Alan V.

Murray

The army of Godfrey of Bouillon, 1096-1099 : Structure and

dynamics of a contingent on the First Crusade
In: Revue belge de philologie et d'histoire. Tome 70 fasc. 2, 1992. Histoire mdivale, moderne et contemporaine
Middeleeuwse, moderne en hedendaagse geschiedenis. pp. 301-329.

Citer ce document / Cite this document :

Murray Alan V. The army of Godfrey of Bouillon, 1096-1099 : Structure and dynamics of a contingent on the First Crusade. In:
Revue belge de philologie et d'histoire. Tome 70 fasc. 2, 1992. Histoire mdivale, moderne et contemporaine
Middeleeuwse, moderne en hedendaagse geschiedenis. pp. 301-329.
doi : 10.3406/rbph.1992.3824
http://www.persee.fr/web/revues/home/prescript/article/rbph_0035-0818_1992_num_70_2_3824

ARTICLES ET MLANGES
ARTIKELEN EN MENGELINGEN

The army of Godfrey of Bouillon, 1096-1099 :

Structure and dynamics of a contingent
on the First Crusade

Alan V. Murray

The army led in the First Crusade by Godfrey of Bouillon, duke of Lower
Lotharingia, set off on its journey to the Holy Land about the middle of
August 1096. It marched up the Rhine, down the Danube and through the
Balkans, arriving at Constantinople on 23 December (}). Only at this point
did the army encounter other groups which had travelled through Illyria or
over the Adriatic from Italy, as well as those crusaders traditionally, if
somewhat inaccurately referred to as the 'People's Crusade', who had arrived
the previous summer and remained in Asia Minor since their defeat at the
hands of the Turks at Nicaea on 21 October. From this point these diverse
groups constituted a united Frankish army, but nevertheless each of the
original contingents, usually described as exercitus by the writers who wrote
about the crusade, clearly retained its separate identity within it, and continued
to function as the basic military unit in battle and on the march at least
until the capture of Jerusalem in the summer of 1099.
Historians have traditionally stressed the composition of these contingents
as an important factor in determining the character of the nobilities of the
(1) Albert of Aachen, Historia Hierosolymitana, Recueil des Historiens des Croi
sades, Historiens occidentaux, IV (Paris, 1879) [henceforth cited as AA], 299, 305306 ; Heinrich Hagenmeyer, 'Chronologie de la premire croisade (1094-1 100)', Revue
de l'Orient latin, 6 (1898), 214-293, 490-549, 7 (1899), 275-339, 430-503, 8 (1890-1),
318-382 [items numbered consecutively throughout], nos. 67, 107.

302

. V. MURRAY

four crusader states subsequently established in Syria and Palestine. However

the composition of only two contingents has been discussed in any detail,
those of the Normans of Normandy, led by their duke, Robert, and the
Normans of Southern Italy, led by Bohemund and his nephew Tancred (2).
The army led by Godfrey has by contrast been surprisingly neglected
considering how it is generally assumed to have contributed to the nobility
of the Latin kingdom of Jerusalem (3). Leaving aside the vexed question of
numbers, the aim of this essay is to examine the composition of this army,
in the first instance by identifying as many of its participants as possible,
and discussing their relationship to Godfrey and to one another. It also
examines those political factors which may have influenced participation in
the crusade, and conversely, non-participation, and goes on to discuss how
the composition and structure of this army developed in the course of its
march from Western Europe to Palestine.
Godfrey was accompanied by his younger brother Baldwin and Baldwin's
wife Godechilde (or Godevere), daughter of the Norman lord Ralph of
Tosny (4). Besides this immediate family group, a central element in Godfrey's
exercitus comprised knights who had been in his service prior to the summer
of 1096. These included the household officers Ruthard the butler (5), Baldric

(2) C. W. David, Robert Curt hose, Duke of Normandy, Harvard Historical Studies,
25 (Cambridge, Mass., 1920), Appendix D, 'Robert's Companions on the Crusade',
pp. 221-229 ; . . Jamison, 'Some Notes on the Anonymi Gesta Francorum, with
Special Reference to the Norman Contingent from South Italy and Sicily in the First
Crusade', in Studies in French Language and Medieval Literature Presented to
Professor Mildred K. Pope, Publications of the University of Manchester, 268
(Manchester, 1939), pp. 195-204.
(3) Joshua Prawer, 'La noblesse et le rgime fodal du royaume latin de Jrusalem',
Le Moyen ge, 65 (1959), p. 42, cites two studies in support of his claim that the
early Jerusalem nobility was predominantly Lotharingian in character. The first, John
C. Andressohn, The Life and Ancestry of Godfrey of Bouillon, Indiana University
Publications, Social Sciences Series, 5 (Bloomington, Illinois, 1947), does give a detailed
account of the march, but primarily from a biographical perspective. Andressohn's
discussion of the participants other than Godfrey and Baldwin is limited to the naming
of twelve crusaders, most of whom are identified either wrongly, as I hope to show,
or not at all. The second, Charles Moei.ler, 'Les Flamands du Ternois au royaume
latin de Jrusalem', in Mlanges Paul Fredericq (Bruxelles, 1903), pp. 189-202, is not
a study of Godfrey's army, but an analysis of the Versus de viris illustrious dioecesis
Tarvanensis qui in sacra fuere expeditione, a short but valuable source dealing only
with crusaders from the Flemish diocese of Throuanne. See also the discussion of
research on Godfrey and his crusading army in Georges Desi'y, 'Godefroid de Bouillon,
mythes et ralits', Acadmie Royale de Belgique, Bulletin de la Classe des Lettres
et des Sciences Morales et Politiques, 5e srie, 71 (1985). 249-275.
(4) AA p. 358.
(5) AAp.481.

THE ARMY OF GODFREY OF BOUILLON

303

the seneschal (6) and Stabelo the chamberlain (7). Others can be identified by
their surnames as Bullonienses, that is, vassals holding fiefs in Godfrey's
allodial territory of Bouillon in the Ardennes. The most prominent of these
were Heribrand, castellan of Bouillon, and a relative of his, Walter of
Bouillon (8). We also know of a Folbertus de castello Bullon and a Philippus
de Bulon either on crusade or from so early a time in Palestine that they
must have been on crusade (9). However we cannot simply assume that
Godfrey's household accompanied him in toto to the East. We will discover
that numerous knights entered his service in the course of the crusade, and
therefore cannot be certain whether the chamberlains Adelolf and Godfrey
and the seneschal Matthew had previously been vassals of Godfrey or whether
they had come to him from other contingents (10).
The army also included various lords from those parts of Lower Lotharingia
which surrounded Godfrey's lands. Warner, count of Grez, later became the
most prominent of Godfrey's vassals in Jerusalem. He was a kinsman of
Godfrey and Baldwin and is mentioned immediately after them in the list
of departing crusaders given by the chronicler Albert of Aachen (n). In 1095
he sold the allod of Vaux (Belgium, Lige, arr. Huy) quod in divisione patrimonii cum fratre suo Henrico in partent suscepit to the church of Fosses
in order to raise funds to go to Jerusalem with duke Godfrey (l2). He and
his brother Henry, also described as count of Grez, were present with other
crusaders at the sale of the allods of Baisy and Genappe by Godfrey (I3).
Warner's name appears immediately after that of Rainer, the advocate of
Lige, among the laid nobiles who witnessed the purchase of the castle of
Couvin by bishop Otbert of Lige on 14 .June 1096, suggesting that he was
a vassal of the ecclesiastical principality. The small county of Grez lay sandCo) A A pp. 300,412,481.
(7) AA pp. 300, 481-482, 593. His name seems to be the basis of the non-existent
and erroneous 'Baldwin of Stavelot' mentioned by Andressohn (p. 53) and Steven
Runciman, History of the Crusades (Cambridge, 1951-5), I, 147.
(8) AA pp.317, 440; La chronique de Saint- Hubert dite Cantatorium, ed. Karl
Hanquet (Bruxelles, 1906) [henceforth cited as Cantatorium], pp. 38, 53, 64-65 ; Alan
V. Murray, 'The Origins of the Frankish Nobility of the Kingdom of Jerusalem,
1100-1 118', Mediterranean Historical Review, 4 (1989), 285.
(9) AA pp. 435-436, 593.
(10) Adelolf : AA p. 481. Geoffrey : AA p. 526. Matthew : AA p. 522.
(11) AAp.299.
(12) 'Documents extraits du caitulaire du chapitre de Fosses', Analectes pour servir
l'histoire ecclsiastique de la Belgique, 4 (1867), no. 1, pp. 369-398. There is thus
no justification for Runciman's identification, 'the Burgundian count, Warner of Gray'
(History of the Crusades, I, 313).
(13) Heinrici IV. Diplomata. Die Urkunden Heinrichs IV, ed. Dietrich von Gladiss
and Alfred Gawlik, MGH Diplomata Regum et Imperatorum Germaniae (Berlin,
Weimar, Hannover, 1941-78) [henceforth cited as MGH DD Heinrich IV], no. 459.

304

. V. MURRAY

wiched between the county of Leuven and the territory of Lige proper and
probably recognised the suzerainty of the prince-bishop ; possibly Warner's
share of the inheritance lay within Lige territory, as was undoubtedly the
case with Vaux, his hereditatis portio in the episcopal county of Huy (I4).
The crusader Henricus de Ascha, who was accompanied on crusade by
his brother Godfrey, is described asfilius Fredelonis, unus de collateralibus
duds Godefridi (15). Their father can be identified with a certain Frithelo,
described by monkish chroniclers at the end of the eleventh century as both
advocate and despoiler of the abbeys of Malmedy and Echternach. Henry
and Godfrey belonged to the family which held the castle of Esch-sur-Sre
in the Ardennes, and were probably vassals of the count of Luxembourg,
although Albert of Aachen indicates that Henry and possibly also his brother
had previously been in the service of Godfrey of Bouillon (l6). With them
on crusade were kinsmen of theirs from further north, the brothers Franco
and Sigemar of Maasmechelen (I7). As in the case of Warner of Grez it is
impossible to determine the exact degree of kinship between these four
crusaders and duke Godfrey.
Cuno (or Cono), count of Montaigu, who was accompanied on crusade
by his sons Gozelo and Lambert, was the eldest son of Gozelo, count of
Behogne (18). Montaigu, the Mons Acutus which was the focal point of his
domains, was a castle on the left bank of the River Ourthe near Marcourt
in the Ardennes. Cuno was one of the most important vassals of the church
of Lige. As well as being advocate of Dinant he was episcopal count of
Huy, the fortress which lay at the centre of the prince-bishop's dominions
and which was his place of refuge in times of danger. His brother, the Arch
deacon
Henry of Montaigu, was dean of the church of St. Lambert, while
one of his sons, another Henry, was also an archdeacon and provost of the
church of Fosses (I9). The claim of the historian Orderic Vitalis that Cuno
(14) St. Bormans and E. Schoolmeesters, Cartulaire de l'glise de Saint- Lambert
Lige, (Bruxelles, 1893-1933), I, n 29.
(15) AA pp. 299, 300, 305-307, 328, 366, 423, 427, 435.
(16) Triumphus S. Remacli, MGH SS, XI, 447-448 ; Thiofrid of Echternach,
Vita S. Willibrordi, MGH SS, XXIII, 26 ; J. Vannrus, 'Les anciens dynastes d'Eschsur-la-Sre', Ons Hmecht, 11 (1905), 308-390, 432-442, 485-493, 532-540 ; 12 (1906),
18-23, 51-56 ; A A pp. 427, 300.
(17) AA pp. 413, 519.
(18) AA pp. 306, 310, 317, 359, 366, 422, 464, 495, 504 ; Alberic of Troisfontaines,
Chronicon, MGH SS XXIII, 804, 815 ; La Chanson d'Antioche, d. Suzanne DuparcQuioc (Paris, 1976), lines 1397, 251 1-2544.
(19) C. G. Roland, 'Les seigneurs et comtes de Rochefort', Annales de la Socit
archologique de Namur, 20 (1893), 63-141 ; Andr Joris, La ville de Huy au Moyen
ge, Bibliothque de la Facult de Philosophie et Lettres de l'Universit de Lige, 152
(Paris, 1959) ; Jean-Louis Kpper, Lige et l'glise imperiale, xie-xiie sicles, Biblio-

