worth, in total, £368 at 1086 values, making him
a Class C baron by Corbett’s reckoning.

Hugh could only have obtained these lands if he had the king’s favour and trust, although

this need not have been as the result of loyal service over a number of years; indeed, that
Hugh attests only seven of King William’s authentic

acta suggests that he was not a regular

attender of the court.

It seems more likely that these lands in Leicestershire and elsewhere

were instead given to Hugh as a result of his support of King William during the rebellion
of his son, Robert Curthose, just as men such as William d’Aubigne´ Brito were later to benefit
from supporting King Henry I at Tinchebray (Orne) in 1106.

Details are lacking, but we do

know from Orderic that Hugh was among the magnates who attempted to make peace between
King William I and Robert Curthose in Normandy in

c. 1079,

even if we cannot place him at

key events such as the skirmish at Gerberoi in January 1079.

Hugh, then, probably obtained these lands as the result of his support for the king in Nor-

mandy in his time of need, rather than because immersion in the royal service was a defining
characteristic of his ‘identity’. This is also suggested by Hugh’s failure to support William the
Conqueror’s designated heir to his English lands, William Rufus. Instead, Hugh de Grandmes-
nil rebelled against King William II in favour of Robert Curthose in 1088.

Orderic barely

mentions his involvement, but the Anglo-Saxon Chronicle records that ‘there was also one Hugh

Orderic, vol. 3, 118, 226, 240.

Orderic, vol. 4, 142. The act by which Guitmund de Maisnil and Robert his son gave the tithe of their ploughs to

Saint-Evroult is dated by Arnold’s return from Apulia (le Pre´vost, 185, no. vi).

Orderic, vol. 2, 264. Orderic is not specific about the date of Hugh’s appointment, and while some of those that he

mentions in this passage, such as Henry de Ferrers’ acquisition of the honour of Tutbury, can be dated to

c. 1070, others,

such as the grant of the earldom of Surrey to William de Warenne, were not made until much later (in that case 1088). For
Hugh’s Domesday holdings see

GDB, f. 52a (Hants

x S3), 138ced (Herts x 26), 169b (Glos x 62), 174a (Worcs x 2. 71),

224c (Northants

x 23), 232ae233b (Leics x 13), 242a (Warks x 18), 291c (Notts x 23), 360c (Lincs x 30. 22); LDB, f. 432a

(Suff

x 49); W.J. Corbett, ‘The development of the duchy of Normandy and the Norman Conquest of England’ in: The

Cambridge medieval history vol. 5, ed. J R. Tanner, C.W. Previte´-Orton, and Z.N. Brooke (Cambridge, 1943), 510.

Regesta, nos 39, 52, 53, 54, 144, 175, 255.

For William d’Aubigne´ Brito’s career see J. Green, ‘The descent of Belvoir’,

Prosopon, 10 (April 1999) at

http://www.linacre.ox.ac.uk/research/prosop/Prspn10.doc

Orderic, vol. 3, 110.

Orderic, vol. 4, 124.

224

M. Hagger / Journal of Medieval History 32 (2006) 212e230

(a toponymic was not required to make it clear who this was) who did not mend matters at all,
neither in Leicestershire nor in Northampton’.

Hugh’s decision to rebel was probably the result

of the fact that he held estates in both England and Normandy, and, like Bishop Odo of Bayeux,
saw the difficulties inherent in serving two masters.

That he chose Curthose over Rufus could

well have been for the reasons put forward by Bishop Ododthat ‘he is older by birth, and of
a more tractable character, and we have already sworn fealty to him during the lifetime of the father
of both men’

but it is likely that there was at least one further reason. Despite his extensive ac-

quisitions in England, Hugh de Grandmesnil was still a Norman in his thinking, in his focus, and in
his identity. As such, his support of the duke of the Normans was virtually a foregone conclusion.

This is revealed first by his religious patronage, which was directed exclusively at Norman

houses. There were 29 new monasteries (counting only the independent Benedictine houses,
Cluniac houses, and alien priories deliberately founded by laymen) established in England in
the years between 1066 and 1100 by men of various rank, from earls such as Roger of Shrews-
bury (Shrewsbury and Much Wenlock) and Robert de Mowbray (Tynemouth), to Eudo dapifer
(Colchester), Gilbert de Gant (Bardney), Peter de Valognes (Binham, begun before 1087),
William de Percy (Whitby), and Robert de Busci (Blyth).

These latter were men of similar

standing to Hugh de Grandmesnil, so that his failure to found a new monastery on his English
estatesdthe priory at Ware (Herts) was established by the monks of Saint-Evroult rather than by
Hugh himself, and probably only after his death

may well tell us something of his sense of

identity. Hugh’s patronage was instead directed for the most part at his own Norman abbey of
Saint-Evroult, which received a lavish endowment in the English midlands.

He also made

small donations to the ducal abbeys at Montivilliers and Saint-E

´ tienne, Caen, and to Le Bec.

That Hugh’s thinking and ‘identity’ had a strong Norman bias is also suggested by the mar-

riages that Hugh chose for his children. These are all listed by Orderic, so that we know that three
of Hugh’s sons and four of his daughters found spouses. Only in the case of William de Grand-
mesnil’s marriage did Hugh not have control over the match that was made, so that in the remain-
ing six instances it may be supposed that the choices made for his children reflected Hugh’s
thinking and priorities. If that is true, then it is notable that of the four marriages made by his
daughters, three of them were with men with only, or primarily, Norman interests: Robert de
Courcy and Hugh de Montpinc¸on, and William de Sai. As the Courcy and Montpinc¸on families
were neighbours of the Grandmesnil, it may be supposed that the marriages were made with the
intention of strengthening the family’s security and influence in the Pays d’Auge, although Hugh
may also have been intending to improve his contacts at the king’s court; Ralph de Montpinc¸on
was one of King William’s stewards, and Richard de Courcy may have had good connections
there too as his son, and Rohais de Grandmesnil’s husband, Robert, was to serve as a steward

The Anglo-Saxon chronicle, trans. Whitelock, 167 (‘E’ Chronicle, s.a. 1088).

Orderic, vol. 4, 122.

Orderic, vol. 4, 122e4.

D. Knowles and R.N. Hadcock,

Medieval religious houses: England and Wales (London, 1953), 58e101; A. Binns,

Dedications of monastic houses in England and Wales, 1066e1216 (Studies in the history of Medieval Religion, 1,
Woodbridge, 1989), 61e117.

There is no evidence that Hugh intended to found a priory at Ware in King William I’s confirmation of his donations

in England to the abbey of Saint-Evroult (

Regesta, no. 255), and very slim evidence that monks had been installed at

Ware before Hugh’s death (Orderic, vol. 4, 336, note 3).

Regesta, no. 255.

Regesta, nos 212, 50, and 166, respectively.

225

M. Hagger / Journal of Medieval History 32 (2006) 212e230

to Henry I.

The eldest of Hugh’s daughters, Adelina, married Roger d’Ivry, but the reasons for

this marriage are difficult to divine. He is a man who seems to have had mainly English interests,
although too little is known of his situation in Normandy to be sure of this, and a warning is
sounded both by his foundation of the priory at Ivry in 1071, and his custody of the castle at
Rouen in

c. 1077, when the young Robert Curthose attempted to seize it.

As would be expected, Hugh seems to have less freedom when choosing wives for his sons.

Robert III de Grandmesnil, who was to succeed to Hugh’s Norman estates, married Agnes, the
daughter of Ranulf de Briquessart,

vicomte of the Bessindanother man withdso far as we knowd

exclusively Norman interests. In contrast, Ivo, the heir to Hugh’s English estates, married
a daughter of Gilbert de Gant, who, as a Fleming, had no interests in Normandy whatsoever,
but was a significant tenant-in-chief in the north of England and in the midlands. If England
could not be ignored altogether, these marriages seem to reveal a clear disparity in the relative
importance of England and Normandy in Hugh’s thinking.