THE ARMY OF GODFREY OF BOUILLON

305

was married to a sister of Godfrey of Bouillon has no basis in fact. Godfrey's

parents, Eustace II of Boulogne and Ida of Bouillon, are not known to have
had any daughters, while the chronicle of the abbey of Saint-Hubert states
that Cuno's only known wife, Ida, was a daughter of Lambert the Old, a noble
manof the territory of Lige who was buried there (20). Nevertheless the name
Gozelo, a variant form of Godfrey, borne by Cuno's father as well as his
eldest son, clearly points to some kind of blood relationship with the ArdenneVerdun dynasty which used both Godfrey and Gozelo as Leitnamen (2I).
The younger Gozelo died on crusade and so Cuno was succeeded by his
younger son Lambert, who also became count of Clermont, a fortress on
the right bank of the Meuse between Huy and Lige. The manner in which
the family acquired the county throws light on the identification of another
crusader who has escaped the attention of historians. The previous count of
Clermont was called Giselbert ; a document of 1083 relates how the church
of St. Paul at Lige possessed a domain at Nandrin near the castle of Clermont.
The count Giselbert and his accomplice Fredelo continually committed de
predations
on this land, so that its inhabitants did not dare to gather wood
or till the soil. The prince-bishop Henry of Verdun gave the advocacy of
this domain to count Cuno, in whose county the land was situated, and
forced Giselbert and Fredelo to restore what they had usurped (22). This
information provides a revealing personal constellation. Cuno was of course
the count of Montaigu ; Fredelo was in all probability the same person as
the despoiler of Malmedy and Echternach and father of the crusaders Henry
and Godfrey of Esch. Giselbert of Clermont appears again with his wife
Longarde and brother Herman in 1091 when they gave the church of SaintSymphorien to the abbey of Cluny (23). By 1095 the castle of Clermont had
become such a menace to shipping on the Meuse that Otbert, the new princebishop, organised an expedition to besiege it as part of the campaign to enforce
the Peace of God in his diocese. The siege lasted from 29 June to 9 August
1095 and ended unsuccessfully, partly because Godfrey of Bouillon and other

thque de la Facult de Philosophie et Lettres de l'Universit de Lige, 228 (Paris,

1981), 396-W3, 146-54.
(20) Orderic Vitalis, The Ecclesiastical History, ed. Marjorie Chibnall (Oxford,
1969-80), V, 166 ; Cantatorium, p. 181 ; Nicolas Huyghebaert, 'La mre de Godefroid
de Bouillon : La comtesse Ide de Boulogne', Publications de la Section historique
de l'Institut Grand-Ducal de Luxembourg, 95 (1981), 43-63.
(21) Michel Parisse, 'Gnalogie de la Maison d'Ardenne', Publications de la
Section historique de l'Institut Grand-Ducal de Luxembourg, 95 (1981), 9-42.
(22) Joseph Daris, Notices historiques sur les glises du diocse de Lige, 12 (Lige,
1885), n 41, pp. 131-133.
(23) A. Miraeus, Opera diplomatica et historica, ed. J. F. Foppens (Bruxelles,
1723-48), II, 812.

306

. V. MURRAY

nobles in the besieging army refused to attack the castle owing to an unresolved
dispute concerning the deposed abbot of Saint-Hubert (24).
In his study of the counts of Behogne and Rochefort the historian Roland
was unable to explain how Lambert of Montaigu acquired Clermont-surMeuse. He assumed that after the death of Giselbert his castle was occupied
by brigands, leading to the siege of 1095, and that his daughter later married
Lambert. However a different interpretation of the known facts provides a
more consistent and plausible explanation. The activities of the anonymous
occupants of Clermont in 1095 are entirely consistent with those of count
Giselbert and his accomplice Fredelo in 1083 ; what a monkish chronicler
regarded as brigandage was no doubt perceived by the noblemen as the
legitimate levying of tolls on river traffic. Nevertheless this noble way of life
was becoming increasingly redundant in the face of an energetic bishop
determined to pacify his diocese. A terse entry in the chronicle of Giles of
Orval for the year 1095 reveals that what Otbert failed to achieve by force
was now accomplished by purchase (25). The acquisition of Clermont and its
subsequent enfeoffment to the episcopal vassal Lambert of Montaigu can be
seen as part of a consistent policy of purchase which also brought to the
prince-bishop the important fortresses of Mirwart, Couvin and eventually,
Bouillon itself (26). For count Giselbert the obvious avenue of opportunity
and escape was the crusade. Far from having died in 1095 he seems to have
in fact departed with Godfrey, appearing as a member of the household of
Baldwin in Cilicia in the winter of 1097, described as Giselbertus de Claro
Monte. His participation in the crusade would also explain that of Milo de
Claro Monte, a knight of Godfrey's household in 1099, who may have been
a relative or follower of Giselbert (27).
The crusaders discussed so far came either from the lands of Godfrey of
Bouillon or from the surrounding areas of the Ardennes. According to a
charter recording the sale of Baisy and Genappe by Godfrey in 1095, whose
text was later incorporated into a diploma of Henry IV for the abbey of
Nivelles, no fewer than eight crusaders were present at the transaction :
Godfrey himself, Baldwin, Cuno of Montaigu, Warner of Grez, Henry and
Godfrey of Esch, and Heribrand and Walter of Bouillon (28). It is possible
(24) Cantatorium, pp. 194-197.
(25) Giles of Orval, Gesta episcoporum Leodiensium, MGH SS, XXV, 84 : Item
Clarimontis castellum beato Lamberto multo precio acquisivit.
(26) Ren Deprez, 'La politique castrale dans la principaut episcopate de Lige
du xe au XIVe sicle', Le Moyen ge, 65 (1959), 501-538.
(27) AA pp. 350, 520, 526 ; William of Tyre, Willelmi Tyrensis archiepiscopi chronicon, d. R. B. C. Huygens, Corpus Christianorum, Continuatio Medievalis, 6363 A (Turnhout, 1986) [henceforth cited as WT], pp. 219, 229.
(28) MGH DD Heinrich IV, n 459.

THE ARMY OF GODFREY OF BOUILLON

307

that the sale and others like it presented an opportunity to formulate and
discuss plans for the journey to Jerusalem.
Geographically removed from this close-knit group was Baldwin II of Mons,
count of Hainaut (29). His preference for the company of Godfrey rather than
that of his neighbour Robert II of Flanders can probably be explained in
terms of dynastic politics. He was the second son of Richilda of Hainaut
and Baldwin VI of Flanders who had jointly ruled both counties. However
on his father's death his uncle Robert I usurped Flanders in a revolt in which
his brother Arnulf III was killed. Baldwin and Richilda were left with Hainaut,
guaranteed to them by the Emperor Henry IV at the diet of Lige in 1071,
by which they became vassals of Godfrey III of Lower Lotharingia (uncle
of Godfrey of Bouillon), and rear-vassals of Lige. On the death of Godfrey
in 1076 Baldwin became a direct vassal of the prince-bishop (30). As ruler of
a large county he was presumably accompanied by vassals of his own, of
whom we can identify at least two : Gerard of Avesnes (France, Nord, arr.
Avesnes-sur-Helpe) and Giselbert of Couvin (Belgium, Namur, arr. Philippeville) (31).
A third element in the exercitus of Godfrey consisted of lords and knights
from the neighbouring duchy of Upper Lotharingia and adjacent areas of
the French kingdom. The decision of these crusaders to join the army led
by Godfrey was probably influenced by the fact that their own duke, Theoderic
I, count of Bar and Montbliard, had been released from his crusading vows
because of illness and did not make the pilgrimage (32). He was, however,
represented by his eldest son Louis, count of Mousson (France, Meurtheet-Moselle, arr. Nancy), and another knight, Ralph of Mousson (33).
The Dudo de Cons mentioned twice as a member of Godfrey's army has
been repeatedly identified in crusading scholarship as originating from Konz
at the confluence of the Mosel and the Saar above Trier (34). He was in fact
(29) AA pp. 305, 315, 434-435 ; Alberic of Troisfontaines, p. 805 ; Giselbert
of Mons, Chronicon Hanoniense, d. Lon Vanderkindere (Bruxelles, 1904), 45 ;
Chanson d'Antioche, lines 1 163, 8986.
(30) MGH DD Heinrich IV, n 242 ; A. Hansay, 'L'infodation du comt de
Hainaut l'glise de Lige en 1071', Bulletin de la Socit d'art et d'histoire du diocse
de Lige, 13 (1902), 45-58.
(31) Gerard of Avesnes : AA pp. 499, 507, 516, 593. Giselbert of Couvin : A A p. 655.
(32) Jacques Choux, Recherches sur le diocse de Toul au temps de la rforme
grgorienne : l'piscopat de Pibon (1069-1107) (Nancy, 1952), pp. 102-105.
(33) AA pp. 317, 526, 531 ; Alberic of Troisfontaines, p. 804 ; Chanson d'Ant
ioche, line 8975.
(34) AA pp. 299, 574 ; Runciman, I, 147 ; Peter Knoch, Studien zu Adalbert von
Aachen, Stuttgarter Beitrge zur Geschichte und Politik, 1 (Stuttgart, 1966), pp. 114,
157. Reinhold Rhricht, Die Deutschen im heiligen Lande (Innsbruck, 1894), p. 17,
Andressohn, p. 52, and Duparc-Quioc {Chanson d'Antioche, p. 73) all mistakenly
give Saarbrcken as a place of origin.