It is only in the case of the Courcy family that the new kin-by-marriage can be seen to have

worked with the Grandmesnil family to their mutual benefit. Orderic introduces the marriage
between Rohais de Grandmesnil and Robert de Courcy in the context of Robert de Belleˆme’s
nefarious activities in Normandy in 1091. While unidentified Norman magnates discussed how
best to resist him, ‘Hugh de Grandmesnil and Richard de Courcy (Robert’s father), because they
were nearest to the tyrant’s boundaries and most threatened by his evil ambitions, took to arms,
provisioning their castles with arms and food and reinforcing their garrisons. Although these
men were both grey-haired, they were distinguished by courage and magnanimity and throve
in a necessary alliance against imminent danger’.

Orderic emphasizes the role of the family

here, continuing: ‘The good Hugh, therefore, relying on the support of his son, sons-in-law, and
many friends, plunged vigorously into a war against Robert, and thanks to the courage of his
distinguished allies put up an impressive resistance to Robert’s tyranny’.

It may be supposed that Robert de Courcy was among the sons-in-law referred to. He was also to

act with Robert III de Grandmesnil (and Hugh de Montpinc¸on) in 1102 and 1118, when their joint
interests were again threatened.

That there was always this element of self-interest, however,

means that it is difficult to see if the families worked together as a result of their kinship-by-marriage,
or simply through expediency. In any event, there is no indication that the marriages had any impact
on the identities of any of the families involved, insofar as this can be gauged through naming pat-
terns and the patronage of religious houses. In addition, Orderic specifically mentions Count Mat-
thew of Beaumont and Theobald, the son of Waleran de Breteuil in the Beauvaisis who had spoken
up for Hugh in

c. 1063, as taking part in these encounters,

revealing that Hugh was still able to draw

on the support of his wife’s relations and friends some 30 years after they had first come to his aid.

Orderic, vol. 3, 164, although it may be noted that Ralph does not attest any of King William’s surviving acta with

that title. Robert de Courcy attests as dapifer in

Reading Abbey cartularies, ed. B. Kemp (Camden Series, fourth series,

31, 1986), vol. 1, 41 no. 7.

Orderic, vol. 2, 358; Bauduin,

La premie`re Normandie, 212e3.

Orderic, vol. 4, 230.

Orderic, vol. 6, 24, 216 (I have supposed that Orderic’s reference to the ‘men’ of Grandmesnil, Courcy, and Mont-

pinc¸on was a tactful way of avoiding naming Robert III de Grandmesnil, Robert de Courcy and Hugh de Montpinc¸on in
relation to such treacherous designs).

Orderic, vol. 4, 232. Orderic notes that Theobald de Breteuil was killed during one of these engagements.

Robert de Belleˆme might have timed his attacks badly here. Hugh’s wife, Adeliza, died on 1 June 1091 at Rouen

(Orderic, vol. 4, 338), and it may be that Hugh would not have obtained this support from the Beauvaisis had she already
expired before Belleˆme attacked.

226

M. Hagger / Journal of Medieval History 32 (2006) 212e230

The son who assisted Hugh was, perhaps unexpectedly, Ivodthe heir to Hugh’s English

landsdrather than Robert III who was to succeed him in Normandy. Orderic tells us that during
the siege of Courcy, Ivo was captured. This provides the only piece of evidence for any of
Hugh’s sons supporting their father. In contrast, there are two occasions where Orderic shows
members of the family displaying a complete lack of ability to work together for their family’s
common good. William de Grandmesnil was Hugh’s second son, to judge from the list
produced by Orderic.

He also seems to have made a favourable impression at the court of

William the Conqueror. Indeed, ‘the king had so great an affection for him that he offered
him the hand of his niece, the daughter of Count Robert of Mortain, hoping in this way to
bind the youth to him by the signal honour of kinship. But the proud young knight thought noth-
ing of the king’s proposal and for frivolous reasons went to Apulia with Robert Giffard and
many others’.

One can only imagine what Hugh de Grandmesnil thought of this decision,

and what others thought of his inability to control his son.

That William made an even greater

marriage in Italy was probably beside the point.

Similarly, after Hugh’s death Robert III and Ivo de Grandmesnil, heirs to Hugh de Grand-

mesnil’s Norman and English estates respectively, took different sides in the civil war that
marred the first six years of King Henry I’s reign. Robert supported King Henry I, as is evi-
denced by his fighting on the king’s side at Tinchebray in 1106,

while Ivo supported Duke

Robert. Ivo may have made this decision, at least in part, because he had been a friend and
companion of Robert Curthose from the 1070s.

He may also have been influenced by the

fact that, following his father’s death and the division of the Grandmesnil inheritance, he
was a man with wholly English interests, which enabled him to become one of those ‘turbulent
magnates, alarmed by the energy of King Henry and preferring the mildness of the sluggard
duke, Robert, which left them more free to pursue their evil ambitions’ and who consequently
rebelled against King Henry I in

c. 1102.

The result was catastrophic. Ivo was harassed by the

king so much that he placed himself under the protection of the count of Meulan, mortgaged his
lands to him, and left on pilgrimage where he died.

Ivo’s actions and sympathies in England, then, were completely opposed to those of Robert

III in Normandy. Ivo was, it seems, acting out of a longstanding friendship with Curthose and
with a view to his own self-interestdCurthose would rule with a lighter hand than King Henry.
His family’s longstanding enmity towards Robert de Belleˆme was not allowed to stand in the
way of that decision, although as Ivo had been out of Norman politics since 1095, and besides
was not a neighbour of Belleˆme’s in England, he may not have been especially troubled by this
consideration. Whatever the reasons for his support of Curthose, however, it is evident that

Orderic, vol. 4, 230, 338.

Orderic, vol. 4, 338.

Robert Bartlett makes the point vividly when talking of Christina of Markyate’s refusal to marry in accordance with

her parents’ wishes: ‘For a girl to deny the authority of her parents and her husband was a rebellion that must be
crushed, lest the family’s credibility founder in this competitive world of wagging and malicious tongues’ (R. Bartlett,
England under the Norman and Angevin kings, 1075e1225 (The new Oxford history of England, Oxford, 2000), 551).

Orderic, vol. 4, 84.

It was Ivo and his brother Aubrey who provoked the fight at the house of Roger Cauchois in L’Aigle in 1077, which

led to Curthose’s impetuous attempt to take Rouen (Orderic, vol. 2, 398; Orderic, vol. 4, 230, 234). He and his brother
also went on crusade with Curthose in 1095, where they earned themselves an unenviable reputation by escaping from
the siege of Antioch down ropes (Orderic, vol. 5, 34, vol. 6, 18).

Orderic, vol. 5, 306e8.

Orderic, vol. 6, 18.

227

M. Hagger / Journal of Medieval History 32 (2006) 212e230

family solidarity played no role whatsoever. Robert III and Ivo took different roads, Robert be-
cause he had suffered at the hands of Curthose’s lax rule, and Ivo because of friendship and
a belief that things would be better under the relaxed rule of the duke. Hugh de Grandmesnil
had divided his estates on his death, and with it he had divided his family.

Hugh de Grandmesnil died seven years after Robert de Belleˆme’s attack on Courcy castle, on

22 February 1098, at his manor of Ware in Hertfordshire. He would have been in his seventies.
Six days previously, he had been received as a monk of Saint-Evroult by the prior of that house,
who had been sent across the Channel to tend to Hugh’s spiritual welfaredsomething that sug-
gests that Hugh had been seriously ill for some time previously. It may be supposed that Hugh
had commanded that his body should be taken back to Normandy, it would certainly be
expected of a man who saw himself first and last as a Norman, and so it was ‘preserved
with salt and sewn securely into an ox-hide’ and taken to Normandy. He was buried in the chap-
ter house at Saint-Evroult, ‘on the south side, beside Abbot Mainer’, and Orderic himself
furnished one of his better epitaphs to commemorate the abbey’s founder.