308

. V. MURRAY

lord of Cons-la-Grandville (France, Meurthe-et-Moselle, air. Briey) on the

River Chiers in the Ardennes, and was a benefactor of the abbey of SaintHubert to which he donated the priory of St. Michael at Cons (35). His father
was Adelo of Dun, Dun being a castle on the Meuse which had been given
to the church of Verdun by Henry IV. On Adelo's death his lands were divided
between his sons Walter who received Dun, and Dudo who received Cons.
Dudo was accompanied on crusade by his wife Hadwida, daughter of Arnulf II
of Chiny (36).
Similar problems of identification have arisen in the case of the crusader
Petrus de Stadeneis, whom historians have almost unanimously referred to
as 'Peter of Stenay' (37). This issue is an important one ; the identification
with Stenay (France, Meuse, arr. Verdun) suggests Peter was a vassal of
Godfrey of Bouillon who was lord of this domain on the Meuse midway
between Bouillon and Verdun. However Peter was in fact count of the pagus
Stadunensis or Astenois, a district on the plains of Champagne in the extreme
east of the French kingdom. He was also known as Petrus de Dunperrun
after his principal stronghold of Dampierre-le-Chteau (France, Marne, air.
Sainte-Menehould) (38). With him on crusade was his elder brother Rainald
III, episcopal count of Toul in Upper Lotharingia (39). They were sons of
Frederick I, count of Astenois, and Gertrude, daughter of count Rainald II
of Toul, and were kinsmen of Godfrey of Bouillon. Several authors allude
to a blood relationship between Frederick's brother Henry of Verdun and
Godfrey's uncle Godfrey III, at whose instigation Henry was installed as bishop
(35) Cantatorium, pp. 150, 164 ; Godefroid Kurth, Chartes de l'abbaye de St.Hubert en Ardenne, I (Bruxelles, 1903), n 63 ; Raymond Pagny, 'Les seigneurs et
la seigneurie de Cons-la-Grandville', Bulletin de l'Association 'Les Amis du Vieux
Longwy', 5 (1962), 3-9, 74-78.
(36) MGH DD Heinrich IV, n 162; Kurth, ibid. : commemoratio Hawidis filiae
comitis Arnulphi de Chisnei, quae cum viro suo Dudone Cunensi ab Hierosolymis
reversa.
(37) Reinhold Rhricht, Geschichte des ersten Kreuzzuges (Innsbruck, 1901), 62 ;
Ren Grousset, Histoire des croisades (Paris, 1934), I, 12 ; Runciman, I, 147 ;
Ferdinand Chalandon, Histoire de la premire croisade (Paris, 1925), 112; H.
Hardenberg, De Nederlanden en de kruistochten (Amsterdam, 1944) 57-58 ; Marcel
Lobet, Godefroid de Bouillon. Essai de biographie antilgendaire (Bruxelles, 1943),
71 ; Knoch, pp. 167-169 ; Kenneth M. Setton et al., A History of the Crusades (Phila
delphia, Madison, 1955-89), I, 296.
(38) AA pp. 299, 310, 317, 365-66, 422, 424 ; Laurence of Lige, Gesta episoporum
Virdunensium et abbatum s. Vitoni, MGH SS, X [henceforth cited as LL], p. 494 ;
Chanson d'Antioche, lines 2260, 8188, 8205 ; Anatole de Barthlmy, 'Le comt
d'Astenois et les comtes de Dampierre-le-Chteau', Revue de Champagne et de Brie,
16 (1888), 410-416.
(39) AA pp. 299, 301, 317, 343, 365, 366, 398, 422 ; LL p. 494 ; Baldric of Dol,
Historia Jerosolimitana, Recueil des Historiens des Croisades, Historiens occidentaux,
IV (1879), 77.

THE ARMY OF GODFREY OF BOUILLON

309

of Lige in 1075. It is likely that Baldwin, father of Henry and Frederick,

had married a daughter of the Ardenne- Verdun dynasty 40). According to
the chronicler Laurence of Lige Frederick was a carus et fidelis of both
Theoderic, bishop of Verdun, and of Godfrey of Bouillon, who held the county
of Verdun from the bishop until his departure on the crusade in 1096. It
is likely that Frederick and his son held lands in the neighbouring county
of Verdun and were thus vassals of both count and bishop ; certainly it is
a significant reflection of the relative strengths of the two institutions that
Peter had served as one of the commanders of the episcopal forces during
Theoderic's campaigns against Godfrey (41). Among those who went to Jeru
salem in the company of Peter and Rainald were one cleric, Louis, archdeacon
of Toul, and six lords from the diocese : Rambert, son of Fraimer of Lironville
(France, Meurthe-et-Moselle, arr. Toul), Bencelin, Aldo of Fontenoy-surMoselle (air. Toul), and the family group of Lanfroi, his son Olri and brother
Hugh H.
The last known member of this category was the crusader known to
contemporary sources as Balduinus de Burgo, later count of Edessa and second
king of Jerusalem (43). Although often referred to by modern historians as
'Baldwin of Le Bourg', his surname in fact derives from Bourcq (France,
Ardennes, arr. Vouziers) in the valley of the Aisne (u). He was a son of count
Hugh I of Rethel and Melisende of Montlhry, and was a kinsman of Godfrey.
The county of Rethel, which had as its nucleus the territory of Omont, a
fief of the church of Reims, was situated both in regno and in imperio and
the counts were consequently involved in the politics of Lotharingia. Baldwin's
grandfather had been an ally of Godfrey of Bouillon during the disputes over
the latter's inheritance, and had invaded the bishopric of Verdun although
his own castle of Sainte-Menehould was captured by episcopal forces (45).
It is likely that most of the crusaders from the other duchies of the empire
had already gone with the various expeditions which had left prior to the
official departure date under the leadership of Peter the Hermit, Walter SansAvoir, Emicho, Gottschalk and Volkmar. However a diploma of Henry IV
(40) Charles Aimond, 'Le ncrologe de la cathdrale de Verdun', Jahrbuch der
Gesellschaft fr lothringische Geschichte und Altertumskunde, 21 (1909), 185; LL
p. 495 ; Giles of Orval, p. 88 ; Lambert of Hersfeld, Annales, ed. O. HolderEgger, NIGH Scriptores rerum Germanicarum in usum scholarum (Hannover,
Leipzig, 1956), p. 225 ; Gesta Treverorum, MGH SS, VIII, 126.
(41) LL pp. 494-497.
(42) AA pp. 375-376 ; Choux, pp. 102-105.
(43) AA p. 299.
(44) His surname was first identified by Jean Richard, Le Royaume latin de
Jrusalem (Paris, 1953), p. 19.
(45) AA p. 527 ; WT p. 547 ; LL p. 494 ; Michel Bur, La formation du comt de
Champagne 950-1150 (Nancy, 1977), pp. 133-134, 412 ff.

310

A.V.MURRAY

dated 26 July 1097 reveals how a noblewoman called Cunihild had sold her
estate of Obermeiser (Germany, Hessen, Kr. Kassel-Land) to the abbey of
Helmarshausen, for which her son Reinhold had received '36 marks to go
to Jerusalem with Duke Godfrey'. Since he came from a fairly distant part
of Germany it was unlikely that Reinhold was a unique case (46).
The other crusading contingents, notably the Provenal army led by
Raymond of Saint-Gilles, included large numbers of clerics, both secular and
regular, some of whom held fairly senior positions in the church. It was these
men who tended to obtain the new ecclesiastical offices created by the crusaders
in Syria. Thus the Provenal Peter of Narbonne became bishop of Albara
while his compatriot Bernard of Valence became bishop of Artah and sub
sequently
Latin patriarch of Antioch. The Norman Robert of Rouen obtained
the see of Lydda while the Flemings Arnulf of Chocques and Achard of
Arrouaise became patriarch of Jerusalem and prior of the Templum Domini
respectively (47). However senior clerics of this calibre were conspicuously
absent from Godfrey's army, a phenomenon which can be explained by the
effects of the Investiture Contest. No Lotharingian or other German bishops
were present at the Council of Clermont ; only Richer of Verdun sent legates
and presents (48). Indeed the first news of the crusade to reach the western
parts of the empire may have been brought by Peter the Hermit and his
followers, which would explain why he rather than Urban II was depicted
as the initiator of the crusade by many German chroniclers (49). It is also
probable that the Lotharingian bishops and abbots were too heavily involved
in the struggle between empire and papacy to consider leaving their charges
at this critical time.
It is therefore not surprising that the only named clerics known to have
travelled in Godfrey's company were the aforementioned Louis of Toul and
Adalbero of Luxembourg, archdeacon of Metz, who was a son of Conrad
I of Luxembourg and aproximus of Henry IV (50). On the other hand William
(46) MGH DD Heinrich IV, n 457.
(47) Bernard Hamilton, The Latin Church in the Crusader States (London, 1980),
11, 17, 23 ; Moeller, 189-202.
(48) LL p. 497 ; Annales S. Vitoni Virdunenses, MGH SS, X, 526.
(49) Knoch, 109-111.
(50) AA pp. 370-371 ; Heinz Renn, Das erste Luxemburger Grafenhaus, Rheinisches
Archiv, 39 (Berlin, 1941), p. 9. The name Adalbero, relatively rare among laymen,
was considered auspicious for bishops after the time of Adalbero, bishop of Augsburg
(d. 909) and seems to have been especially popular in the Ardenne- Verdun, Bar and
Luxembourg families for younger sons destined for the church. Adalbero III, bishop
of Metz from 1047 to 1072, was an uncle of the archdeacon. These facts, and the
details of the anecdotal vignette relating to his death at Antioch told by Albert of
Aachen, suggest that he was a fairly worldly career churchman for whom the crusade
presented a welcome diversion during the wait to secure one of the sees on which
his family had traditional claims.

THE ARMY OF GODFREY OF BOUILLON

31 1

of Tyre relates that Godfrey took with him a group of monks who celebrated
the divine offices for him during the crusade. After the capture of Jerusalem
he settled them at their own request in the abbey of St. Mary in the valley
of Josaphat (51). Where did these monks come from ? The chronicle of SaintHubert agrees that Godfrey took regular clergy with him, and it is possible
that they were at least in part originally members of that community (52).
On the death of Henry of Verdun in 1091, Henry IV installed the royal chaplain
Otbert as bishop of Lige. Soon the passionate imperialist Otbert deposed
Berengar, the reformist abbot of St. Laurence at Lige. Berengar fled with
his supporters to Saint-Hubert, whose own abbot Theoderic was in turn
deposed for offering his protection to the reformers. Eventually the two abbots
and many loyal monks fled to properties in the diocese of Reims where they
enjoyed the support of Godfrey of Bouillon and Dudo of Cons (53). The dispute
was not finally resolved until the autumn of 1096, and therefore it would
seem that in August of that year there were numbers of displaced monks,
supporters of Berengar and Theoderic, who may well have been attracted
by the prospects of a pilgrimage to the Holy Land in the company of Godfrey,
who was advocate of Saint-Hubert.
So far the examination of the composition of Godfrey's exercitus has yielded
the names of thirty-nine individuals who were probably with him from the
beginning of the expedition. Two of these were clerics and two were women.
Undoubtedly the army included many more unnamed knights and their
families as well as clerics, peasants and townspeople. Within the known group
of thirty-nine individuals we can discern certain relationships and bonds which
operated in the course of the expedition, and which give insights into the
structure and hierarchy of the Lotharingian army.
Next in importance to Godfrey himself was his brother Baldwin of
Boulogne. He was left in charge of the army at Brck an der Leitha during
negotiations with King Coloman of Hungary and later acted as hostage to
him. At Constantinople he commanded a detachment of 500 men, and guarded
the hostage John Comnenus (54). The initial embassy to Coloman was en
trusted
to Godfrey of Esch, probably because he had been to Hungary some
years before ; on the second approach he was accompanied by Warner of
Grez, Rainald of Toul and Peter of Dampierre (55). Godfrey of Esch, Cuno
(51) WTpp.431.
(52) Cantatorium, p. 208 : Nee mult o post dux Hierosolymam vadens ... multos
secum nobiles et religiosos abduxit.
(53) Ralph of Sint-Truiden, Gesta abbatwn Trudonensium, MGH SS, X, 236 ff ;
Cantatorium, pp.153 ff. It was on account of Otbert's refusal to restore Theoderic
that Godfrey abandoned the siege of Clermont-sur-Meuse in 1095.
(54) pp. 301-302,307-308.
(55) pp. 300-301.

312

A.V.MURRAY

of Montaigu and Baldwin of Bourcq acted as envoys to the Emperor Alexius ;

the latter two subsequently received the imperial representative John Comnenus (56). Duke Godfrey was then accompanied to his meeting with Alexius
by Warner of Grez and Peter of Dampierre (57). We later find Warner in
charge of a force sent to secure the port of St. Simeon on the coast of northern
Syria, and Rainald of Toul as commander of a reserve division at the Great
Battle of Antioch in 1098 (58).
It is significant that these important military and diplomatic tasks were
not entrusted to any of Godfrey's household officers, suggesting that his own
vassals did not possess sufficient status. On the other hand the fact that such
missions were repeatedly carried out by the same men suggests that they
formed a close-knit inner group within the Lotharingian army which probably
functioned as a council. It is improbable that the bonds which linked them
were of a feudal nature as, apart from the possible cases of Henry and Godfrey
of Esch, there is no evidence to suggest that they were his vassals. Lower
Lotharingia was not a territorial principality, and its counts and lords were
not vassals of the duke (59). On the other hand as military command was
the most pronounced and the least disputed function of the duke it is probable
that the lords from Lower Lotharingia were prepared to accept Godfrey as
leader of the crusading army. In fact two of the Lower Lotharingians, Baldwin
of Hainaut and Henry of Esch, went so far as to break military discipline
by leaving the army in Thrace in order to hurry ahead to Constantinople
in an attempt to share in the rumoured munificence of the Byzantine
emperor (). Baldwin of Bourcq, Peter of Dampierre and Rainald of Toul
all came from France and Upper Lotharingia, areas where Godfrey had no
ducal authority. The most significant ties within the core group seem in fact
to have derived from kinship ; each member of it was related in some way
to Godfrey of Bouillon and his brother Baldwin, a fact repeatedly reflected
in the descriptions of them given by Albert of Aachen. The most important
of Godfrey's kinsmen (apart from his brother) was Warner of Grez, who
assumed the leadership of the domus Godefridi on Godfrey's death in July
1 100. In Albert's initial list of departing crusaders Warner is the only individual
explicitly described as a kinsman of Godfrey and Baldwin, although many
of the others are subsequently marked as such elsewhere in his history. In
the corresponding passage in the history of William of Tyre, who used Albert's
(56) A A pp. 306-307, 310.
(57) AAp. 310.
(58) AA pp. 383, 424 ; Baldric of Dol, p. 77 ; Orderic Vitalis, V, 108-1 10.
(59) Matthias Werner, 'Der Herzog von Lothringen in salischer Zeit', in Die Salier
und das Reich, 1 : Salier, Adel und Reichsverfassung, ed. Stefan Weinf-urter and
Helmut Kluger (Sigmaringen, 1991), 367-473.
(60) AA p. 305.

THE ARMY OF GODFREY OF BOUILLON

313

history or a source common to both, only Baldwin of Bourcq is similarly

distinguished. This would appear to be a retrospective editorial decision
influenced by Baldwin's subsequent importance as king of Jerusalem (6I). This,
and the analysis of their respective military and diplomatic responsibilities,
suggests that at least originally Warner was near the top of the hierarchy
of the core group and Baldwin somewhere near the bottom.
Another significant factor may have been the bonds of alliance forged during
the wars fought by Godfrey to defend his inheritance in the period before
the crusade. In 1076 he had succeeded as designated heir to the Lotharingian
lands of his uncle, Godfrey III (the Hunchback), but claims were also raised
by his uncle's estranged wife, Mathilda of Tuscany, Albert III, count of Namur,
and Theoderic Flamens, count of Veluwe on the lower Rhine. This coalition
was joined by another Theoderic, the bishop of Verdun, and two local dynasts
keen to share in the spoils, Waleran, count of Arlon and Limburg, and Arnulf
II, count of Chiny. By contrast Godfrey's principal support came from his
relative Henry of Verdun, prince-bishop of Lige (62). Turning to the crusade,
we have already seen that Cuno, Gozelo and Lambert of Montaigu as well
as Warner of Grez and Baldwin of Hainaut were all vassals of the ecclesiastical
principality. Baldwin of Bourcq's family had fought on Godfrey's side against
Theoderic of Verdun. Admittedly Peter of Dampierre and Rainald of Toul
had fought on the episcopal side in this conflict ; however as far as the crusade
was concerned this appears to have been outweighed by their particularly
strong kinship ties (via a bishop of Lige) with Godfrey and Baldwin.
Conversely, the continued importance of kinship and alliance would explain
the lack of prominence accorded to Dudo of Cons, a son-in-law of Arnulf
of Chiny, and to Louis of Mousson, cousin of the countess Mathilda.
Considering the peripheral position of Lower Lotharingia within the empire,
as well as its accessibility to France and the preaching of the crusade, Godfrey's
army included relatively few of the major nobles of the duchy, especially those
of comital rank. As we have seen, the nobles of Lower Lotharingia were
not vassals of the duke and were consequently under no obligation to follow
him. In many cases the phenomenon of non-participation was also influenced
by the political constellations obtaining on the eve of the crusade. Notable
(61) AA pp. 299, 520-524 ; WT p. 161.
(62) Walter Mohr, Geschichte des Herzogtums Lothringen, II. Niederlothringen
bis zu seinem Aufgehen im Herzogtum Brabant (11.-13. Jahrhundert) (Saarbrcken,
1976), 63-69 ; P. C. Boeren, 'Overzicht der dynastie van Wassenburg-Gelre en verwante
geslachten gedurende de elfde eeuw', Bijdragen en Mededelingen van de Vereeniging
Gelre, 41 (1938), 1-23 ; P. C. Boeren, De oorsprong van Limburg en Gelre en enkele
naburige heerschappijen (Maastricht, 1938), 71-86 ; H. Vanderlinden, 'Le tribunal
de paix de Henri de Verdun (1082)', in Mlanges Henri Pirenne (Bruxelles, 1926),
II, 589-596.

314

A.V.MURRAY

absentees were Albert of Namur, Arnulf of Chiny, Henry of Arlon and

Limburg, as well as the three heirs of Theoderic Flamens : Gerard of Geldern,
Henry of Kriekenbeek and Gosuin of Heinsberg. These were all members
(or their successors) of the coalition which had waged war on Godfrey and
his principal ally the church of Lige. A similar case was that of the count
of Leuven, traditionally a rival of the Ardenne-Verdun dynasty and an
adversary of the principality of Lige on which his lands bordered (63). In
1095-96, far from taking the cross, Henry of Leuven embarked on a new
conflict with Lige over the county of Brugeron (M). Other important absentees
were the count of Holland and the count of Luxembourg, while we know
of no nobles at all from Lower Lotharingia north of Antwerp. It has been
argued that the high degree of non-participation in Lotharingia was because
most of the vassals of Henry IV were reluctant to become involved in what
was essentially a papal enterprise (65). In fact it would appear that if anything,
exactly the converse was the case. Under the bishops Henry of Verdun and
Otbert the diocese of Lige was one of the areas which consistently offered
the greatest support to the monarchy during the Investiture Contest i66).
Notable non-participants included traditional enemies of the ecclesiastical
principality ; key members of the crusading army were fidles sancti Lamberti.
Pope Urban had proclaimed that the property of crusaders should be placed
under church protection until their return. For vassals of the powerful church
of Lige this undoubtedly represented effective protection. On the other hand
it is probable that many other lords were unwilling to leave home at a time
when Lotharingia was being ravaged by the Investiture Contest and the various
dynastic feuds which accompanied it (67).
From its departure in August 1096 until its arrival at Constantinople in
December the army was therefore essentially Lotharingian in character, a
lthough
it was by no means representative of the whole of Lotharingia. After
crossing over to Asia Minor in the spring of 1097 the crusaders encountered
(63) Vita Balderici episcopi Leodiensis, MGH SS, IV, 727-729 ; Gesta episcoporum
Cameracensium, MGH SS, VII, 467-469 ; Annales Laubienses, MGH SS, IV, 18.
(64) Giles of Orval, pp. 91-92.
(65) C. D. J. Brandt, Kruisvaarders naar Jeruzalem. Geschiedenis van de eerste
kruistocht (Utrecht, 1950), 103.
(66) Kupper, Lige et l'glise impriale, pp. 387-394. For recent numismatic
evidence of Otbert's imperialist stance, see Jrgen Steen Jensen, ' Maastricht Coin
Representing the Investiture of Bishop Otbert of Lige by the Emperor Henry IV
in 109, Spink Numismatic Circular, 98 (1990), 310-311.
(67) A case in point is that of Henry-Hezelin II, count of Grandpr, who had been
Godfrey's principal partisan in the county of Verdun. Even after Godfrey reached
a settlement with Bishop Richer he continued to wage war on the bishopric until
he took the cross sometime after 1099, although he died before he could leave (LL
p. 497).

THE ARMY OF GODFREY OF BOUILLON

315

the defeated remnants of the 'People's Expeditions', which seem to have been
incorporated into the various newly arrived contingents in the course of the
next stage of the march. Thus at Nicaea Godfrey and his army are described :
Godefridus dux Lotharingiae ... cum universo comitatu Lotharingiorum
constitutus est (68). However, by the time of their arrival at Antioch the
descriptions have become fuller : Godefridus dux ... cum innumeris milibus
Lothar ingiis, Saxonibus, Alemannis, Bawariis and Godefridus, cum Theutonicis, Alemannis, Bawariis, Saxonibus, Lotharingiis and again cum Alemann
is,
Bawariis, Saxonibus, Lotharingiis, Theutonicis et Romanis (69). As Knoch
points out, these lists of tribes constitute a rhetorical device employed by
Albert of Aachen to give greater weight to the German elements in the crusade.
Nevertheless, the fact that these Germans are not associated with Godfrey
until after the crossing to Asia Minor suggests that they were new additions
to the army (70).
A section of the chronicle of Zimmern dealing with the First Crusade has
long been regarded as the principal source for these new additions (71). The
chronicle was the work of Froben Christoph, count of Zimmern (now
Herrenzimmern near Rottweil, Baden-Wrttemberg) and lord of Mekirch
and Wildenstein. It was completed around 1566 and survives in two original
manuscripts written in the Swabian dialect of Early New High German (72).
The first historian of the crusades to draw attention to this source was Hagenmeyer, who originally intended to use it to clarify the role played by Peter
the Hermit in the First Crusade (73). On account of its unique and detailed
information it was accepted by Hagenmeyer, Rhricht, Runciman and later
by Riley-Smith as a genuine prosopographical source for German participation
in the crusade (74). The chronicle claims as its own main source for the crusade
(68) A A p. 315.
(69) AA pp. 366, 422, 425. See also Orderic Vitalis, V, 108-1 10.
(70) Knoch, pp. 116-119.
(71) First published as the Zimmerische Chronik, ed. Karl August Barack,
Bibliothek des litterarischen Vereins in Stuttgart, 91-94 (Stuttgart, 1869), 2nd edn
Freiburg im Breisgau, Tbingen, 1881-82. References are henceforth given to the best
modern edition, Die Chronik der Grafen von Zimmern. Handschriften 580 und 581
der Frstlich Frstenbergischen Bibliothek Donaueschingen, ed. Hansmartin DeckerHauff (Konstanz, Stuttgart, Sigmaringen, 1964-).
(72) Hans Baumgart, 'Studien zur Zimmerschen Chronik des Grafen Froben
Christoph und zur Mainzer Bistumschronik des Grafen Wilhelm Werner von Zimmern'
(unpublished doctoral dissertation, Univ. of Freiburg, 1923), 9-30 ; Beat Rudolf Jenny,
Graf Froben Christoph von Zimmern. Geschichtsschreiber, Erzhler, Landesherr. Ein
Beitrag zur Geschichte des Humanismus in Schwaben (Lindau, Konstanz, 1959), 34-50.
(73) Heinrich Hagenmeyer, 'tude sur la Chronique de Zimmern : renseignements
qu'elle fournit sur la premire croisade', Archives de l'Orient latin, 2 (1884), 17-88.
(74) Rhricht, Die Deutschen, 9-21 ; Runciman, I, 121-122, 131-132; Jonathan
Riley-Smith, 'The Motives of the Earliest Crusaders and the Settlement of Latin