Conclusion

We can see Hugh de Grandmesnil in a variety of guises. He was a Norman, the lord of

Grandmesnil, the custodian of a border town, a sheriff, and the holder of a large honour in
the English midlands. He was of the lineage of Grandmesnil, a member of the house of Giroie,
the son-in-law of the count of Beaumont-sur-Oise, the father-in-law of members of the Courcy,
Montpinc¸on, and other families. It is not always possible to judge which of these ‘identities’
were most important to Hugh, although it does seem that his patrimony, and Normandy
more generally, remained central to his thinking throughout his careerdso much seems clear
from his departure from England in

c. 1068, his part in the rebellion in 1088, the pattern of

his monastic patronage, and the marriages chosen for his daughters.

Hugh’s relationships with his kinsmen certainly varied throughout his career. In his youth, he

was dominated by his uncle, William fitz Giroie and by other members of the Giroie family.
Indeed, such is the importance of that relationship that Hugh’s part in the attack on Se´es in

1049 and his exile in

c. 1061 cannot be properly understood without being aware of who Hugh’s

kinsmen were. That exile also reveals how kinship had both upsides and downsides. Hugh and his
brother, Robert II, benefited from the support provided by their kinsmen, for some years forming,
to use J. C. Holt’s phrase, a ‘mutual benefit society’,

but they also suffered in the aftermath of

Arnold d’E

´ chauffour’s rebellion. Similar benefits came Hugh’s way as a result of his marriage

to Adeliza de Beaumont-sur-Oise, with his kinsmen and friends from the Beauvaisis coming to
his assistance on at least two occasions, albeit that these were separated by some 30 years. In
contrast, Hugh seems to have gained much less from the marriages he made for his children.
This may, in part, be the result of Hugh’s own personal relationship with each of these kin-groups.
The strongest link was that with the family to which Hugh was related by blood through his
mother, the next strongest was that with his wife’s family, while the weakest connections were
those with the families to which he married his own sons and daughters. This cannot be the whole
story, however, as Hugh’s relationship with the Tilleul family was no stronger than that with the
house of Courcy, despite the fact that Robert of Rhuddlan was his blood relation. No doubt friend-
ships, disagreeable personalities, and the like also played their part for good or ill in making or

Holt, ‘Feudal society and the family in early medieval England: III. Patronage and politics’, 8.

228

M. Hagger / Journal of Medieval History 32 (2006) 212e230

breaking these associations, so that friends such as the Waleran and Theobald de Breteuil in the
Beauvaisis could play a larger role in Hugh’s career than his own nephews.

It should also be noted that Hugh was able to rely on his own talents. His return to favour,

following his precipitate return to Normandy in

c. 1068, seems to have been unassisted by his

relations and was probably due to his own political acumen, aided no doubt by the king’s need
for experienced men in the years following the Conquest. Familial clout seems to have counted
for little, therefore, in the most important period in the rise of Hugh de Grandmesnil. Indeed,
the fact that Duke William was able to crush the Giroie and send Hugh into exile in the early
1060s is a useful reminder that even strong and well-established kin groups could not stand against
a determined and well-supported ruler. Hugh’s career, then, provides an example of the extent to
which ducal displeasure or favour might affect a man’s career; as well as of how changing circum-
stances might lead the duke to alter his behaviour towards particular individuals.