316

A.V.MURRAY

a codex described as ain alt geschriben buoch in the Black Forest monastery
of Alpirsbach which was founded during the First Crusade by members of
the Zimmern family among others, as well as a tapestry also supposedly
belonging to Alpirsbach. The self-proclaimed intention of this section of the
chronicle is to highlight the role of the High Germans in contrast to that
of the Low Germans and French who, it was argued, had been amply treated
by other writers, notably William of Tyre, Robert the Monk and the other
wise unknown Guido Remensis (75). After describing Pope Urban's appeal
at Clermont, the chronicle goes on to give a detailed list of those Germans
who departed for Jerusalem : 'Als nu die frsten aus Gallia und aus andern
nationen das eerlich, christenlich frnemen vernamen, die auch zuvor durch
den bapst Urbanum zu Clermant in Auergne darzu bewegt, erweiten sie herzog
Gotfriden ainhelligclich zu irem obristen ber den ganzen hufen. Sollichs
mcht so baldt in deutschen landen nit kont werden, es namen etlich bischof
das creuz an sich, als nemlich bischof Conradt von Chur und bischof Otho
von Straburg, herzog Friderrichs von Schwaben breder. Zu disen und
andern bischofen mer verfliegte sich bischof Thiemo von Salzburg, dessgleichen
herzog Egkhart von Bayern, ain sun grave Ottons von Scheyrn, und herzog
Walther von Tegk. Dessgleichen so zogent mit dise nachvolgende graven und
freiherren : grave Hainrich von Schwarzenburg, pfalzgrafe Hugo von Tbin
gen,grave Rudolf und grave Huldreich von Sarwerden, grave Hartman von
Dillingen und Kiburg, grave Thiemo von Eschenloch, grave Hainrich von
Helfenstain, grave Adelprecht von Kirchberg, grave Hainrich von Hailigenberg,
ain grave von Fanen, herr Arnolt freiherr von Busnang, ain freiherr von
Fridow, herr Ruodolf freiherr von Brandis, ain freiherr von Westerburg, grave
Berchtoldt von Neifen, herr Albrecht freiherr von Stoffeln ; item ain grave
von Salm, ain grave von Viernenberg, ain herr von Bolanden ; item grave
Emmich von Lyningen, ain grave von Rttelen und ain grave von Zwaibrucken,
darzu ain merkliche anzal von der ritterschaft, die alle zu errettung des christenlichen glaubens mit denen unglebigen zu streiten begerten' (76).
This list comprises twenty-seven named individuals, the majority of them
from the duchy of Swabia ; to these can be added ain edelman von Embs
und ainer von Fridingen as well as the brothers Conrad, Albert and Frederick
of Zimmern who are all mentioned later (77). Thus this single source gives
a total of thirty-two names, an amazingly high prosopographical yield for
a relatively short account in a work written over four and a half centuries
after the events it describes. The number is even more remarkable if we
Palestine, 1095-1100', English Historical Review, 98 (1983), 725; Riley-Smith, The
First Crusade and the Idea of Crusading (London, 1986), 50-51.
(75) Chronik der Grafen von Zimmern, I, 73-75.
(76) Chronik der Grafen von Zimmern, I, 75.
(77) Chronik der Grafen von Zimmern, I, 79.

THE ARMY OF GODFREY OF BOUILLON

317

compare it with the thirty-nine names for the original Lothanngian contingent
arrived at by examination of all other sources combined, including the fulllength account of Albert of Aachen.
Yet it is difficult to accept unquestioningly the evidence of the Zimmern
list. Ten names are given only with the formula 'a lord of X', with no forename,
and are thus valueless in prosopographical terms since none of these ten can
be confirmed from any other source. Of the remaining twenty-two names,
three are bishops, those of Chur, Strasbourg and Salzburg. At the time of
the council of Clermont the bishop of Chur was Ulrich II of Tarasp. He
died on 30 July 1096 and was succeeded by Guy who reigned until 1122.
The first bishop of Chur to bear the name Conrad was not elected until
1123 (78). Otto, brother of duke Frederick of Swabia, was installed as bishop
of Strasbourg by Henry IV before 1084. His participation in the crusade is
problematic. As he was still in Strasbourg on 12 July 1097 he could not
have left with Peter the Hermit and Walter Sans-Avoir (79). Although he is
attested as having made a pilgrimage he was back in Germany by 9 November
1099, which hardly lends much support to the testimony of the chronicle
of Zimmern (80). Similar confusion seemed to have led to the inclusion in
the list of Thiemo, archbishop of Salzburg, who did not depart for Palestine
until 1101 (8I). Thus none of these three could have taken part in the 'People's
Expedition' whose German component is described in the chronicle.
The name herzog Egkhart von Bayern, ain sun grave Ottons von Scheyrn
raises further problems. The duchy of Bavaria was held personally by Henry
IV from 1077 until the summer of 1096 ; it was then returned to Welf IV
who was succeeded by his son Welf V in 1098 (82). Bavaria did not pass to
the Scheyern family until Otto of Wittelsbach was created duke by Frederick
Barbarossa in 1 180. Count Otto I of Scheyern, who died before 1078, is known
to have had a son called Ekkehard ; however since Ekkehard died before
1091 he could not have been on crusade. The Wittelsbachs (as the Scheyern
line became known) were later involved in crusading and were keen patrons
of crusading literature. One of the main sources of their family tradition were
the tablets known as the Tabula Perantiqua, preserved in the abbey of
Scheyern. One of these tells how Ekkehard forced King Henry II of Germany
(78) A. Bruckner, Helvetia sacra 1/ 1 (Bern, 1972), 474475.
(79) P. Wentzcke, Regesten der Bischfe von Straburg bis zum Jahre 1202
(Innsbruck, 1908), 290, 295 ; E. Schfrer, Die Straburger Bischfe im Investiturstreit
(Bonn, 1923), 111-120.
(80) Saxo Annalista, MGH SS, VI, 730 ; Bernold of St. Blasien, Chronicon,
MGH SS, V, 466 ; Wentzcke, pp. 295-298.
(81) Gesta archiepiscoporum Salisburgensium, MGH SS, XI, 58.
(82) Max Spindler, Handbuch der bayerischen Geschichte, 2nd edn (Mnchen,
1981), I, 328-331.

318

A.V.MURRAY

to make him duke of Bavaria, whereupon both led an army to the east which
captured Jerusalem. The historical core of this fantastic legend was probably
an actual pilgrimage made by Ekkehard, possibly the great German pilgrimage
of 1064-65, which was later conflated with accounts of the First Crusade and
other crusading activities of the Wittelsbach family by Froben Christoph of
Zimmern, or more likely, one of his sources (83).
Realising the difficulties posed by the inclusion of the three bishops and
Egkhart, Hagenmeyer argued that these names could not have been derived
from the claimed ultimate source, the alt geschriben buoch, although there
are no textual grounds for this within the chronicle of Zimmern (84). Yet many
of the other names accepted as genuine by Hagenmeyer also raise numerous
difficulties. The first known duke of Teck was Adalbert, son of Conrad, duke
of Zhringen, appearing with the title in 1187 (85). The duchy of Teck (near
Kirchheim in Wrttemberg) was probably created as a result of a division
of the Zhringen inheritance between Berthold V of Zhringen and his uncles
Adalbert of Teck and Hugh of Ulmburg (Uilenburg) after the death of their
elder brother Berthold IV in 1186. The division also explains the adoption
of the ducal title by all three heirs (86). However the inclusion in this account
of the First Crusade of a Teck with the ducal title, and with the name Walther
which was otherwise unknown in the family, is quite anachronistic.
The counts of Tubingen are known to have used the Christian name Hugh
in the eleventh and twelfth centuries. However this line did not receive the
title comes palatinus until the time of Hugh III, on the extinction of the
counts of Dillingen in 1 146 (87). The claim that the Count Palatine Hugh
acted as obrist or commander of the Germans along with Walter of Teck and
(83) Albert Siegmund and Franz Genzinger, 'Zur Scheyerer Tabula Perantiqua'
in Wittelsbach und Bayern. Beitrge zur bayerischen Geschichte und Kunst, ed. Hubert
Glaser (Mnchen, Zrich, 1980), I/l, 154-163; Rhricht, Die Deutschen, 7. On
the Witteisbachs as patrons of crusading literature, see Alan V. Murray, 'Reinbot
von Durne's Der heilige Georg as Crusading Literature', Forum for Modem Language
Studies, 22 (1986), 172-183.
(84) Hagenmeyer, 'tude', 39^45.
(85) Irene Grnder, Studien zur Geschichte der Herrschaft Teck, Schriften zur
sdwestdeutschen Landeskunde, 1 (Stuttgart, 1963), 3.
(86) E. Heyck, Geschichte der Herzge von Zhringen (Freiburg im Breisgau, 1891),
418; Hartmut Heinemann, 'Das Erbe der Zhringer', in Die Zhringer. Schweizer
Vortrge und neue Forschungen, ed. Karl Schmid, Vortrge zur Zhringer-Ausstellung,
3 (Sigmaringen, 1990).
(87) J. Frderer, 'Die Erbbegrbnisse und Stammgter der Tbinger Pfalzgrafen',
Tbinger Bltter, 36 (1948-49), 12-18 ; Frderer/Wic die Tbingen Grafen zur Pfal
zgrafenwrde
gekommen sind', Tbinger Bltter, 49 (1962), 4-12 ; Heinz Bhler, 'Wie
gelangten die Grafen von Tbingen zum schwbischen Pfalzgrafenamt ? Zur Geschichte
der Grafen und Pfalzgrafen von Tbingen und verwandter Geschlechter', Zeitschrift
fr wrttembergische Landesgeschichte, 40 (1981), 188-220.

THE ARMY OF GODFREY OF BOUILLON

319

died in battle at Nicaea cannot be confirmed from any other source. In view
of the anachronistic title it is possible that the inclusion of Hugh as a crusader
derived from crusading activity of this family in the years 1190-1215 (88). The
two brothers listed as counts of Saarwerden are also dubious ; the first
documented count is known only from 1111, and the names Rudolf and Ulrich
are unheard of in this family (89). Neither can the grave von Zwaibrucken
be accepted as a crusader. The founder of the comital family named after
Zweibrcken was Henry I, second son of Simon I, count of Saarbrcken
(died 1182). The names Rudolf and Ulrich are equally unknown in this
descendance, so that we may exclude any confusion between Saarbrcken
and Saarwerden on the part of the chronicle of Zimmern. These three crusaders
must therefore be regarded as pure fiction i90).
The first documented lord of Bussnang in the Thurgau was Albert I who
appears between 1150 and 1180. The name Arnold is unknown in this
family (9I). The first lord of Brandis, whose core lands were situated in the
Emmental, does not appear until 1239, with the name Conrad, making the
claimed hen Ruodolffreiherr von Brandis even more anachronistic than the
duke of Teck. Indeed, the name Rudolf does not occur in the main line of
this family until the fifteenth century (92). A Henry of Heiligenberg, in the
Linzgau north of Konstanz, is known in the period around 1 100 as the brother
of Arnold, imperialist bishop of Konstanz, and as advocate and despoiler
of the monastery of St. George of Petershausen. However nothing is known
of any crusading activity prior to this ; although the chronicle of Zimmern
calls him a count, a county named after the Mons Sanctus did not appear
until 1135 (93). The first lord of Neuffen near Esslingen am Neckar was
Mangold, son of Berthold I of Sulmetingen-Sperberseck, who appears be
tween
1087 and 1122, and built the castle of Neuffen after 1100. He was
(88) Chronik der Grafen von Zimmern, I, 79 ; J. Frderer, 'Die Tbinger Grafen
und Pfalzgrafen als Reichsfrsten', Tbinger Bltter, 40 (1953), 16-20.
(89) Hans-Walter Herrmann, Geschichte der Grafschaft Saarwerden bis zum Jahre
1527, Verffentlichungen der Kommission fr Saarlndische Landesgeschichte und
Volksforschung, 1 (Saarbrcken, 1957), I, 73.
(90) C. Phlmann, Regesten der Grafen von Zweibrcken, Verffentlichungen der
Pflzischen Gesellschaft zur Frderung der Wissenschaften, 42 (Speyer, 1962), p. 42 ;
Michel Parisse, La noblesse lorraine xie-xiiie sicle (Lille, Paris, 1975), II, 858-859.
(91) F. Stucki, 'Die Freiherren von Bussnang und von Griesenberg', Genealogisches
Handbuch zur Schweizer Geschichte, IV (Zrich, 1980), 51-96.
(92) Placid Butler, 'Die Freiherren von Brandis', Jahrbuch fr schweizerische
Geschichte, 36 (1911), 1-151.
(93) Casus monasterii Petrishusensis, MGH SS, XX, 656 ; Ilse Juliane MiscollReckert, Kloster Petershausen als bischflich-konstanzisches Eigenkloster. Studien
ber das Verhltnis zu Bischof, Adel und Reform vom 10. bis 12. Jahrhundert, For
schungen
zur Oberrheinischen Landesgeschichte, 24 (Freiburg im Breisgau, Mnchen,
1973), 114-117.

320

. V. MURRAY

succeeded by his son Egino in 1122/25. Mangold had a brother, Berthold

II, lord of Sperberseck, documented from 1087, who died a monk at Zwiefalten
after 1145. Apart from the chronicle of Zimmern there is no evidence for
participation in the First Crusade by any of these members of the family.
However it is possible that the basis of the inclusion of a grave Berchtholdt
von Neifen in the chronicle may been a tradition of a pilgrimage to Jerusalem
known to have been made by Berthold U's son Berthold III sometime between
1130 and 1139H.
At first sight the name grave Emmich von Lyningen appears to hold more
promise. Contemporary sources relate that a crusader called Emicho raised
an army which persecuted the Jews of Speyer, Worms, Mainz and Cologne
before departing for the East (95). On the evidence of the chronicle of Zim
mern he has been identified by historians as count of Leiningen (%). However
the first definite mention of an Emicho of Leiningen dates from 1128 (97).
The accounts of Albert of Aachen, Frutolf of Michelsberg and Ekkehard
of Aura do not actually identify the crusader with Leiningen ; they merely
state that Emicho was a count or vir militaris with lands in the area of Mainz,
and that he returned home after his army was defeated by King Coloman
of Hungary in 1096. Furthermore a recent examination of the Hebrew sources
reveals that the surname of the would-be crusader was written by them in
the form VLNHJM, which would tend to exclude any place-name with the
ending -ingen. A more probable identification would be Flonheim (Germany,
Rheinland-Pfalz, Kr. Alzey-Worms) on the middle Rhine. A witness list of
the year 1096 gives the name comes Emicho de Vlanheim (98). This would
agree with Albert's information that Emicho returned home in the summer
of 1096. The lack of prominence accorded to Emmich von Lyningen in the
Zimmern account is another point against this source. According to Albert
and Ekkehard Emicho was the leader of an important contingent ; in the
(94) Hans-Martin Maurer, 'Die hochadligen Herren von Neuffen und von Sper
berseck
im 12. Jahrhundert. Eine personengeschichtliche Untersuchung', Zeitschrift fr
wrttembergische Landesgeschichte, 25 (1966), 59-130.
(95) pp. 292-295 ; Frutolfs und Ekkehards Chroniken und die anonyme Kaiser
chronik, ed. Franz-Joseph Schmale and Irene Schmale-Ott, Ausgewhlte Quellen
zur deutschen Geschichte des Mittelalters. Freiherr vom Stein-Gedchtnisausgabe, 15
(Darmstadt, 1972), 108, 126, 146 ; 'Chronicle of Solomon Bar Simson', in Shlomo
Eidelberg, The Jews and the Crusaders. The Hebrew Chronicles of the First and
Second Crusades (Madison, 1977), 28, 44, 70-71 ; 'Narrative of the Old Persecutions',
in Eidelberg, 107-115.
(96) Adolf Waas, Geschichte der Kreuzzge (Freiburg im Breisgau, 1956), I, 120 ;
Runciman, I, 137 ; Setton, I, 263.
(97) Peter Acht, Mainzer Urkundenbuch (Darmstadt, 1968-71), 1, n 554.
(98) Ingo Toussaint, Die Grafen von Leiningen : Studien zur leiningischen
Genealogie und Territorialgeschichte bis zur Teilung von 1317/18 (Sigmaringen, 1982),
25-28 ; Acht, I, n 395.

THE ARMY OF GODFREY OF BOUILLON

321

chronicle of Zimmern this role is given to the fictitious Walter of Teck and
the doubtful Hugh of Tbingen, while Emicho is only mentioned towards
the end of the list.
Even the participation of the three Zimmern brothers, Frederick, Albert
and Conrad, is open to considerable doubt. The chronicle claims that their
brother Godfrey married Elisabeth, daughter of Frederick of Teck ("). We
know that Count William Werner of Zimmern later recorded the following
information, probably drawn from a lost necrology of the monastery of
Alpirsbach : Obiit Godefridus de Zimbarn liber dominus, sepultus in monasterio Alpirsbach una cum uxore Elisabete ducisse de Deck (10).
We have already seen that the first known duke of Teck is not attested
until 1187. Assuming that this marriage connection did exist (and the only
evidence for it derives from the two Zimmern sources) then it and the supposed
crusading generation must be placed at least a century after the First Crusade,
and most probably in the thirteenth century when we first find a member
of the Teck line with the name Frederick. However the persistence with which
the chronicle glorifies the Teck family could well be explained by such a
marriage connection (101).
In fact the sole name in the Zimmern list which can be confirmed from
other sources is that of grave Hartmann von Dillingen und Kiburg. Hart
mann was the son of Hupold III, count of Dillingen (Germany, Bayern, Kr.
Dillingen) in Swabia. He married Adelheid, daughter of Adalbert of Winterthur-Kyburg, and succeeded to her vast estates in the Thurgau (l02). In 1095
he founded the monastery of Neresheim on his allodial possessions near Nrdlingen. Since Ernest, the first abbot, is known to have been on the crusade
it is likely that his benefactor was identical with the Hartmannus comes
Alemanniae mentioned on three occasions by Albert of Aachen. He is recorded
as having died in 1121 (103). It is quite possible that in this case Froben
Christoph discovered his name in the history of William of Tyre, which he
evidently knew and where he is one of relatively few crusaders explicitly
(99) Chronik der Grafen von Zimmern, I, 71.
(100) F. J. Mone, Quellensammlung der badischen Geschichte, II (Karlsruhe, 1854),
134.
(101) Grnder, 21-23. The Chronik der Grafen von Zimmern (ibid.) also gives
a fabulous story of a duke of Teck who was supposedly elected king in opposition
to Conrad of Swabia on the death of the emperor Lothar II.
(102) On this family see Nikolaus von Salis-Soglio, 'Das Dillinger Grafenhaus
und seine Stiftung Neresheim', Benediktinische Monatsschrift, 3 (1921), 197-214, 269289, although the author does not address the question of Hartmann's participation
in the crusade.
(103) Annales Neresheimenses, MGH SS, X, 20-21 ; pp. 290, 322, 427 {Al
emannia
is probably used by Albert in the sense of 'Swabia*) ; Necrologii Neresheimensis,
ed. F. L. Baumann, MGH Necrologii Germaniae, I (Berlin, 1888), 95-98.