Orderic seems to have emphasized Hugh’s family relationships whenever possible, so that

a failure to portray Hugh’s kinsmen rallying around him is likely to mean that they failed to
do so. Hugh’s career, then, also suggests that such concerted action was the exception rather
than the norm. One may also question the longevity of any supportive, or even active, relation-
ship resulting from the marriages made by the Grandmesnil. Of the ‘alliances’ made, the
evidence suggests that very few lasted for more than a generation. The relationship with the
Giroie, for example, which was close until the exile of

c. 1061, seems to have fizzled out after

c. 1063, when Arnold d’E

´ chauffour was killed and his heirs dispossessed, even though Robert

III and Robert de Saint-Ce´ne´ri together asked the king to grant a confirmation charter to Saint-
Evroult in 1128. Hugh de Grandmesnil’s relationship with the counts of Beaumont-sur-Oise and
with men from the Beauvaisis is evidenced by his being given their assistance in

c. 1063 and in

1091, but there is no evidence that it remained vital after Adeliza’s death in that latter year.
Robert III de Grandmesnil contemplated rebellion with Robert de Courcy and Hugh de Mont-
pinc¸on in 1119,

but there is nothing to suggest that this was not the last gasp, still 17 years

before the death of Robert III, of a relationship that had been established in around 1091. This
fluidity is not especially surprising; all of us have encountered such shifting friendships and
alliances in our own lives. What is surprising is that historians have tended to assume that
kinship and marriages resulted in bonds that lasted for life, if not longer.

It seems extremely

unlikely that this was really the case, making it dangerous to assume that the relationships
forged among members of the secular e´lite of eleventh-century Normandy remained immuta-
ble. The evolution of Hugh’s identity, and the apparent transience of even the strongest of
connections with his kinsmen, provides an example of why we should ensure that we are
not trapped by such stereotypes.

Orderic, vol. 6, 216.

This is at least the impression given by, for example, Peter Fleming, who has said that ‘for the propertied classes,

marriage was the prime means of advancing the interests of the family. A successful match could provide an alliance
with a family which had influence with the king or the nobility, standing and power in the locality, social status, money
and lands’ (P. Fleming,

Family and household in medieval England (Basingstoke, 2001), 31), and by Amy Livingstone

who has opined that ‘marriage between the two families (those of Fre´teval and the vicecomital family of Chaˆteaudun)
would prevent hostilities and provide the viscount with an important ally to the south, the area most vulnerable to the
aggressions of the count of Anjou. Fulcher and Hildeburg’s other two daughters were married into local families of
lords, those of La Ferte´ and Frouville. . All those marriages extended the family’s network of alliances and provided
loyal supporters close to their base of power at Fre´teval’ (A. Livingstone, ‘Kith and kin: kinship and family structure of
the nobility of eleventh- and twelfth-century Blois-Chartres’,

French Historical Studies, 20 (1997), 426). A brief but

useful argument against the permanency of marriage alliances is found in Power,

The Norman frontier, 225e6.

229

M. Hagger / Journal of Medieval History 32 (2006) 212e230

Acknowledgments

The author would like to thank Nicholas Karn and, in particular, Kathleen Thompson for

their comments on earlier drafts of this paper. An abridged version was read on 12 July
2005 at the International Medieval Congress at Leeds.

Mark Hagger

graduated with an M.A. Hons in Mediaeval History from the University of St Andrews in June 1994,

remaining at St Andrews for his doctorate under the supervision of Professor Robert Bartlett. His thesis on the Verdun
family was published as

The fortunes of a Norman family: the de Verduns in England, Ireland and Wales 1066e1316 by

Four Courts Press, Dublin, in April 2001. Between 1998 and 2003 he was a solicitor with the City of London Law firm,
Freshfields Bruckhaus Deringer. Since October 2003, he has been working on a new edition of the charters and writs of
King Henry I (1100e1135); a project run by Professor Richard Sharpe in the History Faculty at Oxford. He has held the
post of Junior William Golding Research Fellow at Brasenose College, Oxford, since October 2004.

230

M. Hagger / Journal of Medieval History 32 (2006) 212e230

Document Outline

Kinship and identity in eleventh-century Normandy: the case of Hugh de Grandmesnil, c. 1040-1098

kinship and identity in normandy

Document Outline