322

. V. MURRAY

identified as a German, and that William in turn derived his information

from Albert or a source common to both (104). With this single exception,
the jumble of anachronistic or non-existent names, as well as those of whom
nothing is known, and above all the lack of external corroboration make
it extremely doubtful whether the chronicle of Zimmern, composed over four
and a half centuries after the First Crusade, can be considered as a reliable
prosopographical source for the Germans who joined Godfrey's army after
Constantinople.
A similar case is that of the sixteenth-century chronicle of Occo Scarlensis
which gives an account of numerous Frisians with Peter the Hermit, and
which was held to be accurate by Dirks and Rhricht (105). Among the Frisians
according to Occo were the nobles Tjepke Forteman, Jarig Ludingaman, Feike
Botnia, Eelke Lyauckama, Sikke Lyauckama, Epe Hartman, Ige Galama and
Obboke Hermanna, most of whom later joined the main expedition. However
a recent study has cast doubt on the reliability of this account, questioning
in particular whether Frisians around the year 1100 could have had the
surnames and forenames given by Occo. Eelke Lyauckama was supposedly
installed as commander of Nicaea after its capture. As the city was returned
to the Byzantine emperor by the crusaders then Eelke, if he existed, was
probably a Varangian in imperial service rather than a crusader. That he was
the leader of 9000 knights in the Frankish army is even more incredible. Thus
while Frisians were undoubtedly present in Godfrey's army we cannot accept
Occo's names as trustworthy (l06).
However contemporary sources do yield the names of some individual
Germans and others who seem to have joined Godfrey or Baldwin after
Constantinople. Apart from Hartmann of Dillingen, discussed above, they
included Gunter (107), Reinhard of Hammersbach (l08), and Wicherius Alemannus, a ministerialis of Fulda (l09). Others can be identified with the help
(104) WT pp. 203-204, 338.
(105) Occo Scarixnsis, Chronyke van Vriesland (Leeuwarden, 1597), 24 ff. ; J.
Dirks, 'Noord-Nederland en de kruistochten', De Vrije Fries, 2 (1842), 135-152;
Rhricht, Die Deutschen, 10, 13-14.
(106) H. Brasst, Die Teilnahme der Friesen an den Kreuzzgen ultra mare
vornehmlich im 12. Jahrhundert (Berlin, 1970), 17-32.
(107) AA p. 526 ; John of Wrzburg, 'Descriptio terrae sanctae', in T. Tobler,
Descriptiones Terrae Sanctae ex saeculo VIII, IX, XII, AT (Leipzig, 1874), 154-155.
(108) AA pp. 422, 424, 435.
(109) AA pp. 507, 522, 526, 531 ; Baldric of Dol, pp. 47, 50 ; Robert the Monk,
Historia Iherosolymitana, Recueil des Historiens des Croisades, Historiens occidentaux,
III, 867 ; Alberic of Troisfontaines, p. 81 1. John of Wrzburg, 'Descriptio Terrae
Sanctae', 154-55 ; Die Kaiserchronik eines Regensburger Geistlichen, ed. E. Schrder,
MGH Deutsche Chroniken, I (Hannover, 1892), 382 ; Metellus von Tegernsee,
Expeditio Hierosolymitana, ed. Peter C. Jacobsen, Quellen und Untersuchungen zur
Geschichte des Mittelalters, 4 (Stuttgart, 1982), 125-126.

THE ARMY OF GODFREY OF BOUILLON

323

of eastern sources. Geoffrey the Monk, later lord of Marash in northern Syria
and regent of Edessa in the 1120s, is called Gufra Almuin in the Anonymous
Syriac Chronicle ; his surname would seem to be a corruption of Alemannus
or Aleman (no). A similar case is that of William, later lord of Diiliik ;
Matthew of Edessa gives him the surname Sancawel, which seems to be an
Armenian rendering of a French name. He may therefore have been identical
with William, brother of the leader of one of the 'People's Expeditions' known
variously as Walter de Pexeio, Walter Sine Habere or Walter Senzavohir.
It is likely that this family came from Boissy-Sans-Avoir (France, Yvelines,
arr. Rambouillet), about forty kilometres west of Paris (! ' ') Another French
man
who joined Godfrey was Drogo of Nesle (France, Somme, arr. Pronne),
who had been released from Byzantine captivity at the duke's intervention (112).
Such lords, it must be stressed, were in straitened circumstances. They
were leaderless, and had lost baggage, arms, mounts and followers in the
debacle at Nicaea. Their adhesion to the newly-arrived contingents is thus
hardly surprising. Yet lords and knights from the other armies were also joining
Godfrey about this time. Godfrey's brother Eustace III of Boulogne had left
Europe in the company of his lord, Robert II of Flanders, and Robert of
Normandy, arriving with them at Constantinople some time after Godfrey (113).
Yet thereafter Eustace seems to have been associated more with his brothers
than with the two Roberts ; he and his men were included in Godfrey's division
of the united crusading army at the Great Battle of Antioch and again at
the siege of Jerusalem where he fought in the same siege-tower as his
brother (I14).
The same development is found with their younger brother, Baldwin. From
the beginning of the crusade he and his wife Godevere were accompanied
by their familia or household (115). During the march he attempted to establish
himself as an independent prince, first in Cilicia, then at Edessa which he
brought under his control by March 1098 (116). With the help of his brothers
he now created a more substantial following of fighting men, described by
(110) 'Anonymous Syriac Chronicle', ed. A.S. Tritton and H. A. R. Gibb, Journal
of the Royal Asiatic Society, 92 (1933), 91-92 ; Matthew of Edessa, Chronicle, ed.
Ara E. Dostourian (Ann Arbor, 1972), 419 ; Walter the ChancelijOR, Galterii
Cancelarii Bella Antiochena, ed. Heinrich Hagenmeyer (Innsbruck, 1896), 87.
(111) Matthew of Edessa, 335; Orderic Vitalis, V, 28; AA pp.274, 286,
Baldric of Dol, 20.
(112) AA pp. 299, 304-305.
(1 13) AA p. 314 ; Andressohn, p. 53.
(1 14) Orderic Vitalis, V, 108-1 10, 168, 176 ; Baldric of Dol, p. 102 ; AA pp. 472,
477, 495 ; WT p. 330 ; Guibert of Nogent, Gesta Dei per Francos, Recueil des
Historiens des Croisades, Historiens occidentaux, IV, 234.
(115) A A p. 302.
(1 16) Hagenmeyer, 'Chronologie', nos. 247-249.

324

. V. MURRAY

Albert of Aachen as the comitatus Baldewini. At the beginning of 1098 this

comprised seventy knights, but after his move to Edessa had grown to at
least two hundred in addition to many others left as garrisons in the towns
he had already captured (U7). Baldwin clearly drew troops from both his
brothers. From Godfrey's contingent came Giselbert of Clermont, Baldwin
of Bourcq, and Drogo of Nesle, and as we have seen, William Sans-Avoir
and Geoffrey the Monk(118). From Eustace's came Udelrard and Pisellus of
Wissant (France, Pas-de-Calais, arr. Boulogne-sur-Mer) (119). Others such as
Peter of Dampierre and Rainald of Touljoined him on a temporary basis (12).
It is also about this time that we first hear of Baldwin's secretary Gerard
who seems to have had a certain responsibility for financial affairs (l21). From
the contingent of Stephen of Blois, who deserted from the army at Antioch,
came Fulcher of Chartres, later lord of Saruj, as well as his namesake the
historian (122). The number of men from Flanders and Artois who first appear
in Jerusalem after Baldwin's accession as king in 1 100 also suggests that many
of the followers of Robert of Flanders joined him at Edessa (123). It would
be understandable for some of the more enterprising of Robert's followers
to be attracted to the service of Baldwin who was the first leader to success
fully
carve out a principality of his own. The resources he obtained were
considerable, including the accumulated treasure of the previous ruler of
Edessa, the Armenian Thoros, part of the dowry received on his marriage
to the daughter of another Armenian prince, Taphnuz, as well as large sums
confiscated or extorted from some of the leading citizens who later conspired
against him (l24). An illuminating passage in the history of Albert of Aachen
reveals how the resources of many knights had been eaten up in the course
of the long march and the siege of Antioch, and directly links this with the
appeal of Baldwin and the resources he had access to as count of Edessa :
(117) AA pp.350, 358, 366; Radulph of Caen, Gesta Tancredi in expeditione
Hierosolymitana, Recueil des Historiens des Croisades, Historiens occidentaux, III, 632.
(118) A A pp. 350, 366, 442.
(119) AA pp. 358, 446.
(120) A A p. 366.
(121) AA pp. 395, 446.
(122) Fulcher (knight) : AA pp. 357, 442, 446 ; Fulcher (historian) : Fulcher of
Chartres, Historia Hierosolymitana, ed. Heinrich Hagenmeyer (Heidelberg, 1913),
pp. 163-164, 206-208. Neither of these two should be confused with a third Fulcher
who was killed in Asia Minor (AA p. 288).
(123) Murray, 'The Origins', pp. 286-269. One factor in this development may have
been the lethargy of Robert compared with the energy of Baldwin. M. M. Knappen,
'Robert II of Flanders in the First Crusade', in The Crusades and Other Historical
Essays Presented to Dana C. Munro by His Former Students, ed. Louis J. Paetow
(New York, 1928), 79-100, concluded that the count only ever showed initiative on
one occasion (in Italy) and allowed himself to be overshadowed by the other leaders.
(124) AApp. 354-355, 360-361,442-443.

THE ARMY OF GODFREY OF BOUILLON

325

Erant enim summa necessitate gravati, et longa expeditione rebus exhausti

necessariis. Affluebant et accrescebant singulis diebus in numero et virtute,
dum fere tola civitas a Gallis obsessa, et eorum hospitalitate occupata est.
Baldewinus singulis, de die in diem, in bisantiis auri, in talentis et vasis argenteis
donaplurima conferebat (125).
Thus while the main army remained bogged down at Antioch Baldwin
was clearly in a position to provide patronage and opportunities for advance
ment
for those who joined him, who were now becoming feudal dependents
receiving salaries or benefits in kind from him.
From around this time, the winter of 1097-98, we can discern a parallel
growth of ties of dependency within Godfrey's exercitus. During the march
across Anatolia numerous horses and draught animals died of thirst. The
loss of horses was especially telling for the knights in the army as it reduced
their military effectiveness, and consequently, their status (126). These losses
were compounded by the privation suffered during the nine-month siege of
Antioch. Henry of Esch had left Godfrey's army in Thrace in order to share
in the presumed munificence of Alexius Comnenus, but was now reduced
to accepting the duke's charity. Hartmann of Dillingen had been obliged to
sell off his horse and armour in order to buy food and could scarcely live
by begging. He was reduced to riding an ass and fighting with a captured
Turkish sword and shield. Godfrey took pity on Hartmann, allotting him
a daily ration of bread and a piece of meat or fish. These circumstances contrast
sharply with Albert of Aachen's description of him as dives et nobilissimus
et unus de praepotentibus in terra Alemanniae (127). The cases of Henry and
Hartmann are particularly telling, since they had evidently been able to finance
the construction of a siege machine from their own resources at Nicaea (128).
It is thus evident that from the time of the siege of Antioch ever-increasing
numbers of knights were penniless and had nothing to bargain with except
their own service. Despite the scarcity of food and the inflationary prices
commanded by what meagre supplies became available Duke Godfrey
apparently possessed the means to provide for such men who offered him
their service. As we have seen, Baldwin had the resources of the county of
Edessa at his disposal ; what resources were available to Godfrey ?
We must first go back to the duke's financial preparations for the crusade
in the winter of 1095-96, which seem to have been rather unsystematic, evolving
gradually as particular needs were perceived. One of his first actions was to
dissolve the priory of St. Peter at Bouillon, a house belonging to the abbey
of Saint-Hubert, and to confiscate its possessions. He was only persuaded
(125)
(126)
(127)
(128)

A A pp. 441-442.
A A pp. 339-341.
AAp.427.
AAp. 322.

326

. V. MURRAY

to make restitution at the urging of his mother Ida who travelled from
Boulogne to intercede with her son (129). However it seemed that the sums
Godfrey wished to raise would call for wider action. He sold the allods of
Baisy and Genappe to the abbey of St. Gertrude at Nivelles (13), while various
smaller estates clustered around the town of Maastricht were either sold or
given to the church as pious donations (131). Further south, his rights in the
county of Verdun, as well as the allods of Stenay and Mouzay, were sold
to the bishop of Verdun for an unspecified sum (l32). However even after
these transactions Godfrey was obliged to proceed to the mortgage of his
allodial territory of Bouillon, along with an adjacent fief, lying to the south,
which was held from the church of Reims (I33). The sources agree that the
sum realised amounted to at least 1300 silver marks (l34). The cash was raised
by Bishop Otbert of Lige who ransacked the churches and abbeys of the
diocese for jewels, plate, and precious metals (135). It is questionable whether
the total amount realised by the lesser sales was greater than that brought
by the mortgage of the strategically important fortress of Bouillon and its
surrounding territory (l36). What is clear, however, is that Godfrey had a
considerable sum at his disposal on the eve of his departure, and probably
(129) Cantatorium, 203-206.
(130) MGH DD Heinrich IV, no. 459.
(131) S. P. Ernst, Histoire du Limbourg (Lige, 1837-52), VI, 113-115 ; Miraeus,
1,77.
(132) LLp.498.
(133) Bormans and Schoolmeesters, I, n 35 ; Cantatorium, 244-245 ; C. ZollerDevroey, 'Fodalit et conomie dans les Ardennes mdivales : le fief de Bouillon
en Sedanais', in Centenaire du Sminaire d'histoire mdivale de l'Universit libre
de Bruxelles, d. Georges Despv (Bruxelles, 1977), 21-57.
(134) Renier of St. Laurence (Triumphale Bulonicum, MGH SS, XX, 584) and
Giles of Orval (p. 91) both give the sum of 1300 marks of silver and three marks
of gold which would equate with the sum of 1300 marks of silver and one pound
of gold given by Laurence of Lige (LL p. 498). The chronicle of Saint-Hubert (Canta
torium, p. 206) gives 1500 pounds of silver, while Alberic of Troisfontaines (p. 804)
gives 1500 marks of silver. The Triumphus S. Lamberti Leodiensis de castro Bulliono
(MGH SS, XX, 499) and the Gesta abbatum Trudonensium continuatoris tertii
auctorum (MGH SS, X, 387) both give the smaller figure of 1300 silver marks.
(135) Giles of Orval, 91; Cantatorium, pp. 207-208, 249-250; Gesta abbatum
Lobbiensium, MGH SS, XXI, 318.
(136) Jean-Louis Kpper, Otbert de Lige : les manipulations montaires d'un
vque d'Empire l'aube du xne sicle', Le Moyen ge, 86 (1980), 353-385. On the
castle of Bouillon and its importance, see Lon Saur, 'Entre Bar, Namur et Lige :
Bouillon, place stratgique', Publications de la Section Historique de l'Institut GrandDucal de Luxembourg, 95 (1981), 258-281, and Andr Matthys, 'Les fortifications
du xie sicle entre Lesse et Semois', in Burgen der Salierzeit, Teil 1 : In den nrdlichen
Landschaften des Reiches, ed. Horst Wolfgang Bhme, Rmisch-Germanisches
Zentralmuseum, Forschungsinstitut fr Vor- und Frhgeschichte, Monographien 25
(Sigmaringen, 1991), 225-280.

THE ARMY OF GODFREY OF BOUILLON

327

took a large part of it with him in the form of coin. Two of his deniers,
similar in form to coinage of Lige struck during the episcopate of Otbert, have
turned up in hoards in Russia. They bear an inscription which should be
read as GODEFRIDUS IEROSOLIMITANUS in the sense of peregrinus,
and must have therefore have been coined between November 1095 and
August 1096, probably in one of the Lige mints (137).
These initial resources were greatly augmented in the course of the
expedition. During his progress up the Rhine Godfrey exploited the antiSemitic frenzy engendered by popular preachers to extort protection money
from the Jewish communities of the middle Rhine. The Hebrew chronicles
relate that Godfrey, 'may his bones be ground to dust', received 500 zekukim
of silver from the Jews of Cologne and another 500 from Mainz, despite
the fact that as duke of Lower Lotharingia he had been ordered by Henry
IV to prevent persecution. The value of the zakuk is placed variously at either
eight or twelve ounces of silver ; the total profits of this short campaign of
extortion must have therefore been something between 8000 and 12,000 ounces
of silver, that is between 1000 and 1500 marks (138).
By the time of his arrival at Constantinople at the latest Godfrey's financial
resources had begun to tighten the bonds of dependence in his army. Like
other leaders he received gifts and money from the emperor, and continued
to receive an imperial subsidy as long as he was encamped on Byzantine
territory. The duke distributed this money among his army according to the
needs of each individual ; most of it seems to have been immediately spent
on food supplies by the recipients (I39). Godfrey thus played a key role as
the channel through which funds passed ; it is likely that a similar system
(137) N. Bauer, 'Der Fund von Spanko bei St. Petersburg', Zeitschrift fr Numism
atik, 36 (1926), 75-94, identifies the first coin as a denier of Godfrey II ('the Bearded1).
However it bears a strong resemblance, especially in the inscription, to the second,
described by Victor Tourneur, 'Un denier de Godefroid de Bouillon frapp en 1096',
Revue belge de numismatique, 83 (1931), 27-30. Recently John Porteous, 'Crusader
Coinage with Greek or Latin Inscriptions', in A History of the Crusades, ed. Setton
et al. (Philadelphia, Madison, 1955-89), VI, 354-420, has made the surprising claim
that 'The dukes of Lower Lorraine as such did not issue coins, nor did Godfrey strike
any for his territory of Bouillon. In France, however, their father Eustace II struck
aalso'
scanty
(p. 356).
coinage
However
as count
it is of
clear
Boulogne,
that the and
Ardennepossibly
Verdun
their dynasty
brother had
Eustace
issuedIIIcoins
did
since the early part of the eleventh century. See Gnter Albrecht, Das Mnzwesen
im niederlothringischen und friesischen Raum von 10. bis zum beginnenden 12. Jahr
hundert,
Numismatische Studien, 6 (Hamburg, 1959), especially 66-67 and 79-84.
(138) 'Chronicle of Solomon bar Simson', pp. 24-25 ; Eidelberg, p. 147 ; E.
Tubler, 'Spuren von Urkunden in den hebrischen Kreuzzugsberichten', Mitteilungen
des Gesamtarchivs der deutschen Juden, 5 (1914), 143-146.
(139) AA pp. 310-311 ; WT pp. 175-176; Anna Comnena, Alexiade, ed. B. Leib
(Paris, 1937-45), II, 220-226.

328

. V. MURRAY

operated in the distribution of booty and forage, as the fighting divisions

of the crusading army corresponded to the original individual contingents.
Another vital source of supply was Baldwin ; once his position in Edessa
was secure he was able to aid the crusaders at Antioch, sending plurima talenti
auri et argenti to the other leaders for distribution. However he clearly favoured
his own brothers, with a sum put at 50,000 bezants, in addition to large
quantities of corn, barley, wine and oil (I40). This massive support from their
younger brother must have given Godfrey and Eustace a certain edge over
the other leaders. Again Godfrey seems to distributed most of his resources
to the needy, including many knights. However in August 1098 while disease
was rife in Antioch he was able to withdraw with his army to Baldwin's
territories of Turbessel and Ravendel for forage and recuperation (I41).
Thus by the time of the siege of Antioch, a time when many in his own
exercitus and indeed in other contingents were in serious financial difficulties,
Godfrey had access to new sources of income and supplies in addition to
whatever reserves had remained from earlier. The growth of ties of dependence
may also have been expedited by the disappearance of intermediate levels
in the command structure of the army. It is surely no coincidence that Gerard
of Avesnes and Giselbert of Couvin, both later found in the service of Godfrey
and Baldwin in Jerusalem, were originally vassals of Baldwin of Hainaut who
disappeared in Asia Minor while he was on an embassy to Alexius Comnenus (l42). The two Fulchers of Chartres, the future lord of Saruj and the
historian, were originally in the Champagne contingent which was left leaderless by the desertion of Stephen of Blois (143). In these instances the removal
of their immediate lord or recognised leader appears to have brought about
a closer bond to Godfrey and Baldwin.
If the arrival at Constantinople marked the beginning of a second stage
in the development of Godfrey's exercitus, the third stage was signalled by
the entry into Palestine in the spring of 1099. The subsequent establishment
of a Frankish state with Godfrey as its ruler allowed him to provide patronage
in the form of fiefs and financial support. Although the actual territory under
his control was small, he could also dispose of substantial amounts of tribute
paid by the Muslim cities of the coast. Thus the revenues of the port of Arsuf
were assigned to the knight Robert of Anzi ; it is probably also significant
that he was a Norman from southern Italy who had originally come on crusade
with Bohemund (144). The original Lotharingian element in Godfrey's army
had been depleted by death in battle, capture and disease ; among the known
(140)
(141)
(142)
(143)
(144)

AA pp. 395-396.
AA pp. 427-428, 440-441.
Murray, The Origins', 285.
AA pp. 347, 442, 446 ; Fulcher of Chartres, 206-207.
A A pp. 514-515 ; Murray, 'The Origins', 289.

THE ARMY OF GODFREY OF BOUILLON

329

casualties from this group were Gozelo of Montaigu, Henry of Esch, Folbert
of Bouillon, Baldwin of Hainaut, Louis of Toul and Adalbero of Metz (145).
Others were with Baldwin, now count of Edessa. Lastly, after the liberation
of the Holy City, large numbers of crusaders regarded their vows of pilgrimage
as having been fulfilled and returned to Europe in the course of the following
year. They included several important Lotharingians : Peter of Dampierre,
Rainald of Toul, Dudo of Cons, Cuno and Lambert of Montaigu and Louis
of Mousson (146). Apart from the short-lived Warner of Grez, none of the
original core group of Godfrey's allies and kinsmen actually remained with
him in Palestine. At the same time other knights whose own lords were
returning to Europe now entered Godfrey's service, and were joined by others
from Edessa when Baldwin succeeded his brother in July 1 100, so that after
1100 the Lotharingian and German elements in the Jerusalem nobility were
outnumbered by Flemings, Artesians, Picards, Normans, and Francians (147).
This examination of Godfrey's exercitus gives a clear picture of how this
situation arose. When it set off in August 1096 the army was almost entirely
Lotharingian in composition, a character which was greatly influenced by
the alliances and animosities which had arisen in the two duchies during the
years of the Investiture Contest and its accompanying feuds. Yet in the course
of the three years it took to reach its ultimate goal, the army was constantly
changing in composition and structure. By the end of Godfrey's short reign
as ruler of Palestine most of the Lotharingians had been killed or returned
to Europe, to be replaced by men from the other original contingents. It
would therefore seem that, ultimately, the character of the early Jerusalem
nobility was determined far less by feudal and kinship ties dating from before
1096 than by events and conditions prevailing during the course of the First
Crusade itself.
(145) Gozelo : AA p. 359 ; Henry : AA p. 435 ; Folbert : AA p. 436 ; Baldwin : AA
pp. 434-435 ; Louis : A A p. 375 ; Adalbero : A A pp. 370-371.
(146) Peter : Barthlmy, pp. 401-406 ; Rainald : Choux, Recherches, Reg. nos. 81,
84, 86, 89 ; Dudo : Kurth, I, n 63 ; Cuno and Lambert : Murray, 'The Origins',
297 . 25 ; Louis : M. Grosdidier de Matons, Catalogue des actes des comtes de
Bar 1022-1239 (Paris, 1922), n 35.
(147) Murray, 'The Origins', 293-294